PB85-207538
EPA/600/2-85/059
May 1985
THE RUTGERS STRATEGY FOjR COMPOSTING;
PROCESS DESIGN AND CONTROL
H. S. Finstein, F. C. Miller,
S. T. MacGregor, and K. H. Psan'tmos
Rutgers, The State University of New Jersey
New Brunswick, New Jersey 08003
Grant Nc, R806829010
Project Officer
Atal E. Eralp
Wastewater Research Division
Water Engineering Research Laboratory
Cincinnati, Ohio 45268
WATER ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
TECHigiCAL REPORT DATA
(I'lcase read Imtniclions on the reverse before completing)
1. REPORT NO.
_ EPA/600/2-85/059
4. TITLE AND OUUTITLE
The Rutgers Strategy for Composting: Process
Design and Control
8. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
M.S. Finstein, F.C. Miller, S.T. Mac^reqor and
_K.M. Psarianos
0. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
The State University of New Jersey, Rutgers
Department of Environmental Science
P.O. Box 231 New Brunswick
New Jersey 08903
li REPORT DATE
10. PROGRAM ELEMENT NO
CAZB1B
11. CON TO ACT/GRANT NO.
R806329010
12. SPONSORING AGENCV NAMt AND ADDRESS
Water Engineering Research Laboratory - Cin, OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final Project Report
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Atal E. Eralp (513) 684-2621
16. ABSTRACT
A strategy for sludge composting was developed to ccuriter the tendency of
other composting systems to operate at high temperatures t,l«at inhibit and slow
decomposition. This method, known as the Rutgers strategy, can be implemented
in a static pile configuration to retain structural and operational simplicity,
or_in a more elaborate enclosed or reactor structure system. The method main-
tains a temperature ceiling that provides e high decomposition rate throunh on-
demand removal of heat by ventilation (thermostatic control of a blower).
Compared with the approach currently in widespread use, the Rutgers strat-
env yields high-rate composting that decomposes four times more waste in half
the time.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDCN TIFIERS/OPEN ENPED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (Tins Report/
UNCLASSIFIED
21 . NO. OF PAGES
284
20. SECURITY CLASS (Thispagel
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION i= OBSOLETE
-------
DISCLAIMER
Although the Information described in this document has been funded
wholly or in part by the United States Environmental Protection Agency
through assistance agreement number R806829-01-0 to Rutgers, The State
Universityof New Jersey, it has not been subjected to the Agency's required
peer and administrative review and therefore does not necessarily reflect
the views of theAgencyand no official endorsement should be Inferred.
ii
-------
FOREWORD
The U.S. Environmental Protection Agency is charged by Congress with
protecting the Nation's Ian;!, air, and water systems. Under a mandate of
national environmental laws, the agency strives to formulate and implement
actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life. The Clean Water
Act, the Safe Drinking Water Act, and the Toxics Substances Control Act
are three of the major congressional laws that provide the framework for re-
storing and maintaining the integrity of our Nation's water, for preserving
and enhancing the water we drink, and for protecting the environment from
toxic substances. These laws direct the EPA to perform research to
define our environmental problems, measure the impacts, and search for
solutions.
The Water Engineering Research Laboratory is that component of EPA's
Research and Development program concerned with preventing, treating, and
managing municipal and industrial wastewater discharges; establishing prac-
tices to control and remove contaminants from drinking water and to prevent
its deterioration during storage and distribution; and assessing the nature
and controllability of releases of toxic substances to the air, v/ater, and
land from manufacturing processes and subsequent product uses. This publica-
tion is one of the iroducts of that research and provides a vital communica-
tion link between t!>e researcher and the user community.
The report concerns composting as a sewage sludge treatment technology.
Unlike the familiar, informal, small-scale "backyard" composting of leaves,
grass clippings, and other plant remains, municipal-scale composting poses
significant, problems in facility design and control. By following scientific
and technical principles, as developed herein, the public acceptability and
cost-effectiveness of such facilities can be greatly improved.
Francis T. Mayo, Director
Water Engineering Research Laboratory
iii
-------
ABSTRACT
The main determinant of composting process performance is de-
composition rate. This rate is negatively affected by temperature
exceeding 60°G, owing to the inactivation of the responsible micro-
bial community. Nonetheless, composting masses typically self-heat
to 80°C, at which point the rate of decomposition is low. The de-
sired temperature ceiling providing a high rate can be maintained
through on-demand ventilative heat, removal (thermostatic control of
a blower). This constitutes, in essence, the Rutgers strategy for
composting process design and control.
This strategy, compared to the approach currently in widespread
use, yields high rate composting in that approximately 4X more waste
is decomposed in half the time. Although the strategy can be im-
plemented in certain enclosed (in-vessel) configurations, the sug-
gested unenclosed static pile configuration is advantageous in being
structurally and operationally simple, and capital non-intensive.
The rational for this strategy may be expressed symbolically as
follows :
where: 0 a heat removed through vaporization (energy/time)
0.9 a approximate proportion of total heat removed through
vaporization
m B dry air mass flow (mass dry air/time)
h ° enthalpy of outlet air (energy/mass dry air)
h. •= enthalpy of inlet air (energy/aass dry air)
The goal is to maximize Ov , because this is functionally equivalent
to maximizing the rate of waste (e.g. sewage sludge) decomposition,
Realistically, this can only be approached through manipulation of
m (^ time-variable adjustment of ventilation rate) such that the
value of hp. t corresponds to a temperature of 60°C or less. This
translates to temperature feedback (thermostatic) control of a
blower.
This report was submitted in fulfillment of Contract No.
R806829010 by Rutgers University under the sponsorship of the U.S.
Environmental Protection Agency. This report covers the period
10/8/79 to 11/7/81, and work completed as of 11/7/81.
IV
-------
CONTENTS
foreword , iii
Abstract „ iv
Acknowledgments . vii
1. Background 1
2. Materials and Methods 6
3. Comparison Between the Rutgers and Beltsville Control
Strategies: Observation and Interpretation of System
Behavior . 22
4. Sequence of Limi*\tions Induced by Control Strategy. ... 57
5. Rutgers Strategy: Replacement of ' odchips with Re-
cycled Compost as the Bulking Agent 65
6. Comparison Between the Rutgers Strategy and Beltsville
Process: Materials Balance 91
7. Mathematical Description o£ Process Control Dynamics. . . 96
8. Drying Associated with Composting, and Non-Biological
Air Drying 101
9. Comparison Between the Rutgers and Beltsville Strategies:
Effect on Curing Stage 116
10. Rutgers Control Strategy: Diagnosis of Processing
Failure. „„,•,...„ „ 122
11. Pathogen Inactivation. .... o ..... 132
12. Uniform Provision of Air Along the Length of a Compost-
ing Pile. ........................ 148
Conclusions. 155
Recommendations , 1S6
References. 157
Appendices
A-l. Temperature Observations for Piles 7, 8, 9A and 9B. . . 162
A-2. Oxygen Observations for Piles 7 and 8. ........ 204
A-3. Moisture Content in th* "Whole Sample" and the "Non-
Woodchip Fraction". .................. 209
-------
A-4. Representativeness of Pile 9B 213
A-S. Forces Driving Vaporization: Metabolically Generated
Heat, and Unsaturation of Inlet Air 21S
A-6. Relative Sensitivity of the Moisture Content and
Volatile Solids Tests. . . . 216
B. Advantages of Fuel Production Through Composting vs.
Direct Combustion of Sewage Sludge Cake. ....... 217
C. Temperature Observations for Piles 11A, 11B, and 11C. . 218
D-l. Air Needed to Remove Heat and Supply Oxygen. ..... 236
D-2 Unsuccessful Attempt by the Beltsville Group to Im-
prove Drying, in Isolation From Considerations of
Process Dynamics. ............ . . 236
E. Temperature Observations for Pil= 12 237
F. Observations on Blower Operation, Temperature, 0-f
C0~, pH, and Moisture Content for Piles 6A, 6B,
and 6C. ................ 241
G. Temperature Observations for Pile 13 266
-------
ACKNOWLEDGMENTS
This investigation was funded by the U.S. Environmental Pro-
tection Agency under Contract No. RS06829010 (Dr. Atal E. Eralp,
Project Officer). Support was also provided by State funds from the
New Jersey Agricultural Experiment Station (publication No. H-07472-
1-84) .
Thanks are expressed to the field crew at the Caraden County
(New Jersey) Municipal Utilities Authority Sewage Sludge Co- posting
Facility (Horace T. Banks, Foreman), for their cooperation in '-Act-
ing up the field trials.
We greatefully acknowledge Dr. Peter F. Strom for many dis-
cussions on the intricacies of composting dynamics.
VI1
-------
SECTION 1
BACKGROUND
SLUDGE MANAGEMENT GOALS
Composting can serve as a sludge treatment technology because it
advances the goals of decomposing putrescible (odor-causing) material,
decreasing volume, weight, and water content, inactivating pathogenic
organisms, and producing a stabilized process residue. This residue
is more easily stored, transported, and disposed of than the sludge.
Thus, composting is a treatment option within an overall sludge
management plan.
A minimum accomplishment of composting would be to convert a
sludge which is not acceptable for disposal at a sanitary landfill to
a process residue of lesser amount, which is acceptable. More desir-
ably, a use would be found for the residue. The traditional use is
as a compost (organic soil amendment) in gardening and horticulture,
and this might be extended to reclamation of disturbed land and the
preparation of 'j. and fill cover material. A novel possibility, result-
ing from composting's capacity to remove water, is to use the residue
as a waste-derived, low grade, solid fuel,
COST-EFFECTIVENESS AND PUBLIC ACCEPTABILITY
Even if a use is found for the process residue its monetary value
will, in all likelihocu, be small relative to capital and operating
costs. Consequently, cost-effectiveness does not reside primarily in
product sales, but rather in economical construction and operation.
Thus, the f-.cility should not be viewed, for example, as a compost
factory, but rather as a sludge management center. This leads to a
process, rather than a product, orientation (1).
In addition to being cost-effective, the facility must be publicly
acceptable in terms of aesthetics. This requires an absence of odor
nuisance, insect breeding, rodent harborage, etc.. Such nuisances
often havs public health overtones.
IMPORTANCE OF DECOMPOSITION RATE
Both public acceptability and cost-effectiveness hinge on the rate
of decomposition. The basic preventative measure against odor produc-
tion is to speed the decomposition of the putrescible organic waste.
Similarly, a high rate of decomposition is consistent with lov; capital
and operating costs. This is because a nigh rate results in a need for
less facility time/space and structural appurtenances to achieve a
-------
given degree of stabilisation, less material to be handled, stored, and
transported, and improved bandleability. It is necessary, therefore,
to identify the operative rate-limiting factors.
OVERRIDING RATE LIMITING FACTOR: INTERACTION BETWEEN HEAT GENERATION
AND TEMPERATURE
With the partial exception of pathogen inactivation (see Section
11), the aforementioned sludge management goals (decomposition of
patrescible material, decreasing the volume, weight, and i-ater content,
producing a stabilized residue) may be equated with the generation of
heat. _This is because the heat is generated microbially through de-
composition of the waste; hence, heat generation is functionally
equivalent to waste treatment. In understanding the physical, chemical,
and biological dynamics which govern the composting system, it is use-
ful to focus on heat generation. In particular, the overriding determi-
nant of system behavior is an interaction between heat generation and
temperature (2-3).
Consider the behavior of a non-managed pile of organic material
having the following characteristics: it is large enough to be self-
insulating; the material is moist and nutritionally supportive of micro-
bial growth; the material is sufficiently porous to allow gas exchange;
the exchange suffices to prevent gross oxygen depletion in the inter-
stitial atmosphere. Such a pile spontaneously increases in temperature
because, for a period of time, the rate of microbial heat generation
within the pile exceeds the rate of h",at loss to the surroundings, This,
"self-heating" phenomenon is the basis of the composting process.
At the outset of self-heating a positive feedback loop becomes
established between microbial heat generation and temperature, in that
higher temperatures favor microbial growth with its associated metabolic
generation of heat. When the temperature begins to exceed approximately
38 C the feedback turns negative, because higher levels are progress-
ively unfavorable to mesophilic* growth and activity. This slows the
temperature ascent and would, in the absence of subsequent events,
terminate it at approximately SO°C.
The temperature ascent is renewed, however, with the initiation of
thermophilic* growth, starting at approximately 4S°C. This reestabli-
shes positive feedback between heat generation and temperature. Since
the thermophilic community in self-hsating organic masses is most active
at approximately SS°C, the feedback starts to become negative when the
temperature exceeds this value. The temperature ascent again slows,
typically peaking at 80°C. At this temperature, heat generation is
slight.
Mesophile - an organism living in the temperature range around that of
warm-blooded -animals; thermophile = an organism living at high tempera-
ture (T.D. Brock, D.W. Smith, M.T. Madigan. Biology of Microorga-
nisms, 4th Edition, Prentice-Hall, Inc., 1984). A competent array of
both types is reliably present in organic wastes.
-------
To realize a high rate of decomposition the system's tendency to
self-limit via inhibitively high temperature must be countered. A means
of doing so is to match ventilative heat removal to heat generation in
reference to a 60°C ceiling, as accomplished through temperature feed-
back control of a blower. This development, which paved the way for rhe
present work, was described earlier (2-7). It constitutes the approach
to composting process control known as the Rutgers strategy.
CONFIGURATION AND STRATEGY: TWO LEVELS OF COMPOSTING PROCESS ORGANIZATION
Two levels of composting process organization can be recognized.
Strategy lies at the conceptual level, in that it represents a plan for
guiding the interacting physical, chemical, and biological events. Im-
plementation of a strategy is through some set of physical and mechanical
elements, including geometry of the composting mass, equipment for venti-
lation, ar:d machinery for material handling. The elements at this level
of process organization are collectively termed configuration. The dis-
tinction between strategy and configuration is basic to this investiga-
tion.
The configuration used was that of unenclosed static pile — meaning
that the material was in the open, was ver.f.Hated by blower, rind wa:i not
agitated mechanically during composting.
SIGNIFICANCE OF THE RUTGERS-BELTSVILLE COMPARISON
An essential feature of this report is a comparison between tht
Rutgers strategy and the strategy embodied in the Beltsville Static Pile
Process. These represent fundamentally different approaches to the man-
agement of the composting microbial ecosystem (TABLE 1).
The Beltsville Process was developed specifically for the treatment
of sewage sludge (8-11), and is in widespread use (12-13). This process
is advantageous in its structural and operational simplicity, owing to
the static pile configuration. Like most composting systems, however,
it suffers from slow decomposition as a result of inhibitively high tem-
perature. By employing the Rutgers strategy, however, in static pile
configuration the structural and operational simplicity can be retained,
while benefitting from rapid decomposition.
The difference in behavior induced by the two strategies originates
in the management of ventilation. The Rutgers strategy focuses on heat
removal for temperature control, whereas the Eeltsville Process focuses
on the maintenance of an oxygenated condition. These operational ob-
jectives are met through the respective approaches taken to blower con-
trol, sizing, and operation mode. The physical, chemical, and biological
dynamics governing the composting system, however, dictate certain con-
sequences that might not be immediately apparent. By focusing on tem-
perature in this manner an abundance of oxygen is automatically provided,
whereas by focusing on oxygen inhibitively high temperatures result.
These principles pertain to composting in general, regardless of the
particular type of waste being treated, the configuration employed, or
the name or proprietary status of the process.
-------
TABLE 1.
FUNDAMENTALLY DIFFERENT COMPOSTING PROCESS CONTROL STRATEGIES
Process control strategy
Process control operational
objective
Rutgers
Maintain 60°C
temp ceiling
Beltsville
Maintain 0, at
S% to 15%
Blower control
Blower sizing
Fixed schedule
initially, fol-
lowed by temp
feedback
Must meet peak
demand for heat
removal
Fixed schedule
throughout
Prescribed as 1/3 hp
per 50 ton pile
Blower o elation mode
Consequences o£ strategy
Forced-pressure
System oxygena-
ted; a high rate
of heat genera-
tion and vapo-
riza.tion; dry-
ness may come
to inhibit acti-
vity, tailess pre-
vented; good
pathogen kill.
Vacuum- induced
System oxygenated;
temp peaks, by de-
fault, at an inhi-
bitively high level
CT ; a low rate
of heat generation
and vaporization;
good pathogen kill.
Both strategies were implemented in unenclosed static pile
configuration.
Heat generation is equivalent to decomposition.
-------
POSITION OF THE PRESENT REPORT
Most of the essential findings reported herein have already been
published (1-7, 14-26) or are in an advanced stage of publication (27-
30,. These papers represent different facets of a coherent line of re-
search, but are scattered. The present Report permits a more integra-
tive treatment in a single volume, and provides an opportunity to docu-
ment the data in full detail.
Finally, our basic view on composting process design and control,
first enunciated in detail in 1980 (4), has received independent con-
firmation by at least four groups. Two of the studies were sponsored
by EPA (31-33) and two were conducted in European countries (34-36).
-------
SECTION 7.
MATERIALS AND METHODS
Materials and methods generally employed in the investigation are
described in the present section. Modifications and additional pro-
cedures pertaining to particular studies are described in the indivi-
dual sections.
Site. Our pilot-scale composting research frcility was located
at the~~CTamden County Municipal Utilities Authority sewage treatment
plant at Second Avenue and Jackson Street, Camden, N.J. , where the
Authority routinely composted the sludge from their Jackson Street
and Baldwin Run treatment plants (total 18-23 dry tonnes/day) (20-
25 tons/day) by means of the Beltsville static pile process (8-11).
Although sludge, woodchips, and supportive services were provided by
the Authority, t'?.s investigation reported herein was independently
designed and executed.
At the pilot research facility (Figure 1) a shed housed the in-
strumentation and control systems (Figure 2). The shed was heated and
ventilated to prevent temperature extremest
SJLudgjs. The sludge consisted of a mixture from two sources.
Approximately 90% of the material came from the Jackson Street plant,
where sewage treatment consists solely of primary settling. The re-
mainder came from Baldwin Run, where "partial digestion" was accom-
plished in an Imhoff tank. Thus, the experimental material was
essentially raw fresh sludge.
The Jackson Street plant served a mixed residential-industrial
area, where a food processing plant was the largest industrial contri-
butor. The sludge was preconditioned with the addition of approxi-
mately 1 kg (2.2 pounds) of polymer (Allied Colloid Percal 728, a
chloride based cationic polyeletrolyte) per metric tonne (1.1 ton) dry
solids, followed by dewatering through a belt filter press. This
generally yielded a cake of approximately 25% solids (oven dry weight
expressed on a wet weight basis). Analyses of the sludge are avail-
able (37).
Woudchips. Virgin woodchips from hard and soft woods having
nominaT~3T5ieiTsTons of 2.5 x 2. S x 0.6 cm (1 x 1 x 1/4 in.) were used
to form a base for the composting pile, as a bulking agent for mixing
with the sludge, and to form a cover over the composting pile.
Mixture. Sludge cake (by wer weight) and woodchips (by volume)
were mTxed in a ratio-of approximately 1 tonne sludge per 1.7 m-5 wood-
chips (1 ton per 2 yd ). Mixing was .done in an industrial pug mill
-------
0 LOWERS
VAN
WITH
f
•63
'&',
*
C C M U A
OPi R ATIONS
AREA
Figure I. Layout of the Department of Environmental Science,
Rutgers University, research facility at the Camden County
Municipal Utilities Authority treatment plant. A typical
layout for three 6- metric tonne piles is illustrated.
-------
i
Figure 2. Shed interior showing instrumentation. Photo by
F.C. Miller, z
-------
(McLanahan Corp., Hollidaysburg, PA). The mixture was discharged to
a dump truck for transfer to the pilot facility, 300 meters distant
(980 feet).
Mechanical ventilation. Corrugated f'lexhose (Agway Inc. ,
EnglisFtoT7nT~N77.J~wTtTrT~HTameter of 10.2 cm (4 in,) was used for
ventilation ductwork. This was nonperforated, or perforated, as
required, A radial blade blower (Dayton, Inc., Chicago, 2C820) with
a 1/3 hp motor (Dayton #SK586a) provided ventilation. Unless noted
otherwise, the ductwork was fitted to the outlet end of the bloi-er,
thus providing ventilation in the forced pressure direction (7).
. f^rst: steP in pile construction was
to form~a~EaTe~bf woodcETps, in which was buried one cr more lengths
of perforated flexhose (Figure 3], This was connected to the blower
via a length of non-perforated flexhose. Over the base was placed
the sludge-woodchip mixture, or the sludge-recycled compost mixture,
which was covered with woodchips (Figure 4). The piles were parabolic
in cross section, a slight longitudinal axis parallel to the flexhose
in the smaller piles, and a distinct axis in the larger piles (see
Figures 3 and 4).
Blower coiitrg_l - Rutgers strategy. The Rutgers strategy (2-7)
Uitiliz"e~d~aT temp^Fature-actuated~~5Towercontrol system consisting of a
temperature controller with an adjustable set point (Fenws.ll Inc.,
Ashland, MA., Series 551), and a thermistor (Fenwall 28-232306-304) in
the pile (Figures 5 and 6). The controller continuously received and
interpreted a signal from the thermistor. To dampen response, the
thermistor was shielded by threading it into a short length of I in.
NPT iron pipe. When the signal indicated a temperature less than the
set point the controller actuated the blower through the timer, as
scheduled. When t-he temperature was above the set point the controller
directly actuated the blower until ventilation decreased the pile
temperature (in the vicinity of the thermistor) to less than the
set ] 'int. Control is thus "on demand," based on the feedback of
a temperature signal.
Monitoring - ^utc^atj^c. Temperature was determined with thermo-
coupJ eT~macle o± 20 gauge copper-constantan wire (Thermoelectric Corp.,
Saddle Brook, N.J.). Thermocouples were taped to wooden dowels at
0.3m (1 foot) intervals, and the dowel vere inserted horizontally into
the pile to establish a vertical 0.3 x J.3m grid (Figure 7). The grid
was perpendicular to the long ax^s of the flexhose. The grid was in
the center of the pile, unless noted otherwise. (For photographic
examples see Figures 44, 45 and 46.)
Thermocouple Isads were; organized through the use of subjunction
boxes at each pile, through a main junction box? and finally were wired
to a recording monitor (Doric Scientific, San Diego, CA. , Digitrend
220). The temperature reading was logged every hour.
Gas sampling probes (Figure 8) were inserted into the pile. The
probes led, via 0.64 cm (j in.) I.D. Tygon tubing, to a condensate trap
-------
Photo by F.C. Miller,
f woodohip
10
-------
l^c _r I T
U>k f*. ./, ' •
:A
r--.
P \
if
,*' .
^•.if r^ ^-
1 it-i fj- ! - IE , 3f I \' ,1 .
W
'* ! ((/ I '
<• \Vv->}//
i^
• ii: -^
cd
^ \ ' i«ix
-x ' • 4 i ^
,[\l_' t>
,^i- • ; ' /
,. - J , T"'«"-M''-v?^ ^^
"k^yr •'./, 'v%; V*"
^(-f * J^^_ . - ^ ^T ' >^ , -
'< „ •- - i ,. . , - ; •,„ vo- ---'^'
•rf * "- r , * ^- •" •& . -; '^n:Kr' ,„ • -•"li-
' ^/% r' ' ^ _ « ^ ^ - ,, *,„ - > ^
• '•" ,f ^ < '•> ', >.:->•*' ,. ^i
" "
Figure 4 Typical pile construction, placement of sludge-
P?C Miller " °VSr thS W°°dchiP base" Ph°to by
11
-------
«• THERMtS TORI-—*
T < T(SET)
Figure 5. Blower control logic flow diagram.
-------
. O
u
ORMAL
EVENT RECORDER
r
HOI
IN
r
NlUT
IN
GRNO
IN
FHERMISTOR
TEMPERATURE
CONTROLLER
H N G
HIOH
HOI
, r
-*•- TO BLOWER
10
WIRING
TERMINAL
STRIP
-*»-TO PILE
TIMER
NO.
N H N
Figure 6. Blower control wiring schematic,
13
-------
j:r^
-*l
\ '-
»•-
Figure 7. Insertion of a dowel (pile 11C). Fixed to the dowel
was the thermistor, and 5 thermocouples at 0.3 m intervals.
The personnel were, from left to right, Messrs. Psarianos,
MacGregor, Miller and Finstein. Photo by F.C. Miller.
14
-------
•END CAP
TO SAMPLER
V "
\70 L
ENGTH PROBE
PIPE NOTES
l" I D S.S. OR PLATED MILD STEEL
4
Figure 8. Construction of the gas sampling probe. The
dimensions are in inches, based on National Pipe Thread
(NPT) standards.
15
-------
(2L vacuum flask) packed with glass wool, and thence to a seven-point
multiplexing system (Figures 9, 10, 11 and 12). The multiplexing system
sequentially interrogated up to six different pneumatic inputs using a
sample pump of nominal l.QL/rain flow. A nominal 5L/min purge pump main-
tained flow in the pneumatic lines that were not being interrogated.
The seventh sample point was used as an ambient system purge between
interrogations. Thirty minute long sample cycles were initiated every
four hours (Figure 13) .
The gas sample collected by the multiplexer was split for passage
through an infrared C02 Analyzer (Beckman model 865) and a paramagnetic
oxygen analyzer (BeckmSn model 755). The multiplexer, analyzer, and
dual channel stripchart recorder (Beckman model 8720A) were housed in a
temperature-controlled cabinet.
Monitoring - manual . A clamshell type pesthole digger having a
working diameter of 12.7 cm (5 in.) was used to obtain samples. The
sample material was put into a plastic bag for transport to the labora-
tory. Determinations were initiated upon return to the laboratory,
usually within 4 hrs of sampling.
Laboratory de t e rm inat i ons . Independent "whole sample" and "non-
woodchip fraction" moisture content determinations were usually perform-
ed. The separation was accomplished by removing non-woodchip material
from the woodchips by hand. Except for Section 4 in which, the data re-
fer to the non-woodchip fraction, the data in the body of the report re-
fer to the whole sample. A comparison of the moisture contents of the
two fractions is given in Appendix A- 3.
To determine moisture content, the material was dried at 104°C to
constant weight. Moisture content is expressed on a wet weight basis
'C% moisture = wet * «0)
To prepare sample material for the pH determination, distilled
water was added to sieved material to make a slurry. The determination
was by pH electrode.
16
-------
TO GAS
ANALYZER
o
Pi
EXHAUST
S A M P I E
No I
SAMPLE
No 3
A I R
Figure 9.
The
no.
Gas multiplexing system, simplified pneumatic flow di
actual multiplexer consisted of six sampling points.
4 represents the ambient system purge.
agram.
Valve
17
-------
TO PURGE AND
SAMPLE PUM^S
n
TO SAMPLE INITIATER
SAMPLE
CYCLE
IH TERMINATION
TIMER
SAMPLE
TIMER
No.]
HN H G
SAMPLE
TIMER
No. 2
HN H G
TIMER
No. 3
HN H G
LE
R
G
r
LE
R
G
}
LE
R
G
r
N
J
r
I™
r~
P—
^
Figure 10. Gas multiplexing system, simplifipd electrical
schematic. The actual multiplexer consisted of six sampling
points. Valve no. 4 represents the ambient system purge.
-------
LEADS FROM
SAMPLE TIMERS
TO PURGE VALVE
Figure 11. Fabrication of slave relay (see Figure 10). Parts
list: Diodes (3), WER-17; resistor, IK 10% (76131RC PW1C);
capacitor, SPRAGUE TE-1509 20-150DC USA-7317H; relay, POTTER
& BRUNFIELD KA501 110 VDC.
19
-------
CYCLE
INITIATING
TIMER
POWER
IN
Figure 12. Gas sampling cycle actuation system,
20
-------
CYCLE INITIATION TIMER
CYCLE CONTROL TIMER
CYCLE CONTROL RELAY
VALVE - SAMPLE No I
VALVE-SAMPLE No 2
VALVE-SAMPLE NO 3
VALVE- PURGE
t - doiod
°-open
8 10 12 14 16 18 20 22 24 26 26 30
TIME IN MINUTES
Figure 15.
-------
SECTION 3
COMPARISON BETWEEN THE RUTGERS AT) BELTSVILLE CONTROL STRATEGIES:
OBSERVATION AND INTERPRET .TION OF SYSTEM BEHAVIOR
IMPORTANCE OF THE COMPARISON
Comparative study of the Rutgers (2-7) and Beltsville (8-11) strat-
egies has universal relevance to composting process design and control.
This is because i, composting processes are based on the self-heating
microbial ecosystem, ii. these strategies represent fundamentally differ
ent approaches to the management o£ this ecosystem, iii. the comparative
analysis yields a coherent theory of the physical, chemical, and biolog-
ical interactions governing the dynamics of this ecosystem.
The basic experimental design was to isolate strategy as the only
variable, by implementing both strategies in a common configuration
(static pile). This permitted orderly interpretation of the data. The
distinction between strategy and configuration was already drawn. (Sec-:-
tion 1) .
MATERIALS AND METHODS SPECIFIC TO SECTION 3
Ventilation
For pile 9B (Beltsville Process) the blower was operated in the
vacuum- induced direction, solely as scheduled by timer. A condensate
trap v/as interposed between this blower and its piles (Figure 14} . For
piles 7, 8, and 9A (Rutgers strategy) the blowers were operated in the
forced-pressure direction, as actuated by the temperature feedback con-
trol system.
The physical features of the piles and the ambient weather conditions
are summarized in TABLE 2, Variations in the control of the ventilation
system were as follows:
hr 500, 1.75/15
hr 500, 1.25/15
hr 120, 1.5/15; hr 120 - hr 240, 1.25/15;
hr 240 - 500, 0/1S
Pile 9B: time zero - hr 70, 4/15; hr 70 - hr 50J, 3/15
Pile 7: time zero
Pile 8: time zero
Pile 9A: time zero
22
-------
55 GALLON DRUM
WATER
OUT
fER
vs
EXHAUST
THRU
BLOWER
PUMP
WITH
FLQAT
SWITCH
Figure 14. Condensate trap used for the pile (9B) managed
according to the Beltsville process.
23
-------
Pile 7:' time zero - hr 56, position 1; hr 56-500, position 6
Pile 8: time zero - hr 500, position 6
Pile 9A: time zero - hr 44, position 1; hr 44-500, position 6
Pile 9B: not applicable
j£jHE?JlilHZ£_££Bl££ii£Zl
Pile 7: time zero - hr 500, 45°C
Pile 8: time zero - hr 500, 4S°C
Pile 9A: time zero - hr 76, 4S°C; hr 76 - hr 500, 48°C
Pile 9B: not applicable
The cross-sectional representation of the piles are given in Fig-
ures 15, 16, 17 and 18.
RESULTS
Three of the piles were controlled according to the Rutgers strategy
(piles 7, 8, and 9A), and one according to the Beltsville Process (pile
9B). The pile weights were: pile 7, 6 tonnes; pile 8, 27 tonnes; pile
9A, 36 tonnes; pile 9B, 36 tonnes. Piles 9A and 9B (Figure 19) were
assembled within two days of each other, to provide a direct comparison
of the alternative approaches to process control.
Blower Operation
Blov/er operation is represented in Figures 20, 21, 22, and 23.
Temperature feedback control (Rutgers piles only) commenced as follows:
pile 7, hr 56; pile 8, hr 10; pile 9*\, hr 10. The maximum blower opera-
tion (% time on) and the time at which the maximum operation occurred
was, respectively: 601, hr 110; 40%, hr 100, 100%, hr 96 to 135. In
pile 7 the period of feedback control ended at hr 344, at which time "de-
mand" for the blower wat less than that scheduled by the timer. There-
after, timer-scheduled operation resumed. In pile 8 the return to sch-
edule operation occurred at hr 170. The timer to pile 9A was disconnect-
ed at hr 240, eliminating the resumption of timer-scheduled blower opera-
tion when feedback control ended. Judging from the dwindling amount of
blov/er operation time, timer scheduled operation for this pile would
have resumed prior to hr 380.
The blower to the Beltsville pile was scheduled for operation 27%
of the time early in the run (Figure 23). At hr 70 this was decreased
to 201, as the 02 content exceeded that suggested for this process (9).
Pile Temperature
In the Rutgers piles the median values were a few degrees higher
than the assigned set point (e.g. for pile 7, 47°C vs 45°C) , and only 10%
to 13% of the individual observations exceeded 60°C. In the Beltsville
pile the median was 70°C, and 91% of the individual observations exceed-
ed 60°C. The observations on temperature and other parameters are summar-
ized in TABLE 3. The individual temperature graphs are given in Appendix
24
-------
TABLE 2. PILE DESCRIPTION - PHYSICAL ASPECTS AND WEATHER CONDITIONS
Item Pile 7
Trial period
Start
End
Ambient air , j.
temperature (°C) '
High
Low
Mean
Range 2
Rainfall
Amt (cm)
Events (no.)
Pile dimensions in meters
(Woodchip base) 3.0 x
21 Apr
12 May
21
9
15
to 31
9.3
10
(LxWxH)
2.7 x 0.2
(Overall1) 4.9 x 4.3 x 2.0
Pile weight (tonnes)
Perforated flexhose
Segments (no.)
Lengths (m)
Blowers (no.)
6
1
2.7
1
Pile 8
27 May
9 June
27
15
21
10 to 31
3.3
6
13x2.1x0.2
15 x 4.3 x 2.0
27
6
5.6
6
Pile 9A
9 Jul
1 Aug
32
21
27
16 to 36
13.1
7
13 x 2.4 x 0.2
15x4.9x2.1
36
6
5.6
6
Pile 98
11 Jul
1 Aug
32
21
27
16 to 36
10.0
&
14 x 5.5 x 0.2
14 x 5.2 x 2. 1
36
2
12.2
1
Direction of
ventilationi
Basis of
control§
Forced pressure Forced pressure Forced pressure
Temp feedback Temp feedback Temp feedback
Vacuum induced
Timer schedule
*The dates refer to 1980. tHigh = mean of the daily highs: low = mean of the daily lows;
mean = mean of the daily means; range = overall range of the daily highs and lows.
^Dimensions before adding a covering of woodchips 25 to 30 era thick. § Forced pressure
and temp feedback = Rutgers strategy; vacuum-induced and timer schedule = Beltsville
process.
-------
Figure 15. Pile 7, cross-sectional representation: textured area, woodchip cover
and base; clear area, sludge-woodchip mixture; circle, perforated flexhose. The
numbers indicate monitoring and control positions: thermocouples, positions 1
through 16; gas sampling probes, adjacent to positions 6 and 14; control ther-
mistor, position 6 for most of the run (see Table III-3). The blower was
operated in the forced-pressure mode.
-------
LONGITUDINAL SECTION
^
I 8
I 5
26
27
28
20
29
30
31
MAIM GR!D
'MAIN
GRID
AUXILIARY
GRID
AUXILLARY GRID
Figure 16. Pile 8, cross-sectional representation: textured area, woodchip cover
and b; se; clear area, sludge-woodchip mixture; circles, perforated flexhoses. The
numbers indicate monitoring and control positions: thermocouples, positions 1
through 31; gas sampling probes, adjacent to positions 11, 20 and 28; control
thermistor, position 6. The circles and tubes in the woodchip base represent
flexhoses. The blowers were operated in the forced-pressure mode.
-------
tsj
CO
Figure 17. Pile ~>Ar cross-sectional representation: textured area, woodchip cover
and base; clear area, sludge-woodchip mixture; circles, perforated flexhoses. The
numbers indicate monitoring and control positions: thermocouples, positions 1
through 19; gas sampling probes, adjacent to positions 7 and 16; control
thermistor position 6 for most of the run (see text). The thermocouple
grid was positioned 6m from an end of the pile. The blowers were operated in the
forced-pressure mode.
-------
Figure 18. Pile 9B, cross-sectional representation: textured area, woodchip cover
and base; clear area, sludge-woodchip mixture; circles, perforated flexhoses. The
numbers indicate monitoring positions: thermocouples, positions 1 through 15; gas
sampling probes, adjacent to positions 7 and 14. The thermocouple grid was
positioned 6m from the end of the pile proximal to the blower. The blower was
operated in the vacuum-induced mode and controlled solely by timer.
-------
J...-L..,,--
Figure 19. Pile 9B (left) and pile 9A (right). Photo by
James A. filler.
30
-------
100-f
SOD
200 300
TIME IN HOURS
400
500
Figure 20. Pile 7, blower operation. The baseline represents operation as
scheduled by timer, and the area above the baseline represents operation through
the temperature-feedback control system.
-------
z
o
u ?5-{:
&
t-
J_
z
iii
o
£50-
z
o
i-
<
er
ec
LJ25-
i?
r^J'
o
_!
S3
o-
1 1
1 !
j ,>
(
*«
> , h - "i
r^ 1 , » s
"1 r- ( " J-l
U ' 'V ' , . \_
9 , • ' *i' i, '., T-
-.1' • . -."X
•,; ' . , ' (, "' , 1
| * >• f ' ' r
Q 100 200 300
V
1
tr,
K;|
1
400 BOO
TIME IN HOURS
Figure 21. Pile 8, blower operation. The baseline represents operation as
scheduled by timer, and the area above the baseline represents operation through
the temperature-feedback control system.
-------
lOO—i
w
w
0-
,
I
0
f ',''• k
I
iOO
I
200
:.-... ,.j, ,.r,,,,,,.
l
300
I (U^-lF-l-1 ""T ^JTT7 Tniih-i-
i
400
!
50C
TIME IN HOURS
Figure 22. Pile 9A, blower operation. The baseline (hr zero to 10) represents
operation as scheduled by timer, and the area above the baseline represents
operation through the temperature-feedback control system. The timer was
disconnected at hr. 240.
-------
100-f
100
200 300
TIME IN HOURS
400
500
Figure 23. Pile 9B, blower operation as scheduled by timer.
-------
TABLE 3. SUMMARY OF THE DATA FROM SECTION .3
17
Temperature
Period (hrs)
Range (°C)
Median (°C)
Observations (no.)
% Observations
Interstitial gases
Period (hrs)
02 Range (!)
02 Median (%)
Observation (no.)
C02 Range (%)
C02 Median (i)
Observations (no.)
Moisture content (%)
Time zero (%)
Selected observation
(hr)
(%)
Process control strate
56-344
26-68
47
1102
11
10-170
-19-70
48
728
10
10-380
24-68
53
175S
13
100-500
45-82
70
1519
91
0-388 0-268
14.S-21 6.0-21
20.0 18.5
1S2 204
62
344
29
0-500
8.0-21
19.8
246
0-9
1.0
246
0-500
12-21
20.0
340
0-7
<0-0.1
240
56
190
24
67
310
33
65
500
61
For piles 7, 8, and 9A the summary is for the period of temperature
feedback control. For pile 9B the summary excludes the period of
temperature come-up.
35
-------
A-l.
In the Rutgers piles the data from the innermost vertical series
of probes (positioned generally in a line above an air duct) indicate
the establishment o£ a systematic temperature gradient in the direction
(upward) of airflow (Figures 24 , 25, and 26). In the Beltsville pile
the lower members of the innermost series (not directly in a line above
an air duct-see Figure 18) recorded a weak gradient in the direction
(d.wnward) of airflow (Figure 27). A gradient is absent at the upper
end of this series (Figure 28), which represents the pile apex. A more
well-defined gradient was established at the adjacent vertical series of
probes (Figure 29), which was directly in a line above a duct-see Figure
18). This gradient was shallow and erratic, however, compared to those
in the Rutgers piles. The temperature gradient patterns are summarized
in Figure 30.
In the Rutgers piles the temperature at the thermistor position
slightly "overshot" the set-point at around the time that blower control
passed from timer-schedule to temperature feedback (i.e. at the start of
the period of feedback control) . The relatively erratic temperature re-
cord of Pile 9A during this transition is attributed to control changes
at this time (see Pile Construction and Control, and Weather). Nonethe-
less, starting at approximately hr 140, in this pile also the tempera-
ture at the terraistor position stabilized at the set point-level. After
termination of feedback control, and with the resumption of tL.-er-sch-
eduled ventilation (piles 7 and 0 only) , the temperature declii ed rapid-
ly. Where scheduled ventilation was not resumed . (pile 9A) the temperature
fluctuated slightly, but no particular trend set in.
All of the. piles. were monitored for 02 „ but only piles 9A and 9B
were also monitored for C02- Since the three Rutgers piles behaved
similarly with respect to 02, only the observations for Pile 9A, which
involved both gases, are reported here (Figures 31, 32, 33 and 34). The
other 02 data are given in Appendix A* 2.
In the Rutgers pile the minimum 02 content (upper position) was 8%
(hr 26) , and in the Beltsville pile (upper position) this was 12% (hr
70) . Minimum QI values at the lower probes were 15% and 19% (hrs 380
and 70, respectively). The f>2 an^ ^2 plots are generally complimentary
in that relatively high values of 02 and low values of C02 occurred simul-
taneously, and vice-versa.
Moisjture Content
The moisture content data reported here are for the whole sample
(includes woodchips) . For a comparison with the non-woodchip fraction,
see Appendix A- 3.
The respective starting moisture contents were (%) : 62, 56, 67,
and 65 (Figures 35 and 36) . In the three Rutgers piles drying was sub-
stantial in that by termination the moisture content decreased to low
levels (29%, 24% and 29%). Relative to piles 7 and 8, the onset of dry-
ing in pile 9A was delayed. Once drying commenced, however, it progress -
36
-------
80
PROBE 7-iS
PROBE 7-14
PROBE 7-6
100
200 300
TIME IN HOURS
400
5OO
Figure 24. Pile 7, temperature at the innermost vertical series of thermocouples.
-------
PROBE 8-18
PROBE 8-11
PROBE 8-6 °=
PROBE 8-1
-S
J ,
0
1 •__„ II „. _ 1 1 I 1 1
iOO 200 30O 400
TIME m HOURS
1
500
Figure 25. Pile 8, temperature at the innermost vertical series of
(main grid).
iherrnocouDles
-------
80|
oSO
LJ
£T
O
t-
<
CC
?
20 -
PROBE 9AI8
PROBE 9A15
PROBE 9AI!
PROBE SAG
PROBE 9AI
;•?».„/
\-« '
!OO
200 300
TIME IN HOURS
400
500
Figure 26. Pile 9A, temperature at the innermost vertical series of thermocouples.
-------
80 —
'3
r
2o!f~
PROBE 98-10
PROBE 9B-6
PROBE 9B-I
B-
100
200 3OO
TIME SN HOURS
400
500
Figure 27. Pile 9B, temperature at the lowest three thermocouples cf the innermos-i
vertical series.
-------
PROBE 9B-I5
PROBE 9B-I3
iOO
200 300
TIME IN HOURS
400
500
Figure 28.
vertica1
Pile 9B, temperature at the highest two thermocouples of the innermost
series.
-------
80!
-------
20
10
0
-10
u
20
10
2
LU
^
LU
U.
U.
LU
SK
-------
20
UJ
S
D
§16
2
UJ
U
o:
UJ
CL
-12
z
UJ
o
o
K
O
CE
UJ
CO?
100
200 300
TIME IN HOURS
400
500
Figure 31. Pile 9A, concentrations of 02 (upper curve) and
The gas sampling probe was adjacent to
CC>2 (lower curve) .
position 7.
44
-------
100
200 300
TIME IN HOURS
400
500
Figure 32. Pile 9A, concentrations of 02 (upper curve) and
C02 (lower curve). The gas sampling probe was adjacent to
position 16.
45
-------
20
z
Ul
o
ac.
LU
Q.
UJ
o
o
E
O
o:
UJ
CO,
100
200 300
TIME IN HOURS
400
500
Figure 33. pile 9B'
C02 (lower curve).
position 7.
concentrations of (^ (upper curve) and
The gas sampling probe was adjacent to
46
-------
20
§16
or
iu
Q.
-12
2
O
100
200 300
TIME IN HOURS
400
500
Fisure 34 Pile 9B, concentrations of 02 (upper curve) and
CO (lower curve). The gas sampling probe was adjacent to
position 14.
47
-------
80
100
200 300
TIME IN HOURS
400
50O
Figure 55. Piles 7 and 8, moisture content. Samples taken from central interior
locations. The data refers to "whole samples" (includes woodchips).
-------
20
100
200 300
TIME IN HOURS
400
500
Figure 36. Piles 9A and 9B, moisture content. Samples taken from central interior
locations. The data refers to "whole samples" (includes woodchips).
-------
ed rapidly. The Beltsville pile (9B) dried only slightly, with the
terminal value being 61$.
The pH data are given in Figures 37 and 38. The pH increase of pile
9A was less than that of piles 7 and 8 and comparable to previously re-
ported piles (2) . Pile 9B was atypical in that the pH decreased.
L_e_aehajte
Liquid was last seen issuing from pile 9A on day 3, and from pile
9B on day 17.
REPRESENTATIVENESS OF PILE 9B
Piles 7, 8 and 9A represented the Rutgers strategy, whereas only
pile 9B was constructed and managed according to the Beltsville pre-
scription. The representativeness of pile 9B was assessed by comparing
its behavior with that of Beltsville-type piles reported by others (9,
38-39). The analysis (see Appendix A-4) demonstrates i. a consistency
of behavior among piles managed according to the Beltsville prescription,
ii. a consistency of behavior among piles managed according to the Rutgers
strategy, iii. dissimilar behavior between the two groups of piles. It
is concluded that pile 9B adequately represented the Beltsville Process.
GENERALIZED COMPARISON OF THE STRATEGIES
The objective of the Rutgers process control strategy is to maxi-
mize the rate of microbial decomposition. Operationally, this translates
into the optimization of temperature via controlled ventilation. For a
brief period at the outset of processing (ea. , i day) the purpose of
ventilation is to promote a rapid temperature ascent. Thus, the need to
provide QZ for heat <~ .eration, through aerobic respiration, temporarily
conflicts with the :_eed to minimize heat removal. The compromise is to
actuate the blower, by timer, on s, schedule to provide an adequate 0?
level (ca. . 54) .
When the temperature reaches a preset level (TggfJ assigned to a
controller and sensed via a thermistor in the pile, the purpose of venti-
lation changes to that of matching heat removal to heat output, such that
pile temperature is poised at a biologically favorable level (<60°C) .
This is achieved automatically through establishment of an interaction
between pile and blower, via the feedback o£ a temperature signal to the
controller. With the onset of temperature feedback control the conflict
described above disappears, as the needs of 02 supply and heat removal
now coincide. The period of feedback control ends at some indeterminate
time (typically 0-15 days), when the loss of heat independent of venti-
lation by "demand" starts to exceed heat generation.
The required ventilation capacity (e.g., blower horsepower) is de-
fined by the peak demand for heat removal, which is primarily a function
of the abundance of readily metabolizable substrate in the waste. The
forced-pressure mode of ventilation is preferred for its greater effieien-
50
-------
100
200 300
TIME IN HOURS
400
Figure 37. Piles 7 and 8, pH. Samples taken from central interior locations
-------
PILE 9A
PILE 98
—O
4 (—
..JSl
100
200
300
400
TIME
HOURS
500
Figure 38. Piles 9A and 9B, pH. Samples taken from central interior locations
-------
cyin removing heat and water vapor (7).
In the Beltsville Process the stated purpose of ventilation is to
maintain (^ at a level of from 5% to 15-t (9) • This is accomplished (8)
through use of a timer to schedule blower actuation and deactuation.
Usually the blower is operated 201 of the time, but this percentage may
be varied manually based on 02 level. Pile temperature typically peaks
at 80°C within a few days and changes little for the remainder of this
operational stage, which is fixed at 21 days. A subsequent period of in-
formal curing typically lasts several months. Blower size is standard-
ized at 1/3 hp per 45 wet tonnes of sludge cakes, regardless of the
type of sludge (e.g. raw, waste activated, anaerobically digested).
UNIFIED INTERPRETATION OF EVENTS INDUCED BY THE TWO STRATEGIES
This interpretation is based on the plug flow of air through a com-
posting matrix, and the interaction between heat generation and tempera-
ture. V/e start by considering two roles of ventilation: supplying $2
and removing heat. In turn this leads to a consideration of water re-
moval. Removal of C02 can be neglected for the present purposes.
The 02 (% vol/vol) at any given point along the airflow pathway re-
presents the balance between the rate of microbial 02 uptake and the rate
of resupply through ventilation. Although the two control strategies re-
sult in similar 02 contents, this reflects dissimilar rates of uptake and
resupply.
In the Rutgers strategy the sequence o£ control-related events is:
02 is consumed through microbial activity, which generates heat; the
temperature increases, resulting in a signal actuating the blower; venti-
lation removes heat and resupplies 0%; the temperature decreases, deac-
tuating the blower. The cycle is repeated until the period of tempera-
ture-feedback control terminates, reflecting a low rate of heat genera-
tion. In a direct sense the response of the blower is time-variable
according to the rate of heat generation, as expressed through tempera-
ture. Indirectly, however, the response is time-variable according to
the rate of 02 upta ., in that 03 uptake and heat generation both result
from organic matter decomposition. The rates of 02 uptake and heat gen-
eration are therefore directly proportional. Since the response thres-
hold (Tset;) is selected to maximize heat generation and 0? uptake, in-
trinsic to the Rutgers strategy is a high rate of ^2 uptake matched by
a high rate of $2 resupply. The result is a well-oxygenated pile.
In the Beltsville process the sequence is: slight ventilation re-
moves heat and resupplies 02 in slight amounts; the temperature ascends
to a level that suppress 02 uptake and other manifestations of raicrobial
activity. The rate of resupply of 0->, though low, suffices to maintain
an oxygenated condition because of tne slight metabolic activity. Thus,
the stated objective of maintaining 92 at a level of 51 to 15% (9) is
met by suppressing microbial metabolism (waste decomposition) through
operation at biologically harsh temperatures.
Ventilation should now be considered from the vantage point of heat
and water removal. Removal through radiation and conduction is insigni-
ficant or minor (4), and these mechanisms may be neglected for the pre-
sent purposes. Heat is transferred from the solid phase (composting
S3
-------
matrix) to the gaseous phase (flowing airstream) through the mecha-
nisms of convection and vaporization of v/ater. Vaporization is the
dominant mechanism, removing approximately nine-fold more heat than
convection (2). For the transfer to occur an enthalpy (heat content)
differential must exist between the phases. Because the matrix is the
site of heat generation, its enthalpy is elevated relative to the air.
As the air flows through the matrix it progressively accumulates heat,
decreasing the differential and thus the potential for transfer. This
set of circumstances induces the progressive storage of heat in the
matrix material along the axis of airflow, resulting in the establish-
ment 'of a positive temperature gradient in the direction of airflow.
The temperature gradient, in turn, maintains an enthalpy differential
between the phases.
Based on the relationship between the enthalpy differential and
the temperature gradient, drying in the vertical dimension is expected
to be relatively uniform. This is because the flowing air continu-
ously approaches a saturated condition yet, provided that the adjacent
matrix remains a site of heat generation, does not reach saturation.
Consistent with this expectation, repeated informal observation indi-
cated uniform drying.
In contrast, ventilation of a metabolically inert mass (no heat
generation) causes vaporization only insofar as the inlet lir is
unsaturated. This is the subject of Section 8.
Meanwhile, it is sufficient to note that in the metabolically inert
system a discrete cooling-drying front (evaporative cooling) is expected
to migrate gradually through the pile, and that this was observed in
a ventilated pile of well-curved, essentially inert, compost. Relat-
ing the data from the pile of compost to those from the Rutgers
composting piles, an estimated 95.5% of the coraposting-associated dry-
ing is attributable to heat generation and the -remainder "(4.5$) to
unsaturation of the inlet air. Similarly, a theoretical calculation
attributes 95% to 98.2% of the composting associated drying to meta-
bolic heat generation (Appendix A-S).
Thus, drying during composting is linked to organic matter decom-
position, in that the vaporization is driven almost exclusively by
heat generation. Since decomposition of putrescible material is the
primary goal of waste treatment, we earlier suggested (2-4), that the
course of drying can serve as a specific, objective, sensitive, and
convenient indicator of process performance. In this application the
moisture content test is more sensitive than the volatile solids test
(Appendix A-6). Figures 35 and 36 thus serve to represent the compara-
tive performance of the two static-pile composting processes.
DISCUSSION
These observations provide a coherent framework for composting
process control based on interactions among microbial heat generation,
temperature, vaporization, and ventilation. This is given below.
i. The composting microbial ecosystem tends strongly to self-limit
via excessive accumulation of mcrtabolically generated heat, leading to
54
-------
inhibitively high tempera tut'o. The threshold to significant inhibition
is approximately S5°-60°C, and its severity increases sh'arply at higher
temperatures. Unless controlled through deliberate heat removal, com-
posting masses typically peak at GQ°C, at which point the rate of de-
composition is extremely low.
ii. This self-limiting tendency must be countered if decomposition
is to be fostered. Consequently, the central problem in the design and
control of composting facilities is heat removal in reference to a 60°C
operational ceiling.
iii. A practical means of removing heat from the composting mass is
through ventilation. The main ventilation-associated mechanism of heat
removal (ca. 90$) is the vaporization of water. Ventilation also sup-
plies 02 for aerobic decomposition (:,iain source of heat).
iv. The forced pressure-mode of ventilation removes heat more effi-
cently than the vacuum-induced mode.
v. During composting the rate of heat generation is time-variable.
Hence, to maintain a given temperature this must be matched by corres-
pondingly time-variable heat removal. Implementation is through tempera-
ture feedback actuation of a blower, using standard control equipment.
Composting mass and blower thereby interact, to seek an assigned set-
point temperature.
vi. To achieve the desired operational ceiling of 60°C, it might be
necessary to assign a lower set-point (e.g. 4S°C).
vii. The blower capacity (head and volume) must suffice to meet peak
demand for ventilation. A strong waste (e.g. raw sewage sludge) demands
more ventilation than a weak one (e.g. digested sludge).
viii. A temperature gradient is established along the axis of air-
flow. This imposes a height limitation, above which a high rate of de-
compostion is not obtainable. With the sludge tested herein, the limita-
tion was approximately 2 meters. Drying is relatively uniform along this
axis of airflow.
ix. Managed thusly, decompostion and drying are related in that the
following chain of causation is established: decomposition generates
heat; the heat vaporizes water; the vaporization causes drying. Pro-
duction of metabolic water replenishes only ca. 10% of the water re-
moved.
x. A consequence of this control strategy is that the composting
mass is well-oxygenated (typical 0? level, 171 v/v). This is because
approximately 9x more air is needed to remove heat than to supply 0? for
aerobic respiration.
This framework results from a comparative analysis of two funda-
mentally different approaches to the management of the composting-micro-
bial ecosystem. These are represented by the Rutgers strategy and the
strategy embodied in the Beltsville Process. It may be doubted whether
a different analytical point of departure could lead to a coherent frame-
SS
-------
work.
The study material was sewage sludge, and the configuration that of
unenclosed static pile. The derived principles are nonetheless relevant
to other materials and configurations. The static pile configuration,
however, is structurally and operationally simple, and is therefore pre-
ferred.
56
-------
SECTION 4
SEQUENCE OF LIMITATIONS INDUCED BY CONTROL STRATEGY
INTRODUCTION
The Rutgers strategy is to maximize the rate of decomposition
via heat removal in reference to temperature. Since the main heat
removal mechanism is the vaporization of water, a strong drying
tendency is induced. Consequently, the prevention of inhibitively
high temperature promotes dryness, possibly to an inhibitive extent.
Conversely, the Beltsville process focuses on the maintenance
of a minimal level of interstitial 02. This leads, by default, to
inhibitively high temperatures and tne retention of water.
The present Section concerns dryness as a limiting factor. The
experimental observations support a general discussion of the sequence
of limitations, and its practical implications in sludge management.
MATERIALS AND METHODS SPECIFIC TO SECTION 4
These observations concern one pile of the sludge-woodchip
mixture (Figure ?9), managed according to the Rutgers strategy. Depar-
tures from"the Materials and Methods described in Section 3 are that
the sludge-woodchip ratio was' 1 tonne sludge per l.Sm^ woodchips (1
ton per 1.8 yd^); the infrared analyzer for C0» determination was a
Beckrnan Model 315; the material tested for moisture content had been
passed through a 0.64m (J in.)-sieve (ASTM E-ll specification).
The trial period was 10 May to 25 June, 1979. Ambient air
temperatures during this period were (°C): mean of the daily highs,
24.6; mean of the daily lows, 14,5; mean of the daily means, 19.6;
range of the daily highs and lows 9 to 31. Rainfall amounted to 26 cm
in 20 occurrences.
The pile is represented by Figure 39. On four occasions (see
arrows, Figure 40) water was added to the central portion of the pile,
through various procedures. Occasion 1: approximately 44SL (120 gal-
lons) of water was pumped through a gas sampling probe positioned
variously in the central portion of the pile. Occasion 2: four 176L
(20 gallon) containers filled with water were carried to :he top of
the pile and emptied. Occasion 3: water was applied to the pile sur-
^ce with a ^'re hose, at a low rate, for approximately 90 minutes
Occasion 4 watc was applied to Chi pile surface with a fire hose,
• a irod^-i.ie r..te, for approxima \y 10 minutes.
57
-------
Figure 39. Pile 4A, cross-sectional representation: textured area, woodchip cover
and base; clear area, sludge-woodchip mixture; circle, perforated flexhose. The
numbers indicate monitoring and control positions: thermocouples, positions 1
through 14; control thermistor, position 1. The blower was operated in the
forced-pressure mode. (This pile was designated 4A in reference 4, and A in
reference 2.)
-------
100
100 200 300 400 500 600 700 800 9OO 1000 1100
TIME IN HOURS
Figure -10. Pile 4A, blower operation,, and temperature at the control thermistor. The
baseline represents operation as scheduled by timer, and the area above the base-
line represents operation through the temperature-feedback control system. The
time of water addition is indicated by arrows.
-------
RESULTS
Hour Zero to Hr 490
Blower operation first exceeded that schedule by timer at
hr 12, indicating that the controller had responded to a thermistor
temperature of >4S°C (Figure 40). This marks the start of the period
of temperature-feedback control. Blower operation v/as nearly contin-
uous from hr t'O to 150. Feedback control terminated at hr 352, at
which time blov/er operation reverted to that scheduled by timer (7%
of the time)„
Based on 1020 observations during the period of feedback control,
the median pile temperature was 48°C, and the range was 25°C to 63°C,
Temperatures in excess of 60°C accounted for 1.81 of these data points.
The temperature at the thermistor control position is shovm in
Figure 40.
The level of C0? peaked at 144 just prior to the start of the
period of feedback control. Thereafters CO,, generally varied between
2% and 4*. L
The starting moisture content of 761 decreased to ?.?.$ in IS
days (Figure 41),, This refers to sieved material (woodchips removed).
Hour 490 to_Hr^lI4j)
Immediately prior to the first water addition the temperature at
the thermistor position had declined to well below the set point value
(24 C observed vs. 45°C set point), and blower operation was solely
as scheduled by timer. At this time the moisture content was 22%.
Water was added, as described above, to the part of the pile
containing the control and monitoring devices (see Figure 39). The
material absorbed water slowly and non-uniformly, therefore repre-
sentative samples for the determination of moisture content were not
obtainable in this part of the study.
Approximately ten hours after the first water addition the
temperature re-ascended, initiating a second period of feedback control.
Before the termination of this period the second \iater addition was
made. Shorrly thereafter feedback control ceased, and blower opera-
tion by timer-schedule resumed., The third and fourth water additions
initiated - lependent., successively weaker, periods of feedback control.
The comparative intensity of the successive periods of feedback
control may be judged on the basis of the amount of blower operation
time. Baseline blower operation, as scheduled by timer, is included
in this comparison. Setting blower operation during the original
period (hr 10 to hr 348) equal to 100, the other values are: hr 519
to hr 69, 20.2; hr 872 ro hr 946, 5.9; hr 1044 to hr 1070, 1.8.
During the part of the trial from hr 490 to hr 1140, rainfall
amounted to 10.8 cis in 8 occurrences. The heaviest rainstorm,
60
-------
100
200 300
TIME IN HOURS
400
500
Figure 41. Pile 4A, moisture content. Samples taken from central interior locations.
-------
depositing 6.2 cm, started on hr 735 and ended on hr 755. This
storm occurred between periods of temperature-feedback control (see
Figure 40). The rainfall wetted the pile only superficially, and
did not revive blower demand.
DISCUSSION
Four factors can come to limit microbial activity in a compost-
ing mass. These are: 0, depletion; development of excessively high
temperature; water depletion; substrate depletion. Except for sub-
strate, whether a particular limitation comes into play or is bypassed
depends on the means of ventilation employed and whether water is
added (.Figure 42) .
The sequence is discussed in reference to specific examples,
Oxygen depletion is exemplified in the composting of leaves in static
piles without mechanical ventilation, as reported elsewhere (40).
A summary of the pertinent observations follows.
Soon after assembling leaves into a pile of substantial size,
e.g., L x W x H ° 7.6x3.1x1.8m (25x10x6 ft), the interior portion of
the pile became 0.,-deficient (defined as 0-, not detectable or barely
detectable). The 02-deficient portion comprised roughly half the
total pile volume. The outer portion (roughly half the volume) re-
mained well-oxygenated. The maximum temperature throughout the pile
rarely exceeded 60°C, presumably reflecting the slight generation of
heat fermentatively compared to O^-based metabolism. Over the ten
month observation period the pile shrunk to approximately half its
original volume. The relative proportions of 0-,-deficient and oxygen-
ated volumes remained approximately equal.
At termination the pile was disassembled for examination. The
material in the oxygenated zone was damp and appeared to be humified,
whereas in the central core the material was wet, gave off an odor of
putrefaction, and retained evidence of the original leaf structure.
Thus, overall the leaves had not yet received adequate treatment.
Nonetheless, it was concluded that satisfactory composting of
leaves is possible without mechanical ventilation (or agitation),
For this type of waste the need is to insure that the material in the
innermost core becomes oxygenated within the processing time available,
in that this signifies that all of the material has undergone thorough
decomposition. Provided that the leaves are moist at the outset,
adequate decomposition throughout occurs over winter.
An inadequate ventilation system, in terms of blower capacity
and/or control system, leads to inhibitively high temperature, whereas
an adequate system leads to dryness and/or substrate depletion. These
circumstances are represented by the Beltsville Process and the Rutgers
control strategy, respectively, as was developed earlier.
The practical implications of this sequence are illustrated
through three hypothetical cases. The first concerns a "dirty sludge,"
contaminated with heavy metals and/or non-biodegradable industrial
62
-------
MATERIAL ASSEMBLED
FOR COMPOSTING
NO FORMAL VENTILATION
SYSTEM
INAPEQUATE VENTILATION
SYSTEM
ADEQUATE VENTILATION
SYSTEM
OXYGEN BECOMES
THE LIMITING FACTOR}
SYSTEM DEOXYGENATED
JL
TEMPERATURE BECOMES
THE LIMITING FACTOR;
SYSTEM OXYGENATED
TEMPERATURE DOES NOT
BECOME THE LIMITING
FACTOR; SYSTEM OXYGENATED
WEAK TENDENCY
TO DRY
STRONG TENDEP4CY
TO DRY
STABILIZATION
PROGRESSES SLOWLY AS
OXYGEN PENETRATES
THE MASS
POORLY
STABILIZED
ORGANIC
RESIDUE
WATER NOT ADDED
ACTIVITY DIMINISHES
AS WATER BECOMES
THE LIMITING FACTOR
WATER ADDED
ACTIVITY DIMINISHES
AS SUBSTRATE
BECOMES DEPLETED
Figure 42. Reprinted bv permission from BIO/TECHNOLOGY, Vol. 1,
No. 4, pp. 347-353. Copyright © 1983. Nature Publishing Co.
Limitations to biological activity induced by different manage-
ment strategies.
63
-------
chemicals. Such a sludge should not be considered as a feedstock
for compost production, yet composting might be the treatment process
of choice. The process residue might be suitable as a landfill cover
material, or as a low grade solid fuel. This residue is lesireable
as a fuel, comps.red to the uncomposted sludge cake (Appendix B) .
The second and third hypothetical cases involve a "clean sludge,"
which affords a wider range of opportunities for ultimate disposal
via resource recovery. Consider the production of a "rough" compost
for restricted bulk application to soil. This product must be aes-
thetically acceptable and remain so after application tr soil,-hence
it should be moderately well stabilized. More extensive stabilization
than required to meet this need is undesireable as it adds to process-
ing costs and results in a loss of agronomically valuable nitrogen and
organic matter. Production of a "rough" compost calls for a relatively
high initial moisture content (consistent with reliable process "start-
up"), to prolong the period of microbial action prior to dryness.
Finally, production of a highly stabilized compost for un-
restricted distribution calls for more extensive biological decomposi-
tion. This might be accomplished in-place, as an extension of a high
rate stage, by adding water to prevent premature dryness. In this
manner the high rate stage v/ould gradually pass to an in-place curing
stage. Alternatively, the material might be pennitted to become dry
in the high rate stage and then moved, and remoistened, for curing.
Other unit process flow schemes can be envisioned.
64
-------
SECTION 5
RUTGERS STRATEGY: REPLACEMENT OF WOODCHI PS WITH
RECYCLED COMPOST AS THE BULKING AGENT
INTRODUCTION
Composting requires gas exchange to remove heat and water vapor
and to supply 02> Mechanical agitation can provide the needed ex-
change only intermittently, and at high energy cost. Thus, agitation
does not lend itself to rate maximization through controlled heat
removal, and its main role in composting is to mix and abrade the
material. Ventilation by blower affords the only practical approach
to rate maximization, but this imposes a requirement for porosity to
permit the passage of air. Since sludge cake by itself lacks porosity,
it is commonly mixed with a "bulking agent" having this property.
The usual bulking agent, woodchips, has serious drawbacks. In
routine Beltsville-type operations, the purchase of woodchips and
associated operations (storage, translocation, mixing, screening)
represent perhaps one-third of the overall costs (41). Furthermore,
woodchip stockpiles are colonized by Ajjjergillus fumigjrtu_s_, a fungus
which can infect the human lung. It wouTdFe~3e'sTre¥ble, instead, to
use internally generated recycled compost as the bulking agent, while
retaining the structural and operational simplicity of the static pile
configuration.
To dp so, the recycle must consist of i) stable aggregates in
the physical sense, to impart porosity, ii) highly stabilized material
in the sense of supporting only slight metabolic heat generation, to
reserve most of the ventilation system's heat removal capacity for the
fresh sludge, iii) dry material, to absorb water from the sludge to
improve porosity. Furthermore, the composting process itself should
promote drying so that once composting is initiated, porosity pro-
gressively improves. These are precisely the tendencies intrinsic to the
Rutgers strategy, hence this trial.
MATEuIALS AND METHODS SPECIFIC TO SECTION 5
Screened material (woodchips removed) from piles 8 and 9A \vas
used as the bulking agent. Screening was by means of a Royer Model 355
shredder-mixer .coupled with a Mogensen sizer (Royer Foundry and Machine
Co., Kingston, PA). After the screening the material was stockpiled
in the open for approximately 4j and 3 months, respectively, without
deliberate remoistening. At the time of use the pile was cool, the
bulk of the material had a moisture content of 391, and the material
had a slight earthy aroma.
Fresh sludge ct'ke and compost were fed by separate conveyer belts
into the pug mill for mixing. The feed rates were adjusted by eye to
65
-------
yield 6ne batch with a low recycle ratio (sludge-rich), one batch
with an intermediate ratio (roughly equal proportions), and one
batch with a high ratio (compost-rich). The ratios actually
obtained were determined by calculation bascid on the moisture content
of the sludge (74%), the compost (391), and the mixture (TABLE 4.).
Dry weight
Approx pile recycle ratio
wet weight (compost/compost
11A 5.4 0.3
11B 4.5 0.6
11C 3.6 0.8
Initial
+ Initial mois- approx. ,
69.5 5.3
61.5 5.3
52.0 5.3
Each compost-sludge mixture was formed into s. pile over parallel
segments of perforated flexhose within a woodchip bed (Figure 43),
fitted with control and monitoring devices (Figures 7 and 44), and
covered with woodchips (Figure 45). (It was thought advisable to
insulate these small, free-standing, piles with a woodchip cover.)
Each pile was ventilated with one 1/3 hp blower operated in the forced
pressure mode. Process control was based on temperature feedback,
with the thermistor located at position 1 (Figure 43), and the tem-
perature controller was set to 45°C. Timer-scheduled operation was
0.75 min (uninterrupted) per IS min.
The trial was started on 24 Oct 1980 and terminated on 7 Nov,
at which time the woodchip cover was removed (Figure 46). Ambient
air temperatures during this period were (°C): mean of the daily highs,
14 ; mean of the daily lows, 4°; mean of the daily means, 9°; range
of the daily highs and lows, -2 to 21°. Rainfall amounted to 8.0 cm
in 3 occurrences.
RESULTS
Pj. c t o ri a 1_Reprg_servtat_io n
The periods of temperature feedback control ended on hrs 228,
186, and 102, respectively (see Blower Operation, below), On hr 330
the woodchip cover was removed from all of the piles. A pictorial
overview is provided by Figures 44 and 46, Other pairs of photos give
"before and after" closeups of each pile (pile 11A, Figures 47 and 48;
pile 11B, Figures 49 and 50; pile 11C, Figures 51 and 52). Shrinkage
in volume is evident, being greatest in pile 11A, intermediate in
pile 11B, and least in pile 11C. This is also the order of sludge (and
water) abundance in the mixtures at time-zero.
66
-------
Figure 45. Piles 11A, 11B, and 11C, cross-sectional representation: textured area,
woodchip cover and base; clear area, sludge-compost mixture; circles, perforated
flexhose. The numbers indicate monitoring and control positions: thermocouples,
positions 1 through 12; control thermistor, position 1. The gas sampling probes
were positioned as follows: pile 11A, adjacent to positions 2 and 7; pile 11B,
adjacent to position 2; pile 11C, adjacent to position 7. The blower was
operated in the forced-pressure mode.
-------
Figure 44. Piles 11A, 11B, and lie, before composting. The
dowels extend from the pile, and the thermocouple leads are
organized at subjunction boxes. The thermistor was at the
end of the lowest dowel. Modified oil drums served to cover
the blowers at the rear of the piles. Photo by F.C. Miller.
68
-------
Figure 45. Same as Fig. 44
Photo by F.C. Mills?. '
covering with woodchips,
69
-------
70
-------
1gy"^ -~^~^-*~~
IH -ill
'IM'
if.
1 ~ t- i
R./F'^.J' '-1
Pila 1U, before costing. Photo by p.c. Mi
71
-------
— _ _ _
If
'r
•*> ?
^
to-** I.-,'9 '- *%--''»
X. ^ J"1 > 1
^
^
f
j
t- 4
n
*"V f
r1
f ''
'
i-- • x , «
A ' K:> /'
' {C ~-^ w /
1 "
V— 1^'
*JS sv>'--!^ r^ J> '
-------
$ J,'-'1 „' -
x^, -
49. Pile 11B, before composting. Photo by F.C. Miller,
73
-------
n
1(0
ii"
-T;
^f
. ^i
p.
^i
i ^ /•"? ••
ii -» *
, I \* X *
• f *-» M-'^i
^ f ^.. . l^
o
»*
t:
I,
1C-'"
*" -*s
K
- ^
- ^-^—
-V
^ d
.- *, 'j^r! f'ii
_*r~\ ^f>'^ 5"''
'; !' <":H-3 ? J>'j f 1
^'%U:/'"^"^,
-^.,.^-?^^?^rv
- '^i '""' ' -''*
\ i
^-*«iv" -:^"5J:TJ
|cv^^ii t^ii>«
P " ' T
"J*^
fr
,«» tv
. 1S?f\'
' '"< f
PJgure 50. Pile 11B, after composting,
been removed. Photo by F.C. Miller,
The woodchip cover has
74
-------
ni
U,
•
X LF-
ll!
xr
., N
f- I ]H
JL
Figure 53. Pile 11C, before composting.. Photo by F.C. Miller.
75
-------
II
E|J,-_
'
nt,
-.
-
k if •
- , : - ,--,-
Figure 52.
cover
76
-------
The duration of temperature feedback control (Figures 5?, 54 and
S3) was longest in pile HA (hr 22 to 22B~) , intermediate in pzle 11B
(hr 16 to 186) and briefest in pile 11C (hr 14 to 102). In piles HA
and 11B the peak demand for ventilation utilized approximately 551 of
the blower capacity. The comparable value for pile 11C was 25%. The
comparative overall demand for ventilation was assessed in terras of
the total blower operation time during the period of feedback control
(baseline included). Setting the greatest demand equal to 100, the
values were: pile HA, 100; pile 11B, 92; pile 11C, 23.
Pile Temperature
The temperature plots for the thermistor and overlying positions
are given in Figures 56, S7, and S8, At the thermistor position the
set point temperature (45°C) was precisely maintained during the period
of feedback control. With some exceptions, notably at probe 11B6, the
higher positions experienced higher temperatures in the usual pattern
of a systematic gradient. The gradient was most clearly established
in pile 11C (compost-rich). In other parts of the pile the tempera-
ture control was less precise (see Appendix C). The temperature
observations at all of the positions during the feedback period, and
other data, are summarized in TABLE 5.
Pile Atmosphere
The lowest 02 levels recorded during the period of feedback
control were as follows: HA (high probe position) r 14.51; HB (low
probe position), 16.8%; 11C (high probe position), 13.3%. The cor-
responding peak CO., levels were 2.8%, 1,84 and 6.8% (Figures 59, 60,
61 and 62).
Moisture Cont_erit
The initial moisture contents (HA, 70%; lib, 61%; 11C, 52%) re-
flected the differences in the sludge-compost mix ratios (Figure 63),
Water losses were positively correlated with the amount of sludge in
the mix. The minimum moisture contents, observed prior to the ter-
minal observation, were: pile 11A, 29%; pile 11B, 21%; pile 11C, 23%,
The terminal values were slightly higher.
Visual, Tactile, and Aesthetic J^ujy.ijj.ej;^
At termination, the piles were bisected and examined. The material
comprising the outer rim of the sludge-rich pile HA ('-he "toes") was
wet and pasty, with an unpleasant odor. This material had not com-
posted appreciably. The bulk of the pile however, had composted ex-
tensively', was dry, and had a greyish cast seemingly imparted by my-
celial growths, This part the pile consisted of chunks of material,
ranging in size from that of golfballs to boulders. The individual
units v/ere resistant to breakage by hand. Unlike the outer rim, the
bulk of tha material had no conspicuous odor.
77
-------
In..,,
200 300
TIME IN HOURS
400
soo
Figure S "S Pile 11A, blower operation. The baseline represents operation as
scheduled by timer, and the area above the baseline represents operation through
the temperature-feedback control system. The blower was operated in the forced-
oressure mode.
-------
100-
200 30O
TIME IN HOURS
400
Figure 54. Pile 11B, blower operatior . The baseline represents operation as
scheduled by timer, and the area above the baseline represents operation through
the temperature feedback control system. The blower was operated in the forced-
pressure mode.
-------
CO
O
icoi
Z
O
O
t-
3
O
ii r N
±< J ' s
>J r ,--'1
100
200 30O
TIfc3E IN HOURS
500
Figure 55. Pile lie, blower operation. The baseline represents operation as
scheduled by timer, and the area above the baseline represents operation through
the temperature feedback control system. The blower was operated in the forced-
pressure mode.
-------
PROBE IIAIO
PROBE HAS
PROBE IIA I —
•N-x
j"
0
L 1 l 1 1 S i
SOO 200 300
TIME IN HOURS
1
400
I
500
Figure S6.
Pile 11A, temperature at the innermost vertical series of thermocouples.
-------
PROBE IIBIO
PROBE IIBS
PROBE IIS I
100
200
300
TIME IN HOURS
400
• j cure
Pile 11B, temperature at the innermost vertical series of thermocouples,
-------
80 5-
PROBE IICIO
PROBE I!C 6
PROBE !IC I
20;
* '' »»
\—* %
R_ X- ^. S -X
1 ^^/ \^-
\ ^*+J^
r
L -I j i i i S ( S i
3 100 200 300 400
TIME IN HOURS
!•
f
\
f
i
5C
)0
Figure 58.
Pile 11C, temperature at the innermost vertical series of thermocouples,
-------
TABLE 5. DATA SUMMARY FOR SECTION s'
Pile
Recycle ratio
Temperature (during period
Period (hr)
Range (°C)
Median (°C)
Observations (no.)
% observations * 60°C
Interstitial gases (time 0
0. Range (1)
Li
02 Median (%)
Observations (no.)
C02 Range (%)
C02 Median ($)
Observations (no,)
Moisture content ($)
Time- zero
150 hr
11A
0.3
of feedback control)
22-228
29-73
53
624
24
to hr 332)
12.8-21
20.8
168
<0.1-8.6
<0.1
168
70
32
11B
0.6
16-186
9-71
46
516
20
16.8-21
20.3
84
<0.1-1.8
0.2
84
61
28
11C
0.3
14-102
11-75
59
264
46
13.3-21
20.8
84
<0. 1-6.1
<0.1
84
52
24
Rutgers process control strategy applied to all of the piles. Dry
weight recycle ratio s recycle/recycle * fresh sludge.
84
-------
20
5
=>
§'6
-------
20
UJ
2
§16
z
UJ
o
-------
20
UJ
5
D
§16
UJ
o
K.
-12
z
o
t-
<
CE
I—
Z
UJ
o
o
cr
O
CO,
100
200 300
TIME IN HOURS
400
.00
Figure 61
Pile 11B,
CCU (lower curve)
position 2 .
concentrations of
The gas sampling
02 (upper curve) and
probe was adjacent to
87
-------
20
o
ce
UJ
0.
z
o
t-
-------
30
PILE IIA
PILE IIB
PILE 1IC
SOO
200 300
TIME IN HOURS
400
SOO
Figure 63. Piles 11A, 11B, and 11C, moisture content. Samples taken from central
interior locations.
-------
At termination the material in piles 11B (intermediate) and
11C (compost-rich) was uniformly dry, granular, and brown. The
material in both of these piles had a slight earthy aroma.
DISCUSSION
A noticeable difference in behavior attributable to the replace-
ment of woodchips with compost is that, lacking the rigidity o£ the
woodchip matrix, these piles shrunk considerably in volume. Other
differences that might also be attributable to the use of compost as
the bulking agent are: a relatively brief period of temperature feed-
back control (faster processing); less precise control of temperature;
faster drying.
The major significance of this trial is that the possibility
of using recycled compost as the bulking agent in static pile con-
figuration was demonstrated. Others have demonstrated use of recycled
compost in this capacity in conjunction with "windrow composting,"
in which the mass is mechanically agitated (42). Agitation is energy
intensive, however, and affords little control over temperature and
oxygen.
It was anticipated that pile HA would fail, since it v/as so
rich in sludge (recycle ratio, 0.3), Yet, the bulk of the pile com-
posted satisfactorily, judging from all of the parameters. The outer
material was isolated from the airstream, and did not compost. For
routine use higher recycle ratios, as represented by piles 11B and 11CS
are indicated. Nevertheless, the performance of pile HA indicates a
degree of processing resilience in conjunction with the use of re-
cycled compost.
Owing to compactability leading to greater resistance to air-
flow, sludge-compo.st mixtures may be more subject to a height re-
striction than sludge-woodchip mixtures. This also indie?*fts use of
higher recycle ratios. Further field experimentation is .. dded to
define the height limitation.
90
-------
SECTION 6
COMPARISON BETWEEN THE RUTGERS STRATEGY AND BELTSVILLE
^ROCESS: MATERIALS BALANCE
INTRODUCTION
The outcome of processing in terms of materials balance impacts
on facility design with respect to composting area, storage area, and
machinery needs for materials handling and transport. Moreover the
amount and nature of the process residue, relative to the sludge,
strongly influences the possibilities for ultimate disposal/resource
recovery. As such, materials balance is a major determinant of
construction and operating costs, disposal options, and indeed of the
utility of composting as a waste treatment technology.
In Section 3, the Rutgers and BeltsviTle approaches were compared
in terms of blower operation, temperature, i , and C02 levels, and
moisture content. This comparison is extended herein, based on certain
of these data and related analyses. Also, materials balance is esti-
mated for the piles described in Section 5. These employed recycled
compost as the bulking agent, and were managed according to the Rutgers
strategy.
PROCEDURE FOR CALCULATING AIRFLOW AND MATERIALS BALANCE
An airflow delivery of 807 m3/hr (28,500 ft3/hr) per blower was
assumed, based on the manufacturer's specifications at 5.1 cm (2 in.)
of water head.
The mass of the mixture after composting was calculated based on
the equation:
where:
M- s initial mass
MC. - initial moisture content
MC,p s final moisture content
k = 8.07 (Rutgers strategy), or 8.33 (Beltsville Process),
The constant k is the ratio mass water vaporized/mass solids decomposed,
as derived in Appendix A-6. With air exiting from the composting mass
at. 60°C and 100% relative humidity (representative of the Rutgers
strategy), the constant is 8.07; at 70°C and 1001 RH (representative
of the Beltsville Process), it is 8.33. Mixture refers to sludge and
woodchips , or sludge and recycled process residue ("compost").
91
-------
Rearranging:
M „ M.[k(l-MC.) - MC.]
(ii)
Since the values for M. and MC- are known, the initial weights of dry
matter and water can be calculated. Having derived Mr., and knowing
MC.p, the final weights of dry matter and water can be calculated. At
this stage the estimates refer to the mixture of sludge and woodchips ,
or sludge and recycled compost.
With respect to the sludge-woodchip mixture the estimate was
converted to total sludge solids decomposition, and to sludge volatile
solids decomposition, based on a sludge-woodchip ratio o£ 1 tonne
sludge cake per 1.7m3 woodchips (1 ton per 2 yd3) , and the following
nominal characteristics: sludge cake moisture content, 75.51; sludge
cake volatile solids content, 73.51; woodchip bulk density, 234 kg/ra3
(394 ib/yd-5) (43). Also, it was assumed for the purpose of calcula-
tion that the woodchips did not decompose. To estimate the volatile
solids decomposition of the mixture of sludge and recycled compost, the
compost was assigned a volatile solids content of 54.5% based on values
given in TABLE 6.
AIRFLOW AND MATERIALS BALANCE ESTIMATES
The airflow and materials balance estimates for the piles using
woodchips as the bulking agent are given in TABLE 6. For piles 7, 8,
and 9A, the mean total air delivery -during 12.4 days of composting was
9,9SOm3/tonne (319,000 £t3/ton). For pile 9B this was 20.8 days and
2330 m3/tonne (74,500 ft3/ton) . Thus, mean air delivery to the Rutgers
piles was 4.3x greater. The comparable values for mean and peak
.delivery were Sx greater and 15x greater, respectively. The need for
more air reflects the more extensive waste decomposition.
As is characteristic of feedback control (Rutgers) , mean and peak
usage differed, being 37.0 and 95. Im^/tonne-hr , respectively (1,190 and
3,050 ft3/ton-hr) „ With timer control (Beltsville) mean and peak
differed slightly, only because th-? blower schedule was adjusted manu-
ally. This was as prescribed (9) , in response to high levels of 02.
With application of the Rutgers strategy, a mean of 16.01 of
the sludge-woodchip mixture was decomposed in 12.4 days. The compar-
able estimate for the Beltsville Process is 4.3%, in 20.8 days. Assum-
ing for calculation purposes that the woodchips were not decomposed
microbd.ally (a reasonable approximation at these temperatures and
times), total sludge solids decomposition amounted to 41,8% (mean
value) and 11.3%, respectively. In terms of sludge volatile solids
the estimates are 56.9% and 15.4%. Regardless of the fraction (sludge-
woodchips mixture, or total sludge solids , or sludge volatile solids),
the Rutgers strategy induced an estimated 307x more decomposition in
approximately half the time. With respect to v/ater the removals
amounted to 78.21 and 19.4%, indicating 4x more removal in approxi-
mately half the time.
92
-------
*
TABLE 6. EFFECT OF CONTROL STRATEGY ON AIR USAGE AND MATERIALS BALANCE
_ .._ . . ..
Period (hr)
•?
Air delivery (m /initial wet
Total
Mean (hr"l)
P O Jl V1 r Tl T* "" ^ ^
i JL
Air delivery (ft /initial wet
Total
Mean (hr"1)
Peak (hr"1)
Pile 7
0-344
tonne)
13,100
38.1
81.2
ton x 10
421
1.22
2.60
Material decomposed (i of initial dry
Overall sludge-woodchip
mixture
Sludge total solidst
Sludge volatile solids*
Water removed (4 of initial)
15.9
41.6
56.6
78.9
Process
___Rutj£erj3___
Pile 8
0-170
7,180
42.1
71. 2
3)
230
1.35
2.28
weight)
11.8
30.8
41.8
74.8
control strategy^"
Pile 9A
0-380
9,580
30.9
133
307
0.990
4.28
20.4
53.1
72.3
80.9
Beltsville
Pi.le 9B
0-500
2,330
4.7
5.6
74.5
0.149'
0.192
4.3
11.3
15.4
19.4
Certain entries given in both metric and English units.
For piles 7 and 8 the calculation is based on the period from time-
zero to the cessation of temperature feedback control ( = resumption
of timer-scheduled blower operation). For pile 9A it was based on
the period from time-zero to hr 380 (scheduled operation not resumed
because the timer was disconnected during feedback control). For
pile 9B it was based on the period from time-zero to hr 500. Where
necessary, the corresponding value for MCf was derived through inter-
polation.
The calculation was made as if decomposition of woodchips were nil.
93
-------
The materials balance estimates for the piles using recycled
compost bulking agent are given in TABLE 7. Performance was compar-
able to that obtained with woodchip bulking agent in conjunction with
the Rutgers strategy (compare to TABLE 7, piles 7, 8, and 9A).
Except that air delivery to pile 11B was somewhat high, the pattern
of results is consistent with the strength of the mixtures. Thus
pile 11A, which had more fresh sludge than pile 11C, exerted a
greater demand for ventilation and lost more solids and water. Being
richer in sludge, its potential for heat generation, blov/er demand,
and vaporization was greater.
DISCUSSION
Composting is a robust process such that it is possible (though
not desireable) to design and operate a facility without benefit of
a coherent process control strategy. The process1 robustness stems
from two factors. First, microbial self-heating (the underlying
phenomenon) commences spontaneously, even if conditions of nutrition,
moisture content, and gas exchange are only marginally adequate. This
reflects the non-specificity, ubiquity, and rapid growth of the microbes
capable of initiating the process and carrying it forward (18). Second,
composting is resistant to outright process failure (24), This stems
from the tendency of the climax population, apparently consisting of
thermophilic members of the genus Bacillus (44-45) to elevate the
temperature to the edge of its tol¥FIHc¥~Timit (^ 7S°C). In this
state decomposition proceeds, but at a mere fraction of the rate
achievable at 60°C and less.
This is evidenced herein in that the Rutgers strategy, compared
to the Beltsville strategy, induced 3.7x more decomposition and 4.Ox
more water removal in approximately half the time. 'This outcome reflects
the fundamental difference between these approaches. The Rutgers
strategy is to maximize the rate of decomposition through ventilative
heat removal in reference to temperature. The Beltsville process, by
default, suppresses decomposition through inhibitively high temperature.
The Rutgers strategy is expected to result in raor' cost-effective
composting in terms of facility construction and routi^d operation. The
reasoning is that i. less facility time/space is required, ii. less
process residue is produced, decreasing materials handling operations,
iii. the residue is easier to handle, store, and transport, Morsover?
this strategy improves composting's utility as a waste treatment tech-
nology, owing to the production of a process residue that is more
amenable to resource recovery/ultimate disposal.
-------
TABLE 7. RECYCLED PROCESS RESIDUE AS BULKING AGENT: AIR USAGE AND
MATERIALS BALANCE*
Period (hr)
Air de]ivery (m /initial wet
Total
Mean (hr'1)
Peak fhr-1}
Air delivery (ft /initial wet
Total
Mean (hr'1)
Peak fhr-lj
11A
0-228
tonne)
9,770
42.8
86.2
ton x 11
313
1.37
2.76
Material decomposed (1 of initial dry
Overall sludge -recycle
mixture
Volatile solids
Water removed (% of initial)
24.7
36.4
85.4
Pile
11B
0-186
10,100
54.3
119
>3)
323
1.74
3.83
weight)
14.8
23.8
76.1
11C
0-102
3,370
33.1
57.1
108
1.06
1.83
7.8
13.4
58.3
Rutgers process control strategy applied to all piles. The calculation
is based on the period from time-zero to cessation of temperature feed-
back control (resumption of timer-scheduled blower operation). Certain
entries given in both metric and English units. *
-------
SECTION 7
*
MATHEMATICAL DESCRIPTION OF PROCESS CONTROL DYNAMICS
In this Section composting process control dynamics are reduced
to mathematic form. The mathematical description focuses on the
generation and removal of heat; an extra-mathematical constraint is the
effect of temperature on heat generation. Tiiis development leads to
testable predictions of system behavior.
Overall heat removal may be expressed in units of energy/time:
fit 3 ficonv
where: (J*. - total heat removal
3 convective heat re
- conductive heat removal
Q. 3 convective heat removal
cor,j
^r =• radiant heat removal.
A calculation based on a ventilated field-scale mass indicates that
Qconj .'o 0.02 Q , and that Q is small and sometimes exceeded by radiant
gains (4). Thus, all but convective removal may be neglected in the
analysis .
The relationship governing convective removal is:
Sconv s 2 £out - hin) Cii)
where: m - dry air mass flow (mass dry air/time)
h K outlet air enthalpy (energy /mass dry air)
""OU 1C
_h. ° inlet air enthalpy (energy /mass dry air).
The enthalpy of the air is a function of its temperature and the amount
of water vapor it contains. Note that to maintain a quasi-steady state
(temperature - constant) , -Qt (and hence QCQnv) must match heat genera-
tion. Note also that the goal is to maximize £Lonv in a sustained
fashion (quasi-steady state maintained), as this is equivalent to
maximizing decomposition rate.
Convective removal can be subdivided into tv/o parts:
(iii)
* ^dac
where: 0^ - heat removal through vaporization
* We~~tEarnr~D~rT Peter ~Fi
developed in Sections 7 and 8.
96
-------
E removal through dry air convection.
At composting temperatures the dominant mechanism of heat removal is
vaporizatio.;, with 0 ^ 9 Ud (2). Removal of heat through vaporiza-
tion thus can be expressed at?
QVH 0.9 m
Note that the goal of maximizing flconv pertains equally to
The overall rate o£ wp.ter removal is expressed as:
I s » ("out • "in3
where:v B mass vapor flow (mass moisture/time)
to - humidity ratio (mass moisture/mass dry air).
With inlet (ambient) conditions of 20°C and 50% relative humidity,
approximately 96% of the vaporization is driven by heat generation,
and the remainder by inlet air unsaturation (Appendix A-5). Thus we
may focus on expression (iv).
The only part of expression (iv) corresponding to a manipulable
physical analogue is the coefficient m, which corresponds to ventila-
tion rate. Thus, m represents a means of matching heat removal to
heat generation in reference to a constant, activity-promoting,
temperature.
Now consider in theory two ways of controlling m which predict
markedly different behavior patterns (TABLE 8). In the first (see
"Prediction" columns), designated R, m is varied such that the pile's
outlet temperature ascends to 60°C and" this value is subsequently
maintained as the operational ceiling. The need for a variable m
st(5ins from the variable rate of heat generation as caused by popula-
tion shifts, nutrient depletion, available water depletion, and other
unidirectional changes characteristic of batch culture. A period of
vigorous heat generation ensues, yielding a large 0 . Consequently
the material dries rapidly, ultimately terminating activity for lack
of microbially available water (Section 4). The pile is well-oxy-
genated, as approximately 9-fold more air is needed to remove heat than
to supply 02 for respiration (Appendix D-l).
The -econd predictive theoretical control approach, designated B-,
does not involve deliberate temperature management. Rather, m is con-
trolled such that interstitial 02 is not less than 5% (v/v) . This is
easily accomplished through minimal ventilation on a fixed schedule
as it leads, by default, to inhibitively high temperature, suppressing
02 consumption. Although water removal per unit air is comparatively
high (large h t), owing to a small m overall removal (0 ) is slight.
Thus, approacR B leads to inhibitively high temperature^nd a prolonged
period of low-level activity. Moreover, attempting to increase 0
by increasing m is not successful. Since heat output, is slight, the
increased ventilation cools the pile, decreasing h
~
97
-------
If.'
TABLE 8. BEHAVIOR OF THE COMPOSTING ECOSYSTEM AS INDUCED BY DIFFERENT APPROACHES TO THE
CONTROL OF m (VENTILATION MANAGEMENT). PREDICTION BASED ON EXPRESSION iv
[fly - 0.9 m (hQut - hin)l, AS CONSTRAINED BY THE INTERACTION BETWEEN HEAT GENERA-
TION AND TEMPERATURE, 'BSERVATION BASFD ON SIDE-BY-SIDE 36 TONNE PILES (PILES
9A AND 9B).
m (m air delivered/wet metric ton)
Total
Peak (hr'1) 1
Mean (hr'1) J
Temperature (°C)
Outlet air
Pile: median (range)
Pile: lobs > 60°C (number of obs)
Heat
Qv
h . (kj /kg dry air) #
— o y £
hin (kj/kg dry air)$
Decomposition
Sludge volatile solids decrease (1)
Water removal
Moisture content decrease (5)
Mass H20 removed (% of initial)
kg H?0/kg dry air
Prediction
Anoroach R Aporoach B'
• Large Small
Peak>mean Peak=mean
60 (by design) SO (by default)1
Large Small
477 1560
74.9 74.9
Large Smal^
Large Small
Large Small
0.1404 0.5394
Observation
Rutgers ^ Seltsville
strategy r>rocessi"
9660 2350
135 6.06*
31.5 4.69*
53(24-68} 70(45-82)
13(1755) 91(1519)
72.3 15.4
67 -»• 29 65 -> 61
8C.9 19.4
kg H20/ra^ ambient air
(continued)
0.05S6
0.0793
-------
TABLE 8. (continued)
Processing period (days)
Respiration
Rate 0- uptake
0,, level (§ v/v)
CG^ level (% v/v)
Prediction
*
Approach R
Short
High
High
Low
Approach B
Long
Low
High
Low
Observation
Rutgers Beltsi'ille
Strategy Process
15. S 21
.161 .16%
1,4 % i/45
Time-variable, interactive, blower operation (temperature feedback control) in reference
to an operational ceiling of 60 C.
"^Fixed schedule blower operation.
'^Difference reflects manual adjustment of schedule (9).
sThe ecosystem brings itself to the edge of its temperature tolerance limit.
Relative humidity (RH) assumed to be 1005.
^Temperature and RH assumed to be 20°C and 50%.
S
-------
The physical analogues of R and B are the Rutgers and Beltsvi1le
approaches, as already compared in detail in previous Sections. In
TABLE 8, the comparison is developed in terms of equation (iv) . Note
that where existing data permit evaluation, prediction i? con firmer7
by observation.
Additional confirmation comes from responses to sudden chanees
in ventilation. In the Rutgers strategy heai generation is intense,
hence accidental loss of blower function is predicted to induce a
tainperatm a upsurge. This response was obser/ed numerous times. In
the Beltsville process heat generation is weak, he:ice a sudden -'ncrease
in ventilation should induce a temperature downturn. Moreover, increased
ventilation is not expected to substantially enhance drying, as the de-
creased h + tends t-o offset the increased m, moderating any change in
—- ~*"
0 . This behavior was obseived in an unsuccessful attempt to improve
drying in field-scale Beltsville pilt~ by increasing ventilation 4-fold
partway through the process cycle (10) > as described in Appendix D-2.
In the Rutgers strategy the system is not permitted to self-limit
via inhiliitively high temperature, but rather is prompted to do so via
substrate and water depletion. Our experience is that demand for ven-
tilation terminated on day 7.1-IS.8 with a mixture of primary sewage
sludge and wnodchips, and on day 4.3-9.5 with a mixture of the sludge
and recycled compost. While the immediate cause of termination was
dryness, the addition of water provoked only a weak revival of demand.
Thus, the essence of composting process control is given by the
expression Qr = 0.9 m (h - Jj-n)j as constrained by the interaction
between heat generation and temperature. Mathematically, in isolation
from this constraint, it might seem thai; a high value of Qy is obtain-
able through a high value of n, or h,,, . , or both. This is unrealistic
-~ — out
however, for two reasons. First, ordinary values of _h. dictate that
an arbitrarily high m leads to a low h t. (see Secticn 8 - next). Second,
values Oj,' h t representing temperatures higher than 60°C inhibit heat
generation. Consequently, the role of m is defined as that of matching
heat ~emoval to heat generation in reference to temperature, such that
maximal sustainable values of h t and 0 are realized.
This mathematical development constitutes a formal rationalization
for on-demand ventilation via temperature feedback control. This is
the basis of the Rutgers strategy for composting process design and
control.
100
-------
SECTION 8
DRYING ASSOCIATED WITH COMPOSTING, AND NON-BIOLOGICAL AIR DRYING
INTRODUCTION
The analysis of Section 7 is extended here'-in to include the'be-
havior of a biologically inert, ventilated, pile o£ organic material
("air drying"). This leads to a reinterpretation of the Rutgers and
the Beltsville approaches, in a broader context.
THEORETICAL TREATMENT
Simplifying Assumptions
Consider the air drying process as a closed system comprised
of a pile (not mechanically'agitated) of biologically inert material
and the air which is passed through the pile. Neglecting gains and
losses of heat through the mechanisms of radiation and conduction,
and fractional gains resulting from the passage of air through the
matrix, the system (pile and air) experiences no change in enthalpy
(- isenthalpic). However, if the inlet air is unsaturated, a redis-
tribution of enthalpy occurs within the system through the vaporiza-
tion of water from the matrix into the flowing air. This results
in a decrease in the enthalpy of the pile exactly balanced Ly an
increase in the enthalpy of the air.
Similarly, consider a static composting pile managed according
to the Rutgers strategy and the air passed through it as a closed
system, accepting the same simplification with respect to radiant,
conductive, and frictional factors. Unlike air drying, the compost-
ing system is not isenthalpic; rather, the system's enthalpy increases.
This is because chemical bonds are broken through biological action,
releasing heat. In common with air drying any unsaturation of the
inlet air causes a redistribution of enthalpy. This contributes
negligibly to *he enthalpy increase of the air, however, compared to
biological heat generation.
Two further simplifications are made for the purposes of this
exercise. First, the redistribution of enthalpy originating in the
unsaturation of the inlet air is neglected. That is, in the air
drying examples the air is treated as if it were isenthalpic. Rela-
tive to actual conditions this overstates the temperature decrease
experienced by the air, and understates the water removal per unit
mass of air. It does not affect the composting examples with respect
to these factors. Rather, the effect is a small overstatement of
the mass of air needed to accomplish a given -mount of cooling. These
simplifications are adopted to permit the use of standard psychometric
data (46).
101
-------
Next, outlet air RH values must- be assigned. For air drying
this is not a problem as a necessary condition (short of "break-
through") is an outlet RH of 1001. For composting an outlet RH
of 100% is adopted, in1 the belief that this is a valid approximation.
Neglect of the enthalpy redistribution originating in unsaturated
inlet air introduces a bias favoring composting relative to air dry-
ing, with respect to the removal of water per unit mass of air. The
stipulation of saturated composting outlet air might introduce a
further bias in this direction. We believe that the bias(es) are
minor in the context of the exercise, and that the hypothetical
examples provide useful approximations of the difference in perfor-
mance to be expected of air drying and composting, with respect to
water removal. Regardless, independent verification based on field
data is offered later in this Section.
As was already seen in Section 7, regardless of whether the
driving force is heat generation, unsaturation of the inlet air, or
both, vaporization can be described as follows:
where: v a mass vapor flow (mass moisture/time)
m s dry air mass flow (mass dry air/time)
uj - humidity ratio (mass moisture/mass dry air)
The outcome of this relationship is exemplified in the non-biological
air drying system using various inlet conditions (TABLE 9) .
If the inlet air is saturated (e.g. 7.2°C- 100% RH) no change
in temperature and RH occurs with passage through the matrix, and
no moisture is removed (Aui s 0). If the inlet air RH is less than
1001, water is vaporized from the aqueous matrix into the flowing
gaseous phase until an RH of 100% is reached. The vaporization
from the matrix causes it to cool, and this is translated into a
cooling of the air as its temperature equilibrates with that of the
matrix. Thus, in the air drying process the magnitude of moisture
removal is solely dependent on inlet air conditions, and the only
factor subject to process control is m. This may be increased to
compensate for poor drying air. For example, roughly equal vapori-
zation rates are obtainable with air at 32°C-20$ RH or 32°C-90% RH,
by using a 10- fold greater m for the more humid air.
Heat drying (see last entry in TABLE 9) is a special case of
air drying, in which the ambient air is preheated in an operation
external to the pile. A gas-fired hot air generator, for example,
might be used to condition the inlet air. Coropared to the ambient
air, this results in a larger Am.
102
-------
TABLE 9. HYPOTHETICAL CHANGES IN AIR AS EFFECTED BY THE AIR DRYING PROCESS (BASED ON
ISENTHALPIC CONDITIONS - SEE TEXT)"
Inlet air
Process
Air drying
Air drying
Air drying
Air drying
Air drying
Heat
dryingt
T
4.4
7.2
7.2
32
32
60
RH
111
60
100
20
20
90
10
ID
(leg H20/
kg dry air)
.0031
.0064
.0012
.006
.028
.013
Inlet and
outlet air Outlet air
h
(kJ/kg
dry air)
30.2
41.2
28.8
65. 1
121
111
1.7
7.2
0
17.3
30,8
29,4
RH
HI
100
100
100
100
100
100
(kg H20/
kg dry air)
.0042
.0064
.0038
.0123
.0286
.026
u/m = Ato'
Tkg H20/
kg dry air)
.0011
0
.0026
.0063
.0006
.0124
ft
Values of h and w derived from Reference 46.
t The inlet conditions are representative of crpp drying applications (47)
is a special case of air drying (See text).
Heat drying
-------
Whereas in air drying (and its variant heat drying) the unsatura-
tion of the inlet air is the only force driving vaporization, in
composting this factor is minor. Here the major factor is the meta-
bolic generation of heat in the matrix, as follows.
Metabolic activity in the aqueous matrix phase generates heat,
establishing an enthalpy differential between the solid-liquid matrix
and the gaseous phases (see Section 3). The differential is sustained
by the flow, which brings in cooler, drier, air. The differential
drives vaporization, and establishes a positive temperature gradient
along the axis of airflow. Thus, as long as heat generation persists,
the enthalpy of the air increases with passage through the pile. The
upper limit to this condition is defined by the heat generation-
temperature interaction. The major components of the enthalpy increase
are i) an increase in the air temperature, and ii) an increase in the
water content per unit mass of air.
Therefore, in composting the rate of vaporization is a function
of heat flow. As was derived in Section 7S this leads to the approxi-
mation:
Qv " °-9 » (iout-hin) " °-9 £ A^ (ii)
where: 0 =» heat flow associated with vaporization (energy/
time)
-out " outlet a^r -nthalpy (energy/mass dry air)
= inlet air enthalpy (energy/mass dry air).
An ef^'jct: of inlet conditions on waiter removal per unit air during
composting is implicit in expression ii, in that ambient temperature
and RH determine the value of h- . (But, variations in h. are
automatically compensated for Tn the Rutgers strategy through tempera-
ture-feedback adjustment of m. ) The effect of inlet conditions is
small, however, as ceen in TABLE 10= For example, only 1.2 times as
much moisture is removed per unit mass of air by the "best ambient
drying air" (7.2°C - 20% RH) than the "worst air" (32°C - 90% RH) .
This illustrates that, given informed process control, composting
process performance is not sensitive to ambient conditions. (Extreme
cold is a separate potential problem - see later.)
Thus, the amount of moisture removed per unit mass of air is much
greater through composting than air dry ig. In this hypothetical exer
cise the amount of water removed through composting is larger by the
following factors: 135x,<«>, S8x, 23x, and 207x (compare the Au's in
TABLES 9 and 10). The factor of infinity is obtained when the inlet
air is saturated. Compared to the heat drying example, the composting
Aw's are lOx to 12x greater.
Continuum Among Air Drying, Rutgers Strategy, and Beltsville Process
It is now possible to define the differences among air drying,
the Rutgers composting process control strategy, and the Beltsville
104
-------
TABLE 10. HYPOTHETICAL CHANGES IN AIR AS EFFECTED BY THE COMPOSTING PROCESS
Inlet air
T
RH
h
0 (kJ/kg
(
4.
7.
7.
32
32
Q.
4
2
2
r ••"
60
100
20
20
90
dry
30
41
2S
65
121
air)
.2
.2
.8
.1
(kg H20/
kg dry air)
.0031
.0064 •
.0012
.006
.028
Tt
1 '
( C)
60
60
60
60
60
Outlet air
RH
111
100
100
100
100
100
h
(kJ?kg
dry air)
477
477
477
477
477
to
(kg H20/
kg dry air)
.152
.152
.152
.152
.152
v/m=Au
Tkg H20/
kg dry
.149
.146
.150
.146
.124
air)
Values of h and w derived from Reference 46, and where necessary by calculation.
Based on the outlet temperature characteristic of the Rutgers strategy.
-------
composting process in terms of ventilation. This is done in reference
to expression ii. The hypothetical exercise involves a static pile of
suitably porous, energy-rich, organic material, different ambient air
conditions, and manipulation of m. The constant in expression ii is
temperature-dependent and would Have to be changed as appropriate,
were numerical solutions sought. This is not a source of error in the
present, qualitative exercise, however, as its outcome is not affected
by the value of the constant.
Cases 1 and 2 involve a very high, arbitrary, value of m, such
that the material is suspended in the airstream ("fluid bed"J. In
Case 1 the ambient air (inlet air) temperature is 1°C and its RH is
100%. Because the temperature is biologically unfavorable, heat
generation is very slight. Because any heat that is generated is
quickly removed, the temperature does not. increase to a more favorable
level. Consequently, for all practical purposes 0. ° 0, and the
system's behavior is described by expression i. However v also is
zero, as this is the special case of saturated inlet air. Were the
RH<100$, evaporative cooling would ensue, v would be grea than
zero, and 0 would be zero.
Case 2 also involves a very high m, but differs in that the
ambient air is 30 C-100% RH. Since the temperature favors biological
activity heat generation ensues, but the heat is promptly removed and
no appreciable temperature elevation results. Nonetheless, the heat
generated drives vaporization, h t>n. » and the system's behavior is
described by expression ii. Although Cases 1 and 2 are o£ theoretical
interest, they would represent extravagant use of energy for ventila-
tion.
Cases 3 and 4 involve composting in the ordinary sense, in that
a temperature elevation is experienced. Expression ii pertains in
both eases.
Case 3 is that of the Rutgers control strategy, in which the
value of m is continuously adjusted through fee.dback control to seek
an outlet temperature of 60°C. Either set of inlet conditions (1°C
-1001 RH, or 30°C -1001 RH) initiates self-heating, although a slow
start is experienced with the colder air. Onr.e underway, a quasi
steady-state is established in which a sustained high rate of decom-
position and vaporization is realized.
Case 4 is that of the Beltsville Process, in which m is set at
some low fixed value consistent with the maintenance of a minimal
residual 02 level. This leads to the highest value of h t support-
able by the system (but note the low m), and a commensurate inhibi-
tively high temperature (^80°C). Consequently, a quasi steady-state
is established characterized by low rates of decomposition and vapori-
zation. This condition is signified by a low £) .
106
-------
FIELD EXPERIMENTATION
Ml!J:££iiLLl_Ji£J:LJ!2^
The experimental study concerned a single pile formed of material
previously composted by the Rutgers method. This was a mixture of
the screened (woodchip-free) material from piles 8 and 9A (Section 3)
which had not been used in the 11 series piles (Section S), and all
of the material from piles 11B and 11C. Prior to its use in the
present study the material was stored in the open for 3i to 8j months.
During this period much of it became moist, presumably resulting in
further stabilization through informal curing.
This material was formed into a pile of approximately 2.2 tonnes,
designated as pile 12, It had a starting moisture content of 65$.
The pile was fitted with one blower, thermocouples, and a gas sampling
probe (Figure 64). Neither a thermistor nor a timer was required, as
the experimental plan called for continuous operation of the blower
(100% time on). The forced pressure mode of ventilation v/as employed.
Samples for the determination of moisture content were taken regularly
from the part of the pile slightly below that represented by position S,
On some occasions the lower part of the pile represented by position 1
v/as also sampled.
Time-zero was 27 February 1981 and termination was on 18 March
1981. During this period the ambient temperatures were (°C): mean
of the daily highs, 9°; mean o£ the daily lows, zero0; mean of the
daily means, 4.5°; range of the daily highs and lows, -7.8° to 13.9 .
Rainfall amounted to 7 cm in four occurrences.
Results
The temperatures of the ambient air and pile position 1 are
plotted in the lov/er graph in Figure 65, and those of positions 1 and •
5 are plotted in the upper graph in this figure. (The other tempera-
ture data are given in Appendix E.) To help bring out the trends the
•plotted data are also tabulated in the form of the mean differential
values for each 24 hour interval (TABLE 11),
Position 1 tended to be cooler than the ambient air during the
first 240 hours, notwithstanding invervals to the contrary (e.g. hr
130 to 160) and intervals of identical temperatures (e.g. hr 180 to
200). Starting at hr 280 position 1 tended to be warmer than the
ambient air. Positions 1 and S v/ere at similar temperatures during
the first 190 hours, and thereafter position S was cooler. Position 5
was generally cooler than the ambient air.
The 0, and C02 levels in the pile v/ere not distinguishable from
the ambient values.
The moisture content data are given in Figure 66. At the upper
sampling level, the moisture content did not change during the experi-
mental period. The lower level experienced a moisture content decrease
subsequent to hr 140. At hr 467 an unusually low value (11.2%) v/as
noted.
107
-------
o
oo
Figure 64. Pile 12, Cross-sectional representation: textured area, woodchip cover
and base; clear area, compost; circle, perforated flexhose. The numbers indicate
monitoring positions: thermocouples, positions 1 through 6; gas sampling probe,
adjacent to position 5. The blower was operated continuously (100% time on) in
the forced-pressure mode.
-------
15
t 1 1 1 I II 1 ! J I ) \ I I ! 1 t i
LJLjLJLJL-jl—JLJLJLJLJ-JLJ-JUJrf.»A^L^i«JU -,
-5
HOURS
Figure 65.
and 5.
Pile 12, temperature external to the pile (ambient) and at positions 1
-------
TABLE 11. TEMPERATURE DIFFERENTIALS AMONG AMBIENT, POSITION 1,
AND POSITION S PROBES*
Interval
(hr)
0-23
24-47
48-71
72-95
96-119
120-143
144=167
168-191
192-215
216 239
2^0-263
264-287
288-311
312-335
336-359
360-383
384-407
408-431
432-455
Position 1
-ambient
™l!£L_
-3
-1.8
-1.8
-I.I
-0.6
-0.2
1.2
-0.5
-0.5
-0.4
-0.5
-0.1
1.3
-0.4
0.3
1.6
2.3
3.0
3.0
Position S
-position 1
(°C)
0.6
0.2
0.7
0.7
0.2
-0.1
0.1
0
-0.7
-0.6
-1.1
-0.2
-0.7
1.1
-2.4
-3.4
-5.3
-1.4
-2.8
Position 5
-ambient
___ffiL_
-2.4
-1.6
-1.1
-0.4
-0.4
-0.3
1.3
-0.5
-1.2
-1.0
-1.6
-0.3
0.6
-1.5
-.2.1
-1.8
-3.0
-1.6
0.2
ft
The temperature was recorded every hour. The differential values
are based on the mean of the twenty-four hour interval.
110
-------
701
60
50
52
o
u
o
e:
u
a
30
20
7© CM HIGH LEVEL
40 CB3 HIGH LEVEL
-1
100
200 300
TIME IN HOURS
400
500
Figure 66. Pile 12, moisture content, samples taken from central interior
locations.
-------
At termination the pile was bisected for visual inspection
(Figure 67). A hemispherical dry zone, with its -origin at the
ventilation duct, comprised approximately half tha cross-sectional
area. This terminated in a 2 cm wide transition zone, evidencing
a moisture gradient. The upper half of the cross-sectional area
was frankly moist. With respect to other characteristics (granula-
tion, odor, color), the material at termination did not differ
noticeably from the starting material.
DISCUSSION
The subject of the experiment was a ventilated pile of £omp_os_t_1
as distinct from the usual ££JS££s^ijig pile. The purpose was to
investigate air drying withautTEe~complication of biological heat
generation. To approach this condition the pile was formed of material
previously composted by the Rutgers method, and subsequently cured
informally for a prolonged period. As such, the material was depleted
o£ readily metabolizable substrate, and potentially supportive of
only slight raicrobial activity. The timing of the trial was fortui-
tous in that the ambient air temperature was unfavorably low for
microbial activity. Furthermore , the strong ventilation imposed
(though not comparable to hypothetical cases 1 and 2) prevented heat
storage, which might otherwise have elevated the temperature to a
biologically more favorable level. Judging from these conditions and
from the pile's behavior as discussed below, heat generation was
negligible as intended.
In the virtual absence of heat generation the system is expected
to behave as follows. A well-defined cooling-drying front develops
at the air inlet point and expands radially along the axis of airflow.
Below the front the moisture content of the material comes into equi-
librium with the RH of the ambient air; above it the moisture content
is unchanged from time- zero. Below the front the temperature of the
material and interstitial air is at the ambient level (except for
frictional heat input - see later); above it the temperature is less
than ambient.
This behavior is predicted because the drying and cooling both
result from vaporization — driven solely by unsaturation of the inlet
air. In this circumstance the unsaturated condition is abruptly ter-
minated through the action of two complimentary forces at the drying-
cooling front: 1) the transfer of moisture into the gaseous phase,
saturating it, and 2) the cooling of the gaseous phase as it comes
into temperature equilibrium with the matrix, decreasing its capacity
to hold moisture at saturation. In combination, these forces bring
vaporization and cooling to a sharp halt5 hence a narrow, well-defined
front. This cooling-drying front migrates through the pile with time,
leaving in its wake an expanding region of air-dried material. The
material ahead of the front is undried.
112
-------
by P. C. Millr
3t t«.in.tion. Note
and drV lo«er area. Photo
113
-------
These predictions were verified in the experimental observations.
Cooling is evident in the temperature differentials between: position
1 and ambient; position S a,:d 1; positions 5 and ambient. The pattern
of the differentials is suggestive of a cooling front passing posi-
tion 1 roughly midway through the experimental time period, but not
reaching position 5 by termination. This is inferred from the sign
change, from negative to positive, in the differential between posi-
tion 1 and the ambient temperature, and the opposite change in the
differential between positions S and 1. Thus the temperature de-
creased and the decrease was discontinuous, as expected of a system
in which biological heat generation is negligible.
In. stipulating isenthalpie air in the hypothetical air drying
example, it was noted that this exagf ~ates the temperature decrease
(see Simplifying Assumptions) . This '!._, because the exercise is based
on stan3ar3~psycHomet,ric 3at¥, which do not take into account the inter-
action between the gaseous phase and the liquid-solid phase (the air
is cooled by the matrix). Additionally there is a second source of
error in the same direction, in the form of frictional heat input,
resulting from the forcing of the air through the niitrix. In the
virtual absence of biological heat generation, frictional heat could
be significant. Neither of these errors can be quantified at present.
It is possible to develop an exercise, not without its own diffi-
culties, comparing the theoretical isenthalpic cooling and the observed
cooling. The exercise is as follows.
During the experimental period the mean ambient temperature (mean
of tae daily means) was 4.5°C. Relative humidity was not measured,
and the ambient value no doubt fluctuated widely. Nevertheless, if
an RH of 60% is taken as a, representative ambient value, and if inlet
conditions were consta\ ^ at 4.5°C - 60% RH, an isenthalpic temperature
decrease of 2.S°C is predicted. The data offer three opportunities
for comparison to the hypothetical value, as follows: position 1 minus
ambient, from time- zero to hr 215; position S minus position 1, from
hr 216 to termination; position S minus ambient, for the entire period.
The observed mean differentials are, respectively, (°C) : -0.8, -1.9
and -1.0. Thus observation conforms to theory in that the means of
the data are negative, and the absolute values are less than 2,5.
at hr 280, above- ambient temperatures were noted at
positio; 1 (TABLE 11). This is attributed to frictional heat input.
Thus, aespite the complications inherent to this exercise, the
temperature data are suggestive of the migration of a cooling front
through the pile.
The moisture content data provide equally suggestive data for the
drying component of a cooling-drying front ("igure 66).
Perhaps the most compelling evidence, however, is in the form
of the visual appearance at termination (Figure 67). This revealed
unambiguously that a distinct, radial, drying front had indeed passed
position 1 but had not reached position 5. It is concluded that the
observations are as predicted by the model of the non-biological air
drying process (expression i).
114
-------
Water
Water removal per se is a goal in sewage sludge treatment. It
is therefore of interest to compare composting and air drying from
this narrow perspective. This comparison neglects the other treatment
goals advanced through composting.
Consider the two approaches to composting. In the Rutgers stra-
tegy, one-third less water was removed per unit air delivery. This
reflects the characteristically lower value of hQut- However, owing
to the high proportion of water removed and the speed of its removal,
only the Rutgers strategy affords a useful means of drying sludge.
(Also, see Appendix D-2).
Compared to non-biological air drying, the biological system
(Rutgers Strategy) removed 22. 3x more water per unit air delivered,
This reflects the dominant role in composting of heat as the force
driving vaporization. Unsaturation of the inlet air plays only a
minor role. In air drying, however, inlet air unsaturation is the
only operative mechanism. In addition to its efficiency in terms of
air usage, composting removes a higher proportion of the water, and the
removal is faster. This .comparison is expressed in terms of the cost
of pumping air (TABLE 12.)
TABLE 12. COST OF WATER REMOVAL THROUGH BIOLOGICAL AND NON-BIOLOGICAL
MEANS
Process
Composting-
Rutgers d
strategy
Composting-
Beltsville
processt
Non-biological
air drying*
Water
removed/
unit air
delivered
(ton x
10-6/ft3
1.85
2.48
0.0828
Process-
ing time
-iiazsJL.
9.8
20.8
18.8
Propor-
tion of
•. initial
water re-
moved
_£!!_
75.7
19.4
46.7
Cost o£ air
delivery/ur.
it
water removed
C$/ton)5
0.31
0.21
6.33
Based on piles 7, 8, 9A, 11A, 11BS 11C.
"("Based on pile 9B,
±Based on pile 12; assumes that half the material was air dried at
termination, and half unchanged from initial moisture content (see
Figures 66 and 67).
§Based on electricity @ $0.06/kw-hr.
115
-------
SECTION 9
COMPARISON BETWEEN THE RUTGERS AND BELTSVILLE
CONTROL STRATEGIES: EFFECT ON CURING STAGE
INTRODUCTION
The greatest benefit from formal process control is realized
early in the composting, when readily metabolizable organic sub-
strates are most abundant and the potential for decomposition is
greatest. As readily available substrate is depleted it becomes
feasible to continue the processing, if necessary to do so, on a
less formal basis. The earlier period may be called the high rate
stage, and the later period the curing stage.
The present section concerns the curing of materials previously
composted by the Rutgers and Beltsville methods.
MATERIALS AND METHODS SPECIFIC TO SECTION 9
Material from piles 9A and 9B (Section 3) was screened, to
remove woodehips, with a Royer Model 365 shredder/mixer, coupled
with a Mogensen sizer (Royer Foundry and Machine Co., Kingston, PA).
Screened-material was transported from the primary composting site
at Camden, N.J, to New Brunswick, N.J., to form two separate curing
piles. Each pile consisted of approximately 3m^ of material in a
conical shape. A stainless steel dial thermometer of 1 meter length
w°" inserted into the ^center.of each pile, where it remained through-
out the trial. The experimental chronology and related matters are
given in TABLES 13 and 14.
On two occasions the piles were remoistened with water from a
garden hose. As part of this operation the material v/as turned and
mixed by shovel.
Samples v/ere removed periodically from_the pile interior and
tested qualitatively for NHt, NO;* and N03- The reagents were
as described in Standard Methods under items 132B and 134 (48).
Devardas alloy was used as a reductant in the test for NO^.
Samples from the pile interior v/ere subjected to an odor test
on the day of sampling. The material was placed into 1 pint
(ca.O.SJl) screw-cap jars, such that they were one-third full. The
jars were coded to conceal sample identity, and randomly selected
individuals (excluding project personnel) were asked to evaluate the
samples on a sc. ' ,• of -5 to +5 (-5 « most unpleasant; 0 - neutral;
+ 5 = most pleasant). Tv/enty people evaluated the first set of
samples, and 30 people the second set.
116
-------
TABLE 13. PERTINENT DATES, ELAPSED TIME, AND WEATHER
Time zero (parent pile formation)
Rutgers: 9 Jul 1980
Beltsville: 11 Jul 1980
Parent pile screening*
Rutgers: day 36
Beltsville: day 26
Curing pile formationt
Rutgers: day 40
Beltsville: day 26
Termination (10 Nov 1980)
Rutgers: day 124
Beltsville: day 122 .
Weather (1 Aug to 10 Nov)*
Ajnbient air temperature (°C)
High: 25
Low: 14
Mean: 20
Range: -S to 37
Rainfall
Amount (era): 21.6
Occurrences (no.): IS
*At Camden, N.J.
^At. New Brunswick, N.J.
This is taken to represent the conditions following the start of the
curing stage. See Table 2 for conditions during the high rate stage.
I
High = mean of the daily highs; low - mean of the daily lows; mean =•
mean of the daily means; range ™ overall range of the daily highs and
lows.
117
-------
TABLE 14. MANIPULATIONS OF THE CURING PILES, AND CHARACTER OF
THE MATERIAL
Moisture content (!) and water addition
Rutgers parent pile: day 0 to 19, see Figure 36; day 24, 251
Rutgers curing pile: day 76, water added; day 84, water added;
day 90, 631; day 107, 64*
Deltsville parent pile: day 0 to 21, see Figure 36; day 22, 65$
Beltsville curing pile: day 74, water added; day 8"2, water added;
day 88, 584; day 105, 53%
Odor test (mean score)
Rutgers curing pile: day 86, • 1.30; day 91, + 0.66
Beltsville curing pile: day 84, - 3.15; day 89, - 2.29
First detection of NO,
Rutgers curing pile: day 86
Beltsville curing pile: day 113
First detection of NO:
Rutgers curing pila: day 99
Beltsville curing pile: day 123
*
The piles were turned as part of the water addition operation.
118
-------
RESULTS
Because of its physical accessibility the parent Beltsville
pile (9B) was screened first, permitting the earlier formation of
its derivative curing pile (TABLE 13). This curing pile was moist
from the outset (see Figure 68 and TABLE 14). A subsequent period
of rainfall delayed the screening of the Rutgers pile (9A) , but
the material eventually obtained was from the interior of the pile
(unaffected by the rain). Hence this derivative curing pile was
dry at the outset (Figure 36 and TABLE 14) . In addition to the
moisture content data, the dry condition of this curing pile over
much of the trial was indicated by visual and tactile examination
of the samples obtained periodically for the nitrogen spot tests.
On day 76 the material was wetted by the addition o£ water and
turning,
The Rutgers curing pile cooled more quickly and in a more
regular pattern than its Beltsville counterpart (Figure 68). In
the Rutgers pile a slight, temporary, temperature descent coincided
v/ith the first water addition-turning operation (day 76), but not
the second such operation. In the Beltsville pile sharp descents
and re-ascents coincided with both addition-turning operations.
Based on the average test panel score, at the time of the
first odor test the material from the Rutgers curing pile was less
unpleasant than the Beltsville material (TABLE 14). At the time of
the second odor test the freshly sampled material from the Rutgers
pile was rated at the low end of the "pleasant" range. Although
the material from the Beltsville pile was improved v/ith respect
to odor, it still v/as considered distinctly "unpleasant."
The end products of both steps of nitrification (NOZ and
NOj) appeared earlier in the Rutgers pile (TABLE 14).
DISCUSSION
In the parent pile 9A there was intensive heat generation
(decomposition) and vaporization, hence the derivative curing pile
was formed of moderately well-stabilized, dry, material (TABLE 14).
Because of the dryness, activity was probably slight from approxi-
mately day IS to day 76 (when the material was wetted). Thus,
dryness presumably delayed curing by as much as two months. In
routine operation timely water addition would be indicated to prevent
a curing hiatus.
In contrast, parent pile 9B experienced inhibitively high
temperatures, with correspondingly slight decomposition and vaporiza-
tion. Its derivative curing pile was formed of poorly stabilized,
moist, material. Although direct observations are lacking, we
suspect that activity in this pile v/as O^-limited for a part of the
curing period.
119
-------
ts)
O
80
60
a:
£40
2
bJ
20 I
PILE 9A-Q
PILE 9B-0
-4
1
40
60
80 100
TIME IN DAYS
120
140
Figure 68. Curing piles, temperature in the center of the pile,
-------
The circumstances at the start of the curing period may be
summarized as follows. The material derived from pile 9A was ready
for curing, as it had reached a moderately well-stabilized condition.
Curing was delayed, however, by dry-ness. The material derived from
pile 9B was not ready for curing, as prior stabilization was slight.
Despite the dryness-induced curing hiatus, 9A material was first
to reach a well-cured condition judging by the cooling rate, odor,
and onset of nitrification. (Although cooling is generally a sign
of substrate depletion, in the present case the issue is clouded by
the effect of the dryness on biological heat generation.) Odor is of
obvious practical operational interest, as well as being indicative
of the degree of stabilization. This odor test result is consistent
with other formal observations on odor (2, 29).
The validity of nitrification as a sign of organic matter
stability is widely appreciated in the sewage treatment field, among
other fields (49). This stems from several characteristics of the
responsible bacteria, such as their chemoautotrophic nature, sensi-
tivity to elevated temperatures, and slow grov/th.
In routine practice the addition of water could be made part
of the transferral operation from the site of the high rate operation
to the curing site. Alternatively, water might be added during the
high rate stage, as needed to sustain microbial action. In this
manner the high rate stage would pass to the curing stage without
moving the material. This mode of operation can be visualized by
reference to Figure 42.
In designing a water addition program, it should be recognized
that the material becomes progressively more difficult to wet as it
dries. As a. rough approximation, it might be advisable to add water
as the moisture content decreases to perhaps 40%.
In employing composting as a waste treatment technology it is
generally desireable, in the initial processing, to strive for a
maximal decomposition rate. Whether it is necessary to subsequently
cure the material is a site-specific matter. This depends on the
nature of the waste and the intended avenue of ultimate disposal/.
resource-recovery.
121
-------
SECTION 10
RUTGERS CONTROL STRATEGY: DIAGNOSIS OF PROCESSING FAILURE
INTRODUCTION .
The original intent of this trial was to examine the effect
of wetness on composting by the Rutgers strategy, by isolating
starting moisture content as the variable. One batch o£ sludge cake
and woodchips was mixed without addition of water, to serve as the
control, while water was added to two batches. During composting,
however, all three piles behaved similarly, exhibiting inhibitively
high temperatures, low rates of activity and decomposition, and slight
drying. These patterns indicated processing failure by the standards
of the Rutgers strategy.
The suspected immediate cause of failure was inadequate pene-
tration of air into the composting mass, leading to inadequate heat
removal. Both a low amount of bulking agent in the mix, and an overly
large woodchip ventilation base contributed to poor air penetration.
This outcome, though, lead to an unplanned interpretative opportunity
to contrast the behavior of a failed pile with that of a successful
one from a different trial.
MATERIALS AND METHODS SPECIFIC TO SECTION 10
Each pile consisted of approximately 6 tonnes of the sludge-
v/oodchip mixture (excluding added water). Pile 6A v/as formed using
unammended sludge cake and woodchips (no water added). For pile 6B
tap water was added from a garden hose at a "moderate rate" as the
sludge and woodchips were mixed in the pug mill, and for pile 6C the
rate of addition was "fast." The ratio of.sludge to woodchips was the
same in all piles (1 tonne sludge to 1.35m woodchips). The added
water is not taken into account in this ratio. The controller set
point for all of the piles was 45 C.
The trial period was 13 March 1980 to 3 April 1980. Ambient
air temperatures during this period were (°C): mean of the daily
highs, 13; mean of the daily lows, 2; mean of the daily means, 11;
range of the daily highs and lows, -3 to 20. Precipitation amounted
to 14.4 cm in 1-2 occurrences.
In a special terminal test (3 April) of the penetration of
air into the sludge-woodchip mixture of pile 6B, hot air was substituted
for ambient air by use of a kerosene-fired catalytic space heater rated
at 24 x 106gm cal/hr (95,000 btu/hr). The heated air was introduced
to the inlet of the blower, which resulted in air at 130°C being intro-
duced to the flexhose. Prior to the start of this test five additional
thermocouples were positioned at 0.3m intervals in the woodchip base
122
-------
midway between the top and bottom of the base, in a line perpendicular
to the flexhose.
RESULTS
All three piles failed. One of these (6B) is compared to a
previously described successful pile of comparable size (pile 7 - see
Section 3). Certain details not included herein are recorded in Sec-
tion 3, Appendices A-l and A- 2 , and Appendix F. Piles 6B and 7 are
shown in cross-section (Figure 69).
Blowr
For pile 6B the period of timer-scheduled operation lasted until
hr 138, at which point temperature-feedback control came into play
(Figure 70). Blower operation was demanded for lOOt of the time, and
the demand did not subside,
For pile 7 the initial timer period lasted until hr 56, followed
by feedback control. Demand built to a peak of 65%, at hr 110, then
gradually subsided. Demand terminated at hr 344, with the resumption
of timer-scheduled operation.
Temperature
In pile 6B the time-zero temperature was 8-11 C (Figure 71).
At most of the positions the temperature ascended gradually, though at
an accelerating rate, so that at the control thermistor (positio. I)
138 hours elapsed before reaching .cet-point (45 C) . The temperature
pattern at some of the outermo t petitions was erratic. Feedback con-
trol did not arrest the temperature ascent. Whereas the temperature
at the thermistor should have stabilized at 45 C, it did so at 68°C.
Elsewhere, higher temperatures generally prevailed.
In pile 7 the time-zero temperature was 18-22°C. The tempera-
ture ascended faster, such that at the control thermistor (position
6) 56 hours elapsed before reaching set-point (45°C). The onset of
temperature feedback control arrested the ascent at the thermistor
at 4S°C, and elsewhere the design ceiling of 60°C was rarely exceeded.
Oxygen
In pile 6B the lov/est 02 level observed v/as 10% (Figure 72).
The comparable observation for pile 7 v/as 14%. Thus, in both piles
oxygenated conditions were maintained.
Moijture Content
In pile 6B the moisture content decreased from 72% to 68%
(Figure 72), The comparable decrease in pile 7 was from 62% to 28%.
Thus, the drying tendency was strongly expressed only in pile 7.
123
-------
PILE SB
PILE 7
Figure 69. Cross-sectional representation of piles 6B and 7 (see also Section 3). Tex-
tured area, woodchip cover and base; clear area, sludge-woodchip mixture; circles,
perforated flexhose; positions 1-16, thermocouples in a 0.3 m x 0.5 m grid. In pile
6B the thermistor was at position 1, and in pile 7 it was at position 6. The wood-
chip base dimensions (L x W x H) were: pile 6B, 5.2 m x 4.6 m x 25 cm; pile 7,
3.1 m x 2.7 m x 25 cm. The bottom-most level of the sludge- VOCK' chip mixture had
areal dimensions (L x W) of 4.9 m x 4.3 m (both piles). The piles ere slightly
longer in the axis parallel to the flexhose than in cross-section.
-------
100
P60
fe
a eo
S40
cc
Q
20
O
CD
PILE 6B
. s
100
200
300
400 0 100
TIME IN HOURS
200
300
400
500
Figure 70 Reprinted by permission from TOXIC AND HAZARDOUS WASTE: Proceedings of the
15th Mid-Atlantic Conference, pp. 463-471. Copyright c 1983. Buckneli Universi.v
Blower ooeration, mean of four hour intervals (see also Section ^;.
-------
PILE 6B
801- PROBE 68-1 —
PROBE 6B-I3 —
PILE 7
PROBE 7-6
PROBE 7-14
200
300
400 0
TIME IN HOURS
100
aoo
300
400
500
Fioure 71 Reprinted bv permission from TOXIC AND HAZARDOUS WASTE: Proceedings o, the
15th Mid-Atlantic Conference, pp. 463-471. Copyright c 1983. Bucknell University
Pi^e temperature at selected interior positions, plotted every four hours. Contiol
thermistors were at positions 6B-1 and 7-6 (see also Section j) -
-------
80
70
SO
S so
o:
40
30
20
20
16
12
02 CONCENTRATION
PILE 6B--
PILE 7 —
100
200
300
400 0
TIME IN HOURS
100
200
300
400
5OO
Figure 72. Reprinted by permission from TOXIC AND HAZARDOUS WASTE: Proceedings of the
15th Mid-Atlantic Conference, pp. 465-471. Copyright, c 1985. Bucknell University.
samples taken from the pile interior (see also Section 5.)
Left, moisture content of
concentration of 0
probes were at position
2 (v:v) in gas
15 (pile 6B) and
samples taken ac four
position 6 (pile 7).
hour intervals. Sampling
-------
With the recognition that pile 6B failed, an effort was made
to diagnose the responsible operational flaw(s). The suspected
immediate cause of failure was inadequate penetration of air into the
mass, fading to inadequate heat removal. The extent of air pene-
tration was tested, on day 20, with the use of externally heated
air (130 C measured in the flexhose duct between blower and pile).
Thirty minrt.es of continuous input of heated air resulted in
a 33,8°C temperature elevation in tV^e woodchip base at the monitor-
ing position nearest the flexhose, and a 10.3 C elevation at the
furthest base position (Figure 73). The maximum elevation in the
sludge-woodchip mixture itself (1.1°C) was immediately above the flex-
hose. This indicates that most of the air passed horizontally through
the woodchip base, rather than passing up through the sludge-woodchip
mixture.
DISCUSSION
Two of the factors influencing air penetration into the compost-
ing mass are the porosity of the sludge-bulking agent mixture, and the
design of the woodchip base. Porosity is affected by the sludge-bulk-
ing agent ratio and by the moisture content, among other factors. Both
insufficient porosity and poor base design probably contributed to the
failure of pile B.
In routine practice, the more favorable porosity represented by
pile 7's sludge: bulking agent ratio and initial moisture content
should be an operational goal. However, variations in po.-osity are
unavoidable. In contrast, woodchip base construction is repeatable
without variation, and its design should be optimized. The base design
represented by pile 7 is superior in that it suppresses short-circuit-
ing, directing the air to the overlying mass.
One aspect of the comparative performance of these piles ic
explained by ambient temperature. This is the length of time needed
to reach the set-point temperature, which was 2.5x lengthier for
pile 6B than pile 7. The lengthier "come-up" is attributable to the
lower, biologically unfavorable, time-zero ambient temperature. The
explanation of the other performance aspects resides in the factors
governing the composting ecosystem, as influenced by feedback ventila-
tion control.
With adequate air penetration, represented by pile 7, demand
for heat removal is met via blower operation. This induces an inter-
action between pile and blower manifested in time-variable blower opera-
tion, leading to the regulation of pile temperature. This results in
rapid decomposition, rapid drying, and an oxygenated condition. The
first two factors are intimately linked, in that decomposition gener-
ates heat - which vaporizes the water. The oxygenated condition re-
flects the balance between a high rate of oxygen consumption and a
commensurately high rate of oxygen resupply.
128
-------
Figure 73. Reprinted by permission from TOXIC AND HAZARDOUS WASTE: Proceedings of the
15th Mid-Atlantic Conference, pp. 463-471. Copyright c 1983. Bucknell University.
Pile 6B; temperature changes after 30 minutes continuous input of externally heated
air.
-------
Non-adequate air peretration, represented by pile 6B, unlinks
blower demand and heat removal, in that the air (or much of it) is
short-circuited. Hence, pile and blower do not interact. The con-
sequent failure to arrest the temperature ascent leads to inhibitive
levels; consequently, decomposition and drying are slight. An oxy-
genated condition prevails, because the slight rate of oxygen re-
supply suffices to roughly match the slight rate of consumption.
Finally, these observations are put into a broader perspective
by comparing the nominal behavior of a composting pile managed accord-
ing to the Rutgers strategy (.e.g. pile 7), the nominal behavior of a
Beltsville-type pile (e.g. pile 933), and a failed pile intended as a
Rutgers pile (6B) (TABLE IS). Nominal Rutgers behavior requires a
ventilation system adequate in three respects. First, the blower must
respond, in a time-variable fashion, to the needToTneat removal in
reference to temperature (temperature-feedback control). Second, the
blower must be adequately sized, to meet the peak demand fo*r~veritila-
t^on' Third, the air must pass through the mass reasonably freely.
The system represented by pile 7 was adequate with respect to all three
factors; the pile 6B system was deficient with respect to the third
factor; the pile 9B system was deficient with respect to the first and
second factors.
Thus, piles 6B and 9B had different deficiencies in terms of
their ventilation systems. The outcome was similar, however, in that
excessive heat accumulation was manifested in inhibitively high tem-
perature coupled vrith a high level of 0^. This demonstrates that a
processing failure by Rutgers standards resembles nominal performance
by Beltsville standards.
130
-------
TABLE 15. NOMINAL AND FAILED BEHAVIOR OF COMPOSTING SYSTEMS
Nominal behavior
Temperature
Drying tendency
Demand for ventilation
Level of peak demand
Duration of peak demand
Interstitial atmosphere
Temperature gradient
Rutgers
Regulated, as
intended
Strong
Builds gradually
to peak
Usually <100%
Hours or days,
then dwindles'
Oxygenated1*
Decisively
established
Beltsville
Regulation not
attempted, iinhibi-
tively high
Weak
N/A
N/A
N/A
t
Oxygenated
Weakly
established
Failed
behavior
Rutgers
Regulation
failed, inhibi-
tively high
Nil
Euilt instan-
taneously to
peak
100%
Sustained
indefinitely
f
Oxygenated
Weakly
established
Result of rapid 02 uptake balanced by rapid resupply (see Section
^Result of slow 02 uptake balanced by slow resupply (see Section 3)
3).
-------
SECTION 11
PATHOGEN INACTIVATION
INTRODUCTION
Composting can serve as a waste treatment technology owing to
its capacity to stabilize and sanitize the material. The components
of stabilization are the decomposition of putrescible matter, the
reduction of volume and weight, and the removal of the water. All
of these are advanced through maximization o£ the rate of decomposi-
tion. The material is sanitized through biological antagonisms that
inactivate or destroy pathogenic organisms, and through temperature-
inactivation. Biological antagonisms, though poorly understood, are
presumably also promoted through maximization of the decomposition
rate, as this is synonymous with the general level of biological
activity. To this extent, therefore, the goals of stabilization and
sanitation are advanced in tandem. With respect to temperature-in-
activation, however, a potential for conflict exists. Inactivation
through this mechanism is positively related to temperature throughout
the range of possible composting temperatures (peak ~ 80 C), whereas
the threshold to a significant decrease in decomposition rate is
approximately 55 -60°C.
This potential conflict would be aggravated by an administrative
regulation calling for "the fastest possible pathogen inactivation"
(15). A regulation so worded would mandate the earliest possible
attainment of the highest possible temperature. However, the Federal
interim final criteria for pathogen reduction (50) are worded in a
mann<;r that permits a degree of operational flexibility in meeting the
indicated goals. The criteria are for "significant pathogen reduc-
tion" (at least 40°C for 120 consecutive hrs and, within this period,
at least 55 C for 4 hrs), and "further pathogen reduction" (at least
55°C for 72 consecutive hrs).
These criteria can be met with little or no penalty in terms
of stabilization and water removal through tactics involving either
the high rate stage or the curing stage. The former is demonstrated
experimentally in the present section; the latter is noted in discussing
the problem of monitoring regulatory compliance.
MATERIALS AWD METHODS SPECIFIC TO THIS SECTION
The trial involved a 6 tonne pile of the sludge-woodchip mix-
ture (pile 13). In addition to the usual thermocouple grid in the
mixture, four thermocouples were positioned in the woodchip bed
(Figure 74). (Note that the mixture was accidentally placed in an
off-center orientation relative to the flexhose duct.) The controller
132
-------
Figure 74. Pile 13, cross-sectional representation: textured area, woodchip cover and
base; clear area, sludae-woodchip mixture; circle, perforated flexhose. The
numbers indicate the monitoring and control positions: thermocouples, positions 1
through 19; gas sampling probes, adjacent to positions ^ and 16; control
thermistor, position 5. The blower was operated in the forced-pressure mode.
(The sludge-woodchip mixture was accidentally offset realtive to the woodchip base
and this pile, lacking the usual slight longitudinal axis, was nearly symetric.)
-------
set point was 4S°C. The experimental plan was to shut down the blower
before termination of the period of temperature feedback control, and
this was done at hr 176.
The trial period was 27 February 1981 to 18 March 1981. Ambient
air temperatures during this period were (°C): mean of the daily
highs, 9; mean o£ the daily lows, zero; mean of the daily means, 4.5;
range of the daily highs and lows, -7.8 to 13.9. Precipitation
amounted to 7.0 cm in 4 occurrences.
RESULTS
Temperature feedback control commenced at hr 48 (Figure 75).
Blower operation time increased sharply, until reaching 1001 at hr 66.
The I time on started decreasing at hr 138, reaching approximately
61% at hr 176, At this time the blower was deliberately shut dovm.
The temperature record at the innermost series of probes is
presented in Figure 76 (all of the temperature data are in Appendix
G) . Early in the period of temperature feedback control (hr SO to
6S) the temperature ascent at position S (site of the control thermistor)
paused at approximately the set point value (4S°C). The ascent resumed,
however, despite 1001 blower on time, reaching a peak of 58QC at hr 100.
The temperature subsequently declined and, starting at hr 140, stabi-
lized at the set point level. As is characteristic of the Rutgers
strategy, a positive temperature gradient in the direction of airflow
became established, Deliberste blower shutdown at hr 176 marks the
start of a new temperature ascent.
Other temperature records representative of this trial are shown
in Figure 77. Probe 1, positioned in the woodchip base adjacent to
the flexhose duct, experienced only a slight temperature ascent (10°C
to 27°C) prior to blower shutdown, at which point there was a sharp
upturn in the temperature (27 C to 7i , 3C). Within the sludge-woodchip
mixture the sharpest post-shutdown temperature increase was at posi-
tion 6 (34°C to 78°C). Two positions near the edge of the pile, 14
and 9, did not increase in temperature subsequent to blower shutdown,
but rather decreased.
The temperature data from all of the positions are summarized in
TABLE 16 to show the temperature increase resulting from blower shut-
down. The mean increase (positions 9 and 14 omitted) v/as 27.7°C in
20.4 hours (!„ 36°C/hr) .
Gas __An_aly^sejS_._
Prior to shutdown C02 was below the limit of detection and 0?
was at the ambient concentration (Figures 78 and 79). Abrupt changes
(CO-, increase and 0,,, decrease) coincided with the cessation of
mechanical ventilation.
134
-------
100-tl
2
O
£
CJ
a:
2
O
(-40
o:
Q
ET
LU
iOO
200 300
TIME IN HOURS
400
500
Figure 75. Pile .13, blower operation. The baseline represents operation as scheduled
by timer, and the area above the baseline represents blower operation through the
temperature-feedback control system. The blower, which was operated in the
forced-pressure mode, was shutdown at hr 176.
-------
ao
r
oSO
PROBE 13-le
PROBE 13-15
PROBE 13-10
PROSE \$-
TIME IN HOURS
400
500
Figure 76. Pile 13, temperature at the innermost vertical series of thermocoup
>les.
-------
Figure 77. Pile 13,
temperature at re-
presentative posi-
tions. Top graph,
position 14; middle
graph, position 6;
bottom graph,
position 1.
oo
60
40
20
0
80
60
80 ~
60 -
40 -
20 -
100
200 300
TIME IN HOURS
400
500
137
-------
TABLE 16. PILE 13:
TB4PERATURE PEAKS RESULTING FROM SIWIDCMN OP THE B3JWER
AT HR 176
Therrao-
couplea
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Pre-shutdown
Temp
iM
27
36
57
53
57
65
65
63
67
64
72
73
72
68
72
73
78
73
74
jpeak
Time
Ihrl
175
110
128
100
100
90
100
100
90
92
85
85
88
88
85
85
85
90
85
Temp (°C)
just prior
to shutdown
27
18
27
31
45
34
37
42
49
54
44
40
47
47
61
50
48
62
49
Post-shutdown
Temp
CQ
71
67
66
58
75
72
66
60
-t
76
75
73
65
-t
74
76
67
76
71
peak
Time
(hr)
240
204
200
186
206
200
190
185
-
208
194
192
189
_
200
194
194
200
197
Differential (°C)**
44
49
39
27
30
38
29
18
-
22
31
33
18
-
13
26
19
14
22
* Thermocouples 1 through 4 were positioned in the woodchip base; the other thermocouples were
in the sludge-woodchip mixture (see Figure .75).
"* Post-shutdown peak temperature minus temperature just prior to shutdown.
t The temperature decreased after blower shutdown.
-------
100
200 300
TIME IN HOURS
400
500
•igure 78. Pile 13,
(lower curve) .
position 6.
concentrations of 02 (upper curve) and CO.
The gas sampling probe was adjacent to
139
-------
100
200 300
TIME IN HOURS
400
500
Figure 79, Pile 13,
(lower curve).
position 16,
concentrations of 02 (upper curve) and C0?
The gas sampling probe was adjacent to
140
-------
During the first 176 hrs the moisture content decreased from
604 to 43% (Figure 80). In the post-shutdown period a further
decrease to 27$ occurred.
DISCUSSION
During the period of comparability (up to hr 176) the temperature
control achieved in this pile was less precise than in the previous
Rutgers piles. This is attributed to the accidental off-center
position of the sludge-woodchip mixture relative to the flexhose
(Figure 74), presumably leading to uneven air distribution. Thus
two factors were non-ideal with respect to the stabilization-water
removal objectives: the non-deliberate offset position of the compost-
ing pile relative to the flexhose, and the deliberate early shutdown
of the blower (hr 176). Despite these factors process performance
was satisfactory, judging from the moisture content decrease and
informal observations of the odor and visual appearance of the material.
Questions of pathogen inactivation aside, the satisfactory performance
of pile 13 is indicative of the reliability ol' the Rutgers strategy,
given indifferent routine field practice.
As intended, deliberate blower shutdown induced a sharp tempera-
ture upturn. The exceptions (sites 9 and 14) were near the pile's
outer edge. The ambient temperature during this trial (mean, 4.5°C)
was the coldest of the entire investigation.
These temperature data were analyzed with respect to the federal
interim final criteria for pathogen reduction in static-pile compost-
ing (48). Three time periods are considered separately (TABLE 17).
During the first period the temperatures at positions 1 through 4 (the
woodchip base) and position 6 djd not meet either of the criteria.
The other positions met either the criterion for significant pathogen
reduction (spr) or the criterion for further pathogen reduction (fpr).
During the post-shutdown period positions 7, 8, 9, 13, 14s and 17
did not meet either of the criteria. Considering the entire trial
period, only position 4 (in the woodchip bed) did not meet the cri-
terion for further pathogen reduction, and this position met the
criterion for significant reduction. Thus, the induction of harsh
temperatures through deliberate blower shutdown improved performance
with respect to the federal interim criteria for pathogen inactivation..
The other Rutgers piles (7, 8, 9A, 11A, 11B, 11C) were not
deliberately subjected to harsh terminal temperatures and did not
perform as well with respect to the federal criteria (TABLE 18).
Overall, 76% of the monitoring positions in these piles met the
significant reduction criterion;, and 411 met the further reduction
criterion. The Beitsville pile (9B) met the criterion for further
pathogen reduction at all of the monitoring positions. Pile 6A met
the criterion for further pathogen reduction at 12 of the 13 sites
(see footnote - TABLE 17).
-------
805
100
200 300
TlfcSE IN HOURS
400
SCO
Figure 80. Pile 13, moisture content. Samples taker from central interior locations
-------
TABLE 17. PILE 13: TEMPERATURE DATA
TERMS OF THE FEDERAL INTERIM FINAL CRITERIA
Pre- shutdown
(hr zero to hr 176)
Thermo-
couple
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
a
Spr =
Fpr =
Hrs Hrs
£40°C ->55 C
None
None
75
70
130
92
124
132
126
130
130
128
128
136
264
132
128
118
' 125
significant
consecutive
further p&tl
None
None
14
None
28
55
72
66
81
74
82
88
92
86
190
90
97
98
98
pathogen
period ^
Criter-n
ion met
None
None
None
None
Spr
None
Fpr
Spr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
reduction
5SdC).
tiogen reduction (72
Hrg
155
128
122
112
230
130
114
31
12
280
236
122
86
6
132
236
118
270
250
(120
Post-shutdown
' (hr 176 to hr 445)
Hrg
120
95
100
5
102
98
40
6
None
121
105
42
20
None
90
104
38
130
90
consecutive hrs
Entire trial period
(hr zero to hr 445)
Criter-* Hrs Sirs
ion met ±40°C >-S5°C
Fpr
Fpr
Fpr
Spr
Fpr
Fpr
None
None
None
Fpr
Fpr
' Spr
None
None
Fpr
Fpr
None
Fpr
Fpr
i 40°C,
155
128
197
182
360
222
238
163
138
410
366
250
214
142
396
368
246
388
215
plus 4
12G
95
114
5
130
153
112
72
81
175
187
130
112
88
280
194
135
228
188
Cnter-e
ion met
Fpr
Fpr
Fpr
Spr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
hrs within the
consecutive hrs L 55°C).
-------
TABLE IS. ALL PILES: SUMMARY OF THE TEMPERATURE RECORDS WITH RESPECT TO THE FEDERAL
INTERIM FINAL CRITERIA
Pile
(report
Section)
7(3)
8(3)
9A (3)
4A (4)
11A (5)
11B (5)
11C (5)
13 (11)
9B (3)
6A (10)
Control
strategy
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Beltsville
N/A*
Mean
ambient
temp ( C)
15
21
27
14.5
9
9
9
4.5
27
11
Terminal
temp
elevation
No
No
No
No
No
No
No
Yes*
N/A
N/A
Temp
Total
(no. )
15
31
19
12
12
12
12
19
IS
13
monitoring
sites
Perforraance
with respect to
stabilization
and water removal
Meeting indicated
criterion
Spr
12
20
17
11
11
8
7
19
15
*
12V
(no . )
Fpr
6
6
13
2
6
6
7
18
15
12*
Good
Good
Good
Good
Good
Good
Good
Good
Poor
Failed
Terminal harsh temperatures induced (present Section)
The intention was to Jnanage this pile according to the Rutgers process
ing failed because of inadequate air penetration (see Section 10).
but the process-
One position (position 5 - see Figure 74) did not literally meet either criterion, but
it experienced the following temperatures: peak, 71°C; 49 consecutive hrs at ^ 55°G;
116 hrs (in three intervals) at > 40°C.
-------
In contrast, with respect to the stabilization-water removal
objective the Rutgers piles performed well, the Beltsville pile
performed poorly, and pile 6A failed.
These observations should be put into perspective. Except for
the present Section, this investigation concerns the stabilization-
water removal objective. As such, the thermocouples were not posi-
tioned specifically for the purpose of determining compliance with
the Federal criteria. For example, pile 9B (Beltsville process)
lacked a series of thermocouples next to the interface with the
woodchip cover (compare Figures 17 and 18). The material in this
region is suspect because of its proximity to the pile's edge, and
because the vacuum- induced direction of ventilation would further
promote coolness. Another deficiency, from the viewpoint of monitor-
ing for compliance with the Federal criteria, is the lack of probes
in the pile "toes." This refers to the part of the pile forming
a triangle with the woodchip base or the concrete pad. Because of
the direction of ventilation, the "toes" of Beltsville piles are
particularly subject to coolness. These considerations illustrate
the obvious point that the reliability of "compliance data" is
strongly dependent on the frequency of monitoring and the selection
of positions for the monitoring.
The problem of monitoring may be less intractable in routine
practice than might appear from a consideration of these freestand-
ing piles. This is because the extended pile geometry would pre-
sumably be used for the Rutgers strategy in static pile configuration,
as it is in routine Beltsville- type operation. The- ratio of exposed
edges to interior volume is less with such geometry.
Despite the accumulation of considerable routine opera.tional
experience with the Beltsville process , published systematic studies
of the capability of this process to ..cr.ply with the Federal interim
final criteria seem to be lacking. Furthermore, other than pile 9B,
we know of no published report on the Beltsville process which provides
sufficiently detailed data to support an interpretation of the time-
temperature observations in reference to the criteria.
An independent side-by-side comparison of the Rutgers and Belts-
ville approaches, involving a mixture of refuse and sewage sludge,
included tests for Salmonella as one of the many points of compari-
son (34). All of tfieTIsts relating to stabilization, including
changes in moisture content, indicated that the Rutgers strategy
gave superior performance. With respect to the decrease in
both processes performed comparably.
TABLE 19 summarizes our thoughts regarding the relative merits
of the Rutgers and Beltsville approaches with respect to the pathogen
inactivation objective. Some of the points are necessarily specula-
tive. The present state of knowledge does not indicate that one, or
the other, of these approaches is superior with respect to sanitation.
Both are highly effective in this respect.
145
-------
TABLE 19. RELATIVE MERITS OF THE RUTGERS HIGH RATE STAGE AND THE BELTSVILLE
"ACTIVE STAGE" WITH RESPECT TO SANITATION.
P_o_int_ of comparison
Comments
Harsh temperatures
(e.g. 7SDC-8QOC)
Biological antagonisms
Cool spots
Dryness as a sanitizing agent
Observed Salmonella reduction
Potential for regrowth of
Salmonella
Intrinsic to the Beltsville pro-
cess; inducible as a terminal step
of the Rutgers strategy.
Presumably more effective in the
Rutgers strategy.
"Jogs*": Beltsville more subject
to pro¥lem.
Outer_e_dge_sj_ Beltsville more
sTTEJFcTto~~p*r o b 1 e m.
Interior_(near_f 1exhose) : Rut-
gers mor£j~s~u B j e~c tto~p~FoFlem (but
not if terminal harsh temperatures
induced).
Operative only in the Rutgers
strategy.
Rutgers and Beltsville approaches
performed comparably (34).
Less potential in Rutgers
strategy, as substrate is more
thoroughly decomposed.
146
-------
Finally, the potential role of the curing stage in assuring
product safety should be considered. The advantages o£ emphasizing
the curing stage to insure sanitation, relative to the high rate
stage, are as follows. 1) The curing stage is closer, with respect
to time, to the point of end-product usage. This decreases the
opportunity for post-processing recontamination or regrowth. 2)
Large well-insulated curing piles presumably will gradually self-
heat to harsh temperatures, even though formed of material thoroughly
stabilized and dried in.the high-rate stage. 3) Large piles have a
relatively low surface to volume ratio. Thi^ minimizes the volume
of "edge material" exposed to ambient conditions. 4) It seems prob-
able from item 3) that, in routine practice, monitoring of the curing
stage would provide the more reliable time-temperature data to insure
pathogen inactivation.
Composting is an excellent means of sanitizing waste. Regard-
less, we believe that the high-rate stage should be managed primarily
to accomplish stabilization and water removal objectives. If the
intended use of the end-product requires additional assurance of
pathogen reduction this can be achieved through a terminal harsh
temperature phase of processing, or through a curing strategy designed
with product safety assurance as one of its objectives.
147
-------
SECTION 12
UNIFORM PROVISION OF AIR ALONG THE LENGTH
OF A COMPOSTING PILE"
INTRODUCTION
If the composting pile is relatively long, and the area of air
holes is uniform along the length of the duct, delivery of air to
the pile is non-uniform. This is because more air exits from the part
of the Juct close to the blower than from the distant part. With the
control thermistor placed midway between the ends of the pile, the
part of the pile close to the blower is overventilated (too cool) and
the more distant part underventilated (too hot).
A means of providing air uniformly in the longitudinal dimension
is to vary the size and/or spacing of the air holes along the length
of the duct, such that the amount of air vented is similar regardless
of distance from the blower. A computer program, presented herein,
was developed to help the user determine the number of evenly spaced
holes, of a given hole size, necessary to supply a given amount of
air to a 5 footTsection of duct. Each succeeding S foot section is
treated separately. A spacing scheme for one to 20 holes per section
is stored in the program. Depending on the length of the pile, it
may.be advantageous to increase the size of the holes in progressive
sections of the duct.
CALCULATION4
The airflow out of each hole in the duct is calculated through
the use of two equations (52). The first involves the pressure drop
in the duct, as follows:
D5
where: f - the Moody friction factor (dimensionless)
L = the length of duct (ft)
p = fluid density (slugs/ft3)
q - flow (ft/sec)
D ~ duct diameter (ft).
Comments on these factors are as follows :
*An early, fTal7e^T~v^i'sTon~~6^f"'tKrs"^5ection was published (51). A
correction was submitted to the publisher.
tTo change feet to meters, multiply by 0.305.
tV/e thank the following individuals for kindly reviewing the calcula
tions: James H. Miller, Roberto C. Leon, Dr. Robert C. Ahlert.
148
-------
Ntoody friction factor (f). Tables of values of this factor are found
in"TiuT3~3ynariu"cs TeTcHooKS . Also, the Fanning friction
factor may be used (Fanning factor = Moody factor/4). For the pre-
sent application (commercial pipe serving as duct, high velocity) the
Moody friction factor is roughly 0.02.
This refers to the total length of duct, includ-
_____
TngpaFalTel~Ten~gtEs where used.
The value used in the program is p a 0.00238
This is the mean cross-sectional velocity.
£iajnete£__(Dj_._ The user selects the duct diameter.
The second equation is used to calculate the airflow out of
each hole, as follows:
/
,y
0 => 0.00?ld2 duct "exit (ii)
e ___ _
where: q = the exit flow (ft/3sec)
P 6 2f
duct ° the pressure in the duct at that point (Ib/ft )
d = hole diameter (in.)
P
exit s the pressure in the pile
K c constant
Comments on these factors are as follows:
Pressure in the pile. Since instantaneous pressure should not vary
aTo~rf g~THe-TengtTioT~t h e pile, it is set equal to zero.
Constant. In this program K is set equal to 1.5.
COMPUTER PROGRAM - BACKGROUND
List of thingj_tjia/t the user^jnust Jcnow to enter the program
Clb/ft ). The pressure at the entrance
_.^.--.--
• t"6 ^He~ fTFsT~JeTtTon~T2^°aucT~mu s t be kn n wn . The pressure at this
point is the pressure at the blower (fc ^. , as specified by the
manufacturer) minus the pressure drop over the distance to the first
section. The user should also subtract the pressure loss from the
ductwork to the outer edge of the pile (the pile backpressure). If
this is not known, a conservative estimate is 1.0 Ib/ft2. The
To change slugs/ft3 to kg/m3, multiply by 3.99 x 10"4.
tTo change Ib/ft2 to kgm/ra2, multiply by 4.873.
149
-------
pressure at the exit of the first section is the entrance pressure at
the second section, etc. The pressure at the blower must provide
adequate pressure to the terminal section of duct. This is satisfied
when the number of holes required in the terminal section is <20,
and the required hole diameter is reasonably less than the duct (pipe)
I.D.. The use of a blower delivering higher pressure than necessary
yields a solution involving fewer holes and/or smaller holes, but this
represents an uneconomical approach. A_suggested first approximation
for pressure at the blower is 6.0 Ib/ft .
to duct (cfs) * . The estimation of the required air
~
the" piTe is based on pile 9A (Section 3)
as this is the largest pile in our experience (40 ton) ,"*' and as it
yields the largest (hence the most conservative) estimate. Further-
more, the estimate is conservative for use with other sludges, as the
experimental material was primary sludge with a high volatile solids
content (^ 75%). This pile was ventilated by six 1/3 hp blowers at
each end of the pile. The blowers were operated in unison. Each
blower discharged into a 20 foot length of perforated duct, which was
capped at the distal end. Since the total length of perforated duct
was 120 feet, for calculation purposes there are twenty-four S foot
lengths .
$
The peak demand exerted was 71.3 cfm per wet ton (initial
weight) , which is rounded off to 80 cfm per ton for purposes of the
estimation. Since the pile weighed approximately 40 tons, the total
peak demand was 3200 cfm^. The total length of pipe was 120 feet,
arranged in three parallel branches of equal length. Thus there were
twenty-four S foot sections, with each section requiring 133.3 cfm.
For use in the program this is converted to cfs (2.22 cfs) .
The use of standard pipe ID's is recommended.
This depends on duct size, material, and airflow,
.
, L__in . J2nds of an inch. Different values should be entered
™ the~*piro'gf am to find the best combination of number
and size of holes for each S foot section of duct. An input of 99
will end the program.
To change from cfs to m-^/sec, multiply by 2.83 x 10"2.
-j.
To change ton to tonne, multiply, by 1.1023.
t ^
To change from cfm/ton to m /sec-tonne (metric), multiply by 4.232 x
10-4.
To change from cfm to m /sec, multiply by 4.720 x 10" .
ISO
-------
Output
XXXX/32 inch* XX holes M£I§, XXX.XXPSF
The hole size The number of Exit flow Pressure at the
(inputted) holes selected from the end of the
by computer holes section
Two output values are provided each '.time the program is run. The
first is the value for the number of holes giving slightly less than
the desired exit flow. The second is the value for the number of
holes giving slightly greater than the desired exit flow.
Npjte_s
a. The first step in putting the program on line is to es-
tablish the file on hole spacing. This is provided in TABLE 20
(the "H.DAT" file).
b. All of the values inputted to run the program must be real
numbers, rather than integers. Alv/ays include a decimal point.
List o f
AL (20,21) Array storing distance between each hole for 1-20 holes.
Does not change through the program.
PREF Duct pressure in PSF. Changes in line 290 } 330 and 240.
PREM Duct pressure in PSF. Does not change. Used for program to
remember initial pressure.
CFS Flow in pipe (CFS) . Changes in lines 270 and 310.
CFSM Flow in pipe (CFS). Does not change (used like PREMJ
FF Friction factor. Does not change.
DUCT Duct diameter in feet. Does not change.
EXF Exit flow from holes In CFS. Does not change.
I (First use) Counter for filling "AL" array (Line ISO).
J Counter for filling "AL" array (Line 150) and line 320.
DIA Exit hole diameter (32nd1 s of an inch). Changes in line 180.
N Counter used in pulling data from "AL",, (Depicts number of
holes.) Changes in lines 200 and 220.
FL Exit flow from holes (CFS). Changes in lines 250 and 300.
FS Exit flow from holes (CFS) 0 Acts to remember most recent value
of "FL". Changes in line 230.
to change from inches to cm, multiply by 2.54.
151
-------
TABLE 20. "H. DAT' »iOLE DATA) FILE
0.
0.
0.
0.
0.
0.
0.
0.
0.
5. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0o 0. 0
2.5 2.5 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
1.667 1.667 1.667 0. 0. 0. 0. 0. 0. 0. 0. 0
1.25 1.25 1.25 1.25 0. 0. 0. 0. 0. 0. 0. 0.
1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0, 0. 0. 0
0. 1.667 0. 1.667 0. 1.667 0. 0. 0. 0. 0. 0
.7083 .7083 .7083 .7083 .7083 .7083 .7502 0
0. 1.25 0. 1.25 0. 1.25 0. 1.25 0. 0. 0. 0.
.5521 .5521 .5521 .5521 .5521 .5521 .5521 .
. 0.
0.
. 0.
0.
. o.
. 0.
. 0.
0.
5521
0
0
0
0
0
0
0
0
0 0.
0.
. 0.
0.
. 0.
. 0.
. 0.
0.
5832
0
0
0
0
0
0
0
0
0
. 0
0.
. 0
0.
. 0
. 0
. 0
0.
. 0
0.
. 0
0.
. 0
. 0
0.
. 0
. 0
0.
. 0
. 0
0.
. 0
. 0.
0.
. 0.
. 0. 0. 0. 0. 0.
0.
. 0. 0. 0. 0'. 0. 0.
0.
0. 0. 1. 0. 1. 0. 1. 0. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. .4583 .4583 .4583 .4583 .4583 .4583 .4583 .4583 .4583 .4583 .4170 0. 0. 0. 0. 0. 0.
0. 0. 0.
0. 0. .8333 0. .8333 0. .8333 0. .8333 0. .8333 0. .83?5 0. 0. 0. 0. 0. 0. 0. 0.
0. .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3752 0. 0.
0.0.0.0.0.
0. 0. .7083 0. .7083 0. .7083 0. .7083 0. .7083 0. .7083 0. .7502 0. 0. 0. 0. 0. 0.
0. .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333
,3338 0. 0. 0. 0. 0.
0. 0. .625 0. .625 0. .625 0. .625 0. .625 0. .625 0. .625 0. .625 0. 0. 0. 0.
0. .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917
.2517 .2917 .3328 0. 0. 0.
0. 0. .5521 0. .5521 0. .5521 0. .5521 .0. .5521 0. .5521 0. .5521 0. .5521 0. .5832 0.
0.
0. .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604
.2504 .2604 .2604 .2604 .3128 0.
0. 0. .5 0. .5 0. .5 0. .5 0. .5 0. .5 0. .5 0. .5 0. .5 0. .5
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
-------
PRES Duct pressure in PSF. Acts like "FS". Changes- in line 240.
I (Second use) Counter to pull data from "AL". Changes in line 280.
NC Simply N-l used to modify output of liie 370. Changes in line 360,
THE PROGRAM
@ Type Pipos, for
10 REAL AL(20,21)
20 WRITE(Sf440)
30 READ(5,*)DUCT
40 WRITE(5,4SO)
SO READCS/3 EXF
60 WRITE(5,400)
70 READ(5,*)PREF
80 PREM^PREF
90 WRITE(5,410)
100 READ(S,*)CFS
110 CFSM^CFS
120 WRITE(5,420)
130 READ(5,*)FF
140 OPEN (UNIT=29,DEVICE='dsk:'SFILE^'H.DAT')
150 READC29/) ((AL(I ,J) ,J=1,21) ,1 = 1,20)
160 CLOSE(UNIT=29)
170 WRITEC5,430)
180 READ(S,*)DIA
190 IF(DIA,EQ.99.)GO TO 390
200 N=l
210 GO TO 250
220 N=N+1
230 FS=FL
240 PRES=PREF
250 FL^O
260 PREF-PREM
270 CFS=CFSM
280 DO 3101=1,N
290 PREF=PREF-(AL(N;I)MCFS**2)/(DUCT**Sr'FF/519)
153
-------
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
FL=FL+(0. 000126* (DIA**2) * (PREF**0.
CFS=CFSM-FL
J-N + 1
S))
PREF=PREF- (AL(N,J)A (CFS**Z) / (DUCT**S) *FF/519)
IF(FL.LT.EXF)GO TO 220
WRITE (S ,460) N,DIA,FL,PREF
NC=N-1
WRITE(5,460)NC,DIA,FS,PRES
GO TO 170
STOP
FORMAT (' ', 'INPUT ENTRANCE
FORMAT (' ', 'INPUT ENTRANCE
FORMAT (' ', 'INPUT FRICTION
FORMAT (' ', 'INPUT DIA IN
INCH OR 99 TO END')
FORMAT (' ' , 'INPUT DUCT
FORMAT (' ', 'INPUT EXIT
OF DUCT IN CFS')
FORMAT (' ',12,' HOLES' ,F4.0,'
CFS' ,F8.2' , PSF')
END
PRESSURE IN PSF1)
FLOW IN CFS')
FACTOR')
32NDS OF AN
DIAMETER IN FEET')
FLOW PER 5 FT
732 IN.',F7.2',
DISCUSSION
In the developmental phase o£ the Rutgers strategy non-uniform
distribution of air was not a significant problem, as this involved
relatively small (pilot scale) piles. Moreover, the largest pile
(9A-40 tons) was served by blowers at both ends of the pile, and the
longest continuous length of perforeated duct was only 20 feet. Routine
operation, however, would involve longer piles, and a facility layout
having blowers at only one end is preferred. In the absence of a
specific design remedy, this would result in non-uniform ventilation and
degraded performance. One such remedy was developed herein.
1S4
-------
CONCLUSIONS
The composting system tends to accumulate metabolically generated
heat excessively, leading to in.iibitively high temperature. The
threshold to significant inhibition is approximately 60°C, and
its severity increases sharply at higher temperatures. At 80°C
(common peak temperature) the rate of decomposition is extremely
low.
This tendency can be controlled through ventilative heat removal
in reference to temperature. The main mechanism o£ heat removal
is evaporative cooling; establishing a drying tendency. Implemen-
tation is via temperature feedback control of a blower(s), using
standard (non-proprietary) equipment. The forced-pressure mode
of ventilation is moi3 efficient than the vacuum-induced mode.
In this manner an operational ceiling of 60°C is maintained.
Blower capacity (head and volume) must suffice to meet peak de-
mand for ventilation, &s expressed through feedback control. A
strong waste (e.g. raw sewage sludge) demands more ventilation
than a. weak one (e.g. digested sludge).
A temperature gradient is established along the axis of airflow,
whereas drying is relatively uniform. The temperature gradier.t
imposes a height limitation, above which a high rate of decom-
position is not realizable.
Managed thusly, decomposition and drying are related in that the
decomposition generates heat, the heat vaporizes water, and the
vaporization causes' drying. Hence, the stronger the drying
tendency the faster the sludge decomposition.
A consequence of temperature feedback control is that the compost-
ing mass is well-oxygenated, because more air is needed to remove
heat than to resupply 0^.
This strategy permits the use of recycled compost as the bulking
agent, in static pile configuration.
This strategy, compared to a conventional approach, resulted in
4x more sludge decomposition in half the time
The cost of ventilation for water removal through composting and
non-biological air drying was as follows: composting, $0.32/tonne
water removed ($0.31/ton); air drying, $6.43/tonne water removed
($6.33/ton). This difference results from the biological genera-
tion of heat at the expense of putrescible organic material in
the sludge.
155
-------
RECOMMENDATIONS
Maximization of decomposition rate should be the explicit goal of
composting process design and control.
Rate maximization should be approached through temperature feedback
control of a blower(s).
The rate of decomposition should be assessed in terms of the demand
for ventilation, and the course o£ drying.
This strategy (v/hieh focuses on temperature feedback control) should
be implemented at lowest possible construction cost, consistent with
operational considerations.
The unenclosed static pile configuration is structurally simple and
eminently suitable for implementation, and should be the preferi'ed
configuration.
156
-------
REFERENCES
1. Finstein, M.S., and F.C. Miller. Distinction Between Composting (the
Process) and Compost (the Product). Letter, BioCycle, 23 (6):56, 1982.
2. MacGregor, S.T., F.C. Miller, K.M. Psarianos, and M.S. Finstein. Com-
posting Process Control Based on Interaction Between Microbial Heat
Output and Temperature. Appl. Environ. Microbiol., 41:1321-1330,
1981.
3. Finstein, M.S., F.C. Miller, P.P. Strom, S.T. MacGregor, and K.M.
Psarianos. Composting Ecosystem Management for Waste Treatment. Bio/
Technology, 1:347-353, 1983.
4. Finstein, M.S., J. Cirello, S.T. MacGregor, F.C. Miller, and K.M.
Psarianos. Sludge Composting and Utilization: Rational Approach to
Process Control. Report to U.S. EPA, N.J. DEP, C.C. MUAS pp. 211,
New Jersey Experiment Station, New Brunswick (U.S. Dept. Commerce,
NTIS, Springfield, VA, No. PB 82 136243), 1980.
5. Miller, F.C., LJ.T. MacGregor, M.S. Finstein, and J. Cirello. Bio-
logical Drying of Sewage Sludge - A New Composting Process. In:
Proceedings of the 1980 National Conference on Environmental Engineer-
ing 3 Amer. Soc. Civil Engineers, New York, New York, p. 40-49, 1980.
6. Finstein, M.S., J. Cirello, S.T. MacGregor, F.C. Miller, D.J. Suler,
and P.F. Strom. Discussion of Paper by R.T. Haug (J. Water Pollut.
Control Fed., 51:2189-2206, 1979)". J. Water Pollut. Control Fed.
52:2037-2041, 1980.
7. Miller, F.C., S.T. MacGregor, K.M. Psarianos, and M.S. Finstein.
Direction of Ventilation in Composting Wastewater Sludge. J. Water
Pollut. Control Fed., 54:111-113, 1982.
8. Epstein, E., G.B. Willson, W.D. Surge, D.C. Muller, and N.K. Enkiri
A Forced Aeration System for Composting Wastewater Sludge. <7. Water
Pollut. Control Fed. 48:688-694, 1976.
9. Willson, G.B., J.F. Parr, E. Epstein, P.B. Marsh, R.C. Chaney, D.
Calacicco, W.D. Burge, L.J. Sikoras C.E. Teste, and S. Hornick.
Manual for Composting Sewage Sludge by the Beltsville Aerated-Pile
Method. U.S. EPA/U.S.D.A. Report, EPA-600/8-80-022, MERL/ORD,
Cincinnati, pp. 65, 1980.
10. Willson, G.B. and J.C. Thompson. Dewatering of Sludge Compost Piles.
In: Proceedings of the National Conference on Municipal and Industrial
Sludge Composting, Hazardous Materials Control Research Institute,
Silver Spring, MD.f p. 46-54, 1980.
157
-------
11. Sikora, J., G.B. Willson, D. Calacicco, and J.F. Parr. Materials
Balance in Aerated Static Pile Composting. J. Water Pollut. Control
Fed. 53:1702-1707, 1981.
12. Anonymous. Composting Processes to Stabilize and Disinfect Municipal
Sewage Sludge. EPA Water Programs Operations (WH-547) 430/9-81-011
MCD-79, Washington, D.C., pp. 43, 1981.
13. Willson-, G.D. and D. Dalmat. Sewage Sludge Composting in the U.S.A.
BioCyele, 24 (5) : 20-23, 1983.
14. Suler, D.J., and M.S. Finstein. Effect of Temperature, Aeration, and
Moisture on CO? Formation in Bench-Scale Continuously Thermophilic
Composting of Solid Waste. Appl. Environ. Microbiol.* 33:34S-350,
1977.
IS. Finstein, M.S. Composting Process Temperature: Conflict Between
the Fastest Possible Disinfection and Organic Matter Stabilization.
Report distributed at the "Workshops on the Health Effects and Legal
Implications of Sewage Sludge Composting," Cambridge, MA. pp. 12, 1978,
16. Finstein,- M.S.. Composting Hazardous Wastes.
1366-1367, 1979.
Letter, Science^ 204:
17. Finstein, M.S.. Composting Microbial Ecosystem: Implications for Pro-
cess Design and Control. Compost Science/Land Utilization, 21(4):
25-39, 1980.
18. Finstein, M.S., J. Cirello, D.J. Suler, M.C. Morris, and'P.P. Strom.
Microbial Ecosystems Responsible for Anaerobic Digestion and Com-
posting. J. Water Pollut. Control Fed., 52:2675-2685, 1980.
19. Finstein, M.S.. Heat Output - Temperature Interaction in Composting -
A Summary Report distributed at the "EPA Enclosed Municipal Sludge
Composting Meeting," Cincinnati, OH, pp. 5, 1981.
20. Finstein, Melvin S., Frederick C. Miller, Steven T. MacGregor, and
Kevin M. Psarianos. Rational Design and Control of Composting
Facilities. In: Proceedings of the Conference on Energy Conserva-
tion: Retrofit of Municipal Wastewater Treatment Facilities,
Argonne National Laboratory ANL/CNSV-TM-95, p. 19-25, 1982.
21. Finstein. M.S., F.C. Miller, S.T. MacGregor, K.M. Psarianos. Com-
posting Sludge: A Means of Linking Decomposition and Water Removal.
In: Proceedings of The Sisth Mid-America Conference on Environmental
Engineering Design, University of Missouri, Columbia, MO, p. 112-119,
1982.
22. Psarianos, K.M., S.T. MacGregor, F.C. Miller, M.S. Finstein. Design
of Composting Ventilation Systems for Uniform Air Distribution.
BioCycle, 24(2): 27-31, 1983.
23. Miller, Frederick C., Steven T. MacGregor, Kevin M. Psarianos, and
M.S. Finstein. Static Pile Sludge Composting with Recycled Compost
as the Bulking Agent. In: Industrial Vasts, Proceedings of the
Fourteenth Mid-Atlantic Conferences Ann Arbor Science, p. 35-44. 1982.
158
-------
24. Miller, Frederick C., Steven T. MacGregor, Kevin M. Psarianos, and
M.S. Finstein. A Composting Processing Failure: Diagnosis and Remedy.
In: Toxic and Hazardous Waste: Proceedings of the Fifteen th Mid-
Atlantic Industrial Waste Conference, Ann Arbor Science, p, 463-471,
1983.
25. Finstein, M.S.. Economic Motives for Managing the Composting Micro-
bial Ecosystem. Bio/Technology 1:341-342, 1983.
26. Miller, Frederick C., and Melvin S. Finstein. Equipment for Control
and Monitoring of High Rate Composting. In: Proceedings of the In-
ternational Symposium on Biological Reclamation and Land Utilization
of Urban Wastes, Universities o£ Naples and Pisa, p. 551-560, 1983.
27. Finstein, M.S., F.C. Miller, and P.P. Strom. Evaluation of Compost-
ing Process Performance. In: Proceedings of the Conference on Com-
posting of Solid Wastes and Slurries, The University of Leeds, 1983,
In Press.
28. Strom, P.P., F.C. Miller, and M.S. Finstein. Problem of Scale in
Composting Research. In: Proceedings of the Conference on Composting
of Solid Wastes and Slurries, The University of Leeds, 1983, In Press.
29. Finstein, M.S. and F.C. Miller. Principles of Composting Leading to
Maximization of Decomposition Rate, Odor Control, and Cost Effective-
ness. In: Proceedings of the Seminar on Composting of Agricultural
and Other Wastes, Commission of the European Economic Community,
Oxford University, 1984, In Press.
30. Miller, F.C. and M.S. Finstein. Materials Balance in the Composting
of Sewage Sludge as Affected by Process Control Strategy. Submitted
for publication.
31. Vestal, J. Robie, and Vicky L. McKinley. Operational Parameters of
Composting: Microbial Activity in Composting Municipal Sludge. Pre-
liminary Report to EPA, Contract No. CR-897852-01-0, pp. 145, undated.
32. MeKinley, Vicky L., and J. Robie Vestal. Biokinetic Analysis of
Adaption and Succession: Microbial Activity in Composting Municipal
Serfage Sludge. Appl. Environ. Microbiol. 47:933-941, 1984.
33. Hoitink, H.A,J., and G.A. Kuter. Factors Affecting Composting of
Municipal Sludge in a Bioreactor. Draft Report to EPA-MERL-ORD
CR-807791-010) , pp. 100, undated.
34. DeBertoldi, M., G. Vallini, A. Pera, and F. Zucconi. Comparison of
Three Windrow Compost Systems. BioCycle 23(2):45-50, 1982.
35. Stentiford, E.I., D.D. Mara, P.L. Taylor, and T.G. Leton. Forced
Aeration Co-Composting of Domestic Refuse and Sewage Sludge. In:
Proceedings of the Conference on Compacting of Solid Wastes and
Slurries, The University of Leeds, 1983, In Press.
36. Seton, T.G., P.L. Taylor, D.D. Mara, and E.I. Stentiford. Tempera-
ture and Oxygen Control of Refuse/Sludge Aerated Static Pile Systems.
In: Proceedings of the Conference on Composting of Solid Wastes and
Slurries, The University of Leeds, 1983, In Press.
1S9
-------
37. Toth, S. and N. Nocitra. Sludge Composting and Utilization: Chemical
Composition and Agricultural Value of Sewage Sludge Compost. Re-
port to U.S. EPA, N.J. DEP, C.C. MUA, pp. 115, New Jersey Experiment
Station, New Brunswick, N.J. undated.
38. Willson, G.B., J,F. Parr, and D.C. Casey. Criteria for Effective
Composting of Sewage Sludge in Aerated Piles and for Maximum Effici-
ency of Site Utilization. In: Pvoceedingo of the National Conference
on Deoign of Municipal Sludge Composting Facilities, Information
Transfer, Inc., Rockville, MD, p. 79-87, 1978.
39. Olver, W.M. Jr., Static Pile Composting of Municipal Sewage Sludge;
The Process as Conducted at Danger, Maine. U.S. Environmental Pro-
tection Agency, Washington, D.C. (Grant No. 803828), undated manu-
script.
40. Strom, P.P., M.L. Morris, and M.S. Finstein. Leaf Composting Through
Appropriate, Low-Level, Technology. Compost Science,, 21(6):44-48,
1980.
41. Kasper, V. Jr., and D.A. Derr. Sludge Composting and Utilization:
An Economic Analysis o£ the Cainden Sludge Composting Facility. Re-
port to U.S. EPA, N.J. DEP, C.C. MUA, pp. 342, New Jersey Experiment
Station, New Brunswick, 1981.
42, lacoboni, M.D., T.J. LeBrun, and J. Livingston. Windrow and Static
Pile Composting of Municipal Sewage Sludge. Report to EPA-MERL-ORD
Contract No. 14-12-150), pp. 124-, 1982,
43. Alcock, R., G.H. Nieswand, M.E. Singley, M.P. Bolan, and B.L. WMtson.
Sludge Composting and Utilization: Systems Analysis of the Cainden
Composting Operation. Report to U.S. EPA, N.J. DEP, C.C. MUA, pp.
128, New Jersey Experiment Station, New Brunswick, 1981.
44. Strom, P.P. The Thermophilic Bacterial Populations of Refuse Com-
posting as Affected by Temperature. Ph.D. Thesis, Rutgers University,
New Brunswick, N.J., 1978.
45. Strom, P.P. and M.S. Finstein. Thermophilic Bacterial Populations
of Solid Waste Composting. Abst. Annu, Meet. Amer. Sos. MioTobiol.
Q89, 1979.
46. American Society of Heating, Refrigeration, and Air Conditioning
Engineers, Inc.. Psychometric Chart Nos. 1 and 3, 1963.
47. Anonymous. Drying Corn at the County Elevator. Univ Illinois Co-
operative Extension Service Circular 1053, March 1972.
48. American Public Health Association. Standard Methods for the Ex-
amination of Water and Wastewater, 13th Edition, 1971.
49. Alexander, M.. Introduction to Soil Microbiology, 2nd Edition, John
Wiley § Sons, N.Y. 1977.
160
-------
50. Federal Register. Environmental Protection Agency. Part IX, p.
53438-53468, September 1979.
51. Psarianos, K.M., S.T. MacGr^gor, F.C. Miller, and M.S. Finstein.
Design of Composting Ventilation System for Uniform Air Distribu-
tion. BioCyele, 24(2):27-31, 1983. (Reprinted in Managing Sludge
By Compoeting, p. 281-287, The JG Press, Inc., Einmaus, PA. (1984),
under the title Ventilation System for Uniform Air Distribution.)
52. Crane Technical Paper #410: Flow of Fluids. Engineering Division,
Crane Co., Chicago, 1975.
161
-------
80
60
o
o
2: 40
UJ
H
80
60
o; 40
20
APPENDIX A-l
Temperature Observations for
Piles 7, 8, 9A and 9B
100 200 300
TIME IN HOURS
400
500
100
200 300
TIME IN HOURS
400
500
Figure 81. Pile 7, temperature at position 1 (upper plot) and
position 2 (lower plot).
162
-------
QO
60
o_
I 40
UJ
h-
20
0
80
60
I 4°
UJ
H-
20
0
100 200 300
TIME IN HOURS
100
400
500
200 300
TIME IN HOURS
400
500
Figure 82. Pile 1, temperature at position 3 (upper plot) and
position 4 (lower plot).
163
-------
00
60
u
20
100
200 300
TIME IN HOURS
400
500
80
60
40
20
100
20C 300
TIME IN HOURS
400
500
Figure 83. Pile 7, temperature at position 5 (upper plot) and
position 6 (lower plot).
164
-------
00
60
u
9
K. 40
2
UJ
t-
20
0
100 200 300
TIME IN HOURS
400
SOO
80
60
o
@
UJ
"T
T
100 200 300
TIME IN HOURS
SOO
Figure 84. Pile 7, temperature at position 7 (upper plot) and
position 8 (lower plot).
165
-------
80
60
u
fi,
I' 40
20
80
a; 40
UJ
20
0
100
JL
J
200 300
TIME IN HOURS
100 200 300
TIME IN HOURS
400
500
400
500
Figure 85. Pile 7, temperature at position 9 (upper plot) and
position 10 (lower plot.) .
166
-------
80
60
o
o
§• 40
UJ
i-
20
100
200 300
TIME IN HOURS
400
500
80
60
o
a: 40
a
UJ
I-
20
0
0
100 200 300
TIME IN HOU^S
400
300
Figure 86. Pile 7, temperature at position 11 (upper plot) and
position 12 (lower plot).
167
-------
80
GO
o
0
Si 40
20
0
100
200 300
TIME IN HOURS
400
500
100 200 300
TIME IN HOURS
400
500
Figure 87. Pile 7, temperature at position 13 (upper plot) and
position 14 (lower plot).
168
-------
80
60
o
UJ
K
40
20
100
200 300
TIME IN HOURS
400
500
80
60
40
20
0
0
100
200 300
TIME IN HOURS
400
500
Figure 88. Pile 7, temperature at position 15 (upper plot) and
position 16 (lower plot).
169
-------
BO
60
o
o
40
20
100 200 300
TIME IN HOURS
400
500
80
60
o: 40
S
UJ
20
100
200 300
TIME IN HOURS
400
500
Figure 89. Pile 8, temperature at position 1 (upper plot) and
position 2 (lower plot).
170
-------
80
60
o
o
20
00
60
I
UJ
20
100 200 300
TIME IN HOURS
400
500
100 200 300
TIME IN HOURS
400
500
Figure 90. Pile 8, temperature at position 3 (upper plot)
position 4 (lower plot).
and
171
-------
80
60
4°
20
100 200 300
TIME IN HOURS
400
500
80
60
40
UJ
20
100
200 300
TIME IN HOURS
400
500
Figure 91. Pile 8, temperature at position 5 (upper plot) and
position 6 (lower plot).
172
-------
80
60
u
-' 40
20
100 200 300
TIME IN HOURS
400
500
80
60
a: 40
20
100 200 300
TIME IN HOURS
400
500
Figure 92. Pile 8, temperature at position 7 (upper plot) and
position 8 (low^.r plot).
173
-------
o
o
80
60
40
20
iOO 200 300
TIME IN HOURS
400
500
80
60
o
1 40
20
0
100
200 300
TIME IN HOURS
400
300
Figure 93. Pile 8, temperature at position 9 (upper plot) and
position 10 (lower plot).
174
-------
00
60
UJ
20
0
100 200 300
TIME IN HOURS
400
500
80
60
CJ
40
UJ
20
100
200 300
TIME IN HOURS
40O
'600
Figure 94. Pile 8, temperature at position 11 (upper plot) and
position 12 (lower plot).
175
-------
80
60
o
0
UJ
20
100
200 300
TIME IN HOURS
400
500
80
60
u
a: 40
UJ
20
O
100
200 300
TIME IN HOURS
400
500
Figure 95. Pile 8, temperature at position 13 (upper plot) and
position 14 (lower plot).
176
-------
80
60
o
o
2! 40
UJ
H
100 200 300
TIME IN HOURS
400
500
80
60
40
20
100
200 300
TIME IN HOURS
400
SOO
Figure 96. Pile 8, temperature at position 15 (upper plot) and
position 16 (lower plot).
177
-------
60 -
60 ~
o
o
I 4°
20
—r
100 200 300 400
TIME IN HOURS
500
80 -
60
a; 40
HJ
H
20
100
200 300
TIME IN HOURS
400
500
Figure 97. Pile 8' temperature at position 17 (upper plot) and
position 18 (lower plot).
178
-------
80 ~
60 ~
o
o
a: 40
UJ
(-
20
100
200 300
TIME IN HOURS
400
500
60
60
o
40
20
100
200 300
TIME IN HOURS
400
500
Figure 98. Pile S, temperature at position 19 (upper plot) and
position 20 (lower plot).
179
-------
80
60
a
g_
a 40
s
UJ
20
0
100
200 300
TIME IN HOURS
400
500
80
60
o: 40
S
UJ
20
100
200 300
TIME IN HOURS
400
500
Figure 99. Pile 8, temperature at position 21 (upper plot) and
position 22 (lower plot).
180
-------
80
60
40
UJ
20
100
200 300
TIME IN HOURS
400
500
80
60
o
: 40
20
0
100
200 300
TIME IN HOURS
400
500
Figure 100. Pile 8, temperature at position 23 (upper plot) and
position 24 (lower plot).
181
-------
80
60
o
0
a: 40
20
0 L.
0
100
200 300
TIME IN HOURS
400
500
80
60
a: 40
20
100
200 300
TIME IN HOURS
400
500
Figure 101. Pile 8, temperature at position 25 (upper plot) and
position 26 (lower plot).
182
-------
UJ
80 h
60
40
20
80
60
CJ
a: 4O
UJ
(-
20
0
100
100
200 300
TIME IN HOURS
400
500
200 300
TIME IN HOURS
400
SOO
Figure 102. Pile 8, temperature at position 27 (upper plot) and
position 28 (lower plot).
183
-------
80
60
o
o
4C
ui
100 20O 300
TIME IN HOURS
400
500
80
60
o
o
40
20
IOO 20O 300 400
TIME IN HOURS
500
Figure 103. Pile 8, temperature at position 29 (upper plot) and
position 30 (lower plot).
184
-------
80
60
o
UJ
H
20
100
200 300
TIME IN HOURS
400
500
Figure 104. Pile 8, temperature at position 31
185
-------
60
o
o
4°
20
100 200 300
TIME IN HOURS
400
500
SO
CJ
40
20
0
100 200 300
TIME IN HOURS
400
500
Figure 105. Pile 9A, temperature at position 1 (upper plot) and
position 2 (lower plot).
186
-------
o
UJ
H
40
20
0
80
SO
K 40
s
t-
20
100 200 300
TIME IN HOURS
100
400
500
200 300 400 300
TIME IN HOURS
Figure 106. Pile 9A, temperature at position 3 (upper plot) and
position 4 (lower plot).
187
-------
so
60
o
0
UJ
(-
40
20
100 200 300
TIME IN HOURS
400
500
90
60
CJ
S»-
a: 40
UJ
H
80
100 200 300
TIME IN HOURS
400
300
Figure 107, Pile 9A, temperature at position 5 (upper plot) and
position 6 (lower plot).
188
-------
80
60
CJ
40
20
100 200 300
TIME IN HOURS
400
500
80
60
3:
UJ
t-
40
20
0
100 200 300
TIME IN HOURS
400
300
Figure 108. Pile 9A, temperature at position 7 (upper plot) and
position 8 (lower plot).
189
-------
80
60 ~
o
o
o: 40
UJ
I-
20
100 200 300
TIME IN HOURS
400
500
80
60
CJ
a 40
UJ
20
100
200 300
TIME IN HOURS
400
Figure 109. Pile 9A, temperature at position 9 (upper plot) and
position 10 (lower plot).
190
-------
80
o
0
I 40
UJ
20
0
100
200 300
TIME IN HOURS
400
500
60
UJ
i-
20
100
?,00 300
TIME IN HOURS
400
SOO
Figure 110. Pile 9A, temperature at position 11 (upper plot) and
position 12 (lower plot).
191
-------
60
L)J
h-
20
100
200 300 400
TIME IN HOURS
500
80
60
40
UJ
20
0
100
200 300
TIME IN HOURS
400
500
Figure 111. Pile 9A' temperature at position 13 (upper plot.) and
position 14 (lower plot).
192
-------
80
60
o
0
4°
20
100 200 300
TIME IN HOURS
400
500
80
60
o
o
40
UJ
20
100
200 300
TIME IN HOURS
400
500
Figure 112. Pile 9A, temperatnie at position 15 (upper plot) and
position 16 (lower plot).
193
-------
80
o
o
9=" 4r
UJ
t-
20
0
100 200 300
TIME IN HOURS
400
500
00
o
o
40
UJ
t-
20
0
100
200 300
TIME IN HOUR3
400
800
Figure 113. Pile 9A, temperature at position 17 (upper plot) and
position 18 (lower plot).
194
-------
80 -
o
100 200 300
TIME IN HOURS
400
500
Figure 114. Pile 9A, temperature at position 19
195
-------
80
60
o
0
£ 40
UJ
20
" too"
200 300
TIME IN .UOURS
400
800
€0
'G
o
o: 40
UJ
i_
20
100
200 300
TIME IN HOURS
400
900
Figure 115. Pile 9B, temperature at position I (upper plot) and
position 2 (lower plot).
196
-------
80
60
40
UJ
20
100
200 300
TIME IN HOURS
400
500
60
e>
a: 40
I—
20
100
200 300
TIME IN HOURS
400
300
Figure 116. Pile 9B, temperature at position 3 (upper plot) and
position 4 (lower plot).
197
-------
80
40
20
100 200 300
TIME IN HOURS
400
500
60
SL,
g;
e£,
UJ
20
100
200 300
TIME IN HOURS
400
500
Figure 117. Pile 9B, temperature at position 5 (upper plot) and
position 6 (lower plot).
198
-------
60
o
I 40
ui
t-
20
80
60
40
UJ
h-
20
100 200 300
TIME 'N HOURS
400
500
100
200 300
TIME IN HOURS
400
500
Figure 118. Pile 9B, temperature at position 7 (upper plot) and
position 8 (lower plot).
199
-------
80
60
o
UJ
h-
40
20
100 200 300
TIME IN HOURS
400
500
60 -
& 40
UJ
h-
20
100
200 300
TIME IN HOURS
400
300
Figure 119. Pile 9B, temperature at position 9 (upper plot) and
position 10 (lower plot).
200
-------
80
60
u
I 40
LU
t-
100 200 300
TIME IN HOURS
400
500
80
SO
I 4°
UJ
20
0
100
200 00
TIME IN HOURS
400
500
Figure 120. Pile 9B, temperature at position 11 (upper plot) and
position 12 (lower plot).
201
-------
80
60
o
Si 40
UJ
1-
20
0
80
60
0__
o: 40
tu
t~
20
100 200 300
TIME IN HOURS
400
500
100
200 300
TIME IN HOURS
400
500
Figure 121. Pile 9B, temperature at position 13 (upper plot) and
position 14 (lower plot).
202
-------
80
60
%' 40
20
100 200 300 400
TIME IN HOURS
500
Figure 122. Pile 9B, temperature at position 15
203
-------
n>
in
-~4
05
O
X
X
SQ
(D
3
O
CT
C/l
(D
0)
rt
•X)
m
o
i—i
x
>
o
t-o
[00
200 300
TIME IW HOURS
400
500
Figure 123. Pile 7
position 6.
concentration cf 07. The gas sampling probe was adjacent to
-------
o
Cn
20
O
>
LJ IS
O
<£
111
0.
12
HE
I-
O
O
O
o
at
100
2OO 300
TIME IN HOURS
4OO
500
Figure 124, Pile 1, concentrations of 0_. The gas sampling probe was adjacent to
position 14.
-------
100
200 300
TIME IN HOURS
400
500
Figure 125. Pile 8, concentrations of 0-. The gas sampling probe was adjacent to
position 11.
-------
100
200 300
TIME IN HOURS
400
500
Figure 126. Pile
position 26.
concentration of Op. The gas sampling probe was adjacent to
-------
ro
a
oo
200 300
TIME IN HOURS
400
5OO
Figure 127. Pile
position 28.
8, concentration of 0_. The gas sampling probe was adjacent to
-------
APPENDIX A-3
MOISTURE CONTENT IN THE "WHOLE SAMPLE" AND THE "NON-WOODCHIP
FRACTION"
When the material being composted was a mixture cf sewage sludge
and woodehips, two types o£ samples were usually processed for the
determination of moisture content. These were the whole sample 'and the
non-woodchip fraction. The non-v/oodchip fraction was prepared by re-
moving the woodehips by hand. Except where not otherwise noted, the
data reported in the body of the report refer to the whole sample.
Both sets of data are reported in Appendix A-3, for comparative pur-
poses.
At time-zero the moisture content of the non-v/oodchip fraction
was higher than that of the whole sample (Figures 128, 129, 130). The
diffe^-iiee between the moisture content of the two kinds of samples at
time-zero was: pile 8, 201; pile 9A, S%; pile 9B, 8%. The difference
disappeared in the piles which dried extensively (Rutgers strategy),
but not in the pile which experienced only slight drying (Beltsville
process).
209
-------
o
IjO
•
-------
WHOLE SAMPLE
NON-WOODCHIP FRACTION
100
200 300
TIME IN HOURS
400
500
Figure 129. Pile 9A, moisture r-ontent.
-------
t-o
I—I
ISJ
70
ui
K.
t-
-------
APPENDIX A-4
REPRESENTATIVENESS OF PILE 9B
Since the Beltsville Process was represented herein by only one
pile (9B), its representativeness was evaluated in reference to com-
parable data reported by other investigators. Also, the grouped Belts°
vine data was compared to grouped Rutgers data. The analysis is
summarized in TABLE 21=
The first entry for temperature (>70°C) is based on terminal
(day 21) observations (38), or on mean values (9). One of the reports
(39) does not provide usable temperature data. The second entry for
the Beltsville process (80°C) js based on the extensively monitored
pile 9B, in which the bulk of the material peaked at 78°C to 82°C.
The entry for the Rutgers process is based on the piles described in
Section 3, in which the bulk of the material generally did not exceed
60°C.
The first entry concerning ventilation (0, level, 5-15%) is
based on one pile at the end of the 21 day period. The second entry
OlOt) is an "average" value, the precise meaning of which is diffi-
cult to evaluate. The third entry (12-214) is based on pile 9B over a
21 day period, in which the gas sampling probe -was centrally positioned.
The entry for the Rutgers process (16-21%) is from pile 9A at a compar-
able position, and represents a period of 21 days. On average, the
Beltsville piles v/ere provided with 134 cfh/v/et ton during the standard
21 day process period. For the Rutgers piles the blowers provided, on
average, 1232 cfh/wet ton during the 12 day period of the feedback
control.
The first entry concerning moisture content (7% decrease) is
based on samples taken from four locations (six replicates per location)
of one pile on the terminal day of standard processing (38). Reference
9 does not include suitable moisture content data. The second entry
(14$ decrease) represents 114 piles on the terminal day (processing was
usually for 21 days). Th .- third entry (4% decrease) is the final value
of the series representing pile 9B. The fourth entry (.341 decrease)
represents the mean final value of the indicated Rutgers piles.
213
-------
TABLE 21. COMPARISON OF FIELD DATA FOR RUTGERS AND BELTSVILLE PROCESSES.
Peak temperature Ventilation Decrease i
representing the requirement moisture
bulk of the pile to maintain an content
Process
Beltsville
Beltsville
Beltsville
Rutgers
( C)
>70
_.
80
60
oxygenated
02 level
(1)
5-15
10
12-21
16-21
condition (£)
cfh/ton
60-133 7
212* 14
132 4
12323 34
.n
Process
time Reference or
(days) pile
21 9, 38-39
21 39
21 Pile 9B
12 Files 4AiJ, 1]
8, 9A
Woodchips included in the sample, except that for pile 4A the saiaple was sieved before
the determination.
Mean delivery per ton (wet wt) of the initial iludge-woodchip mixture. To convert from
cffh to m^/Sj, multiply x 7,87 x 10" ; to convert from ton to tonne (metric), multiply
The published value is 123 cfSi/yd . This was converted to the common means of expression
by assuming that 1.74 yd3 weighs 1.0 ton.
Peak demand ^ 4800 cf Hi/ton
See Section 4.
-------
APPENDIX A-S
FORCES DRIVING VAPORIZATION: METABOLICALLY GENERATED HEAT,
AND UNSATURATION OF INLET AIR
The experimentally derived value of 4.5% was calculated as the
H-,0 removed per unit volume air from a metabolically inert pile o£
cored compost x lOO/HjO removed per unit volume air from an actively
composting pile, (See Section 8 for observations on a pile o£ well-
cured compost -- pile 12.) The calculation leading to the theoretical
values (1.8% to 5.0%) assumes that ambient inlet air at 20°C and 30%
relative humidity (RH) is introduced into two physically identical
piles: one an actively composting pile; the other a metabolically
inert pile. Upon exiting from the composting pile the air is at 60 C
and 100% RH, hence water removal amounts to 0.14S kg/kg dry air (46).
For the metabolically inert pile two extreme cases are developed.
The first assumes that the air experiences the maximum theoretical
temperature decrease with passage through the pile. Upon exiting
the air is at 13.3°C and 100% RH, hence water removal is 0.0026 kg/kg
dry air. The second case assumes no temperature decrease, giving
exit air at 20°C and 100% RH and water removal of 0.0072 kg/kg'dry
air. Thus, in this calculation non-biological air drying caused by
unsaturation of the air accounts for only 1.8% to 5.0$ of the water
removal from the composting pile. The remainder is attributable to
uiierobial heat generation.
21S
-------
APPENDIX A-6
RELATIVE SENSITIVITY OF THE MOISTURE CONTENT AND VOLATILE
SOLIDS TESTS
Consider the relationship between the decrements of organic
matter and water on a mass basis. Assuming a release of 6000 cal
from the micTobial oxidation of 1 g organic matter to CO* and H^O,
and with the heat of vaporization of water at 20°C (the assumed
temperature of the inlet air and starting material) equal to 586
eal/g, 10.2 g water could be vaporized per g volatile matter de-
composed. A small part of the water loss is made up through metabolic
water production (ca. , 0.8 g) . Also, &A (at 80°C exit temperature) to
14% (at SO°C) of the heat removal is through dry air convection, and
another 21 (at SO C) to 41 (at 80°C) through raising the temperature
of the vaporized water. Hence, the mass of water rsmoved provides
an indication of organic matter cecomposition that is 7.8 (at 50°C)
to 8.6 (at 80 C)-fold more sensitive than the mass of volatile solids
decomposed.
The mass of water removed is rarely known, however, whereas the
$ moisture content is easily determined. Under realistic conditions,
the ratio change in I moisture/change in % volatile solids exceeds -
unity at the outset of composting, and widens progressively due tt> the
different relative changes in the regaining masses of water and vola-
tile solids. The widening of the ratio is mainly a function of the
initial moisture content and the initial volatile solids content, hence
no further generalization regarding its numerical value is possible.
In the ease of an initial moisture content of 75% (wet weight basis)
and volatile solids content of 7S% (dry weight basis), as the moisture
content traverses the 30% level the cumulative ratio is 3.S. In
association with a decrease in moisture content from 31% to 30%, the
ratio is 9.8.
216
-------
APPENDIX B
ADVANTAGES OF FUEL PRODUCTION THROUGH COMPOSTING VS.
DIRECT COMBUSTION SEWAGE SLUDGE CAKE
The residue of the composting process is metabolically inert,
and therefore relatively easy to handle, stockpile., and transport.
The process residue is dry and granular, or can be pulverized, to
granulate it, making it relatively easy to feed into the combus-
tion chamber.
Separation of the drying and combustion functions improves over-
all system reliability, in that a breakdown of the combustion
system does not interfere with the sludge treatment (ifuel prepara-
tion) .
Thejnn£dyjriami c_A_dv an ta_ ge£
Separation of the drying and combustion functions decreases
the amount o£ energy spent in the vaporization o£ water. This
stems from a comparison o£ biological drying (composting) at 140°F
versus direct combustion at 1SOO°F. Note that for every pound o£
water vaporized at 140°F (via biological drying) , approximately
680 BTU's are saved compared to combustion at 1500°F.* This reflects
the approximately O.S BTU's/lb/°F required to raise the temperature
of water vapor over this range.
A further advantage, not as easily quantitated in the general
case, is that less total air is required for the combustion. For
every pound of air not required, approximately 370 BTU's are saved,
reflecting the 0,26 BTU/lb/°F specific heat o£ air over this range.
Energy must be expended in biological drying (composting) , of
course and this decreases somewhat the thermodynamic advantage,
No exact accounting is readily accessible, but it 'is evident that
biological drying is advantageous compared to air drying (see
TABLE 12.) _ ______ —
'~"~ 37th Edition, 1963. The Babcock §
* New YorTc
217
-------
a:
UJ
h-
80
60
40
20
0
60
o
e
Q-- 40
UJ
I-
20
APPENDIX C
Temperature Observations
for Piles 11A, 11B, and 11C
IOO
200 300
TIME IN HOURS
400
500
100
200 300
TIME IN HOURS
400
500
Figure 131. Pile 11A, temperature at position 1 (upper plot)
and position 2 (lower plot).
218
-------
eo
60
o
o
UJ
H
20
0
100 200 300
TIME IN HOURS
400
500
80
o
SL,
& 40
UJ
0
100 200 300
IN HOURS
400
900
Figure 132. Pile HA, temperature at position 3 tupper plot)
and position 4 (lower plot).
219
-------
80
60
o
40
UJ
H
20
0
100 200 300
TIME IN HOURS
400
300
80
60
u
4°
UJ
t-
20
0
100
200 300
TIME IN HOURS
400
SOO
Figure 133. Pile 11A, temperature at position 5 (upper plot)
and position 6 (lower plot).
220
-------
80
60
o
o
UJ
20
100 200 300
TIME IN HOURS
400
500
80
60
o
0
20
100
200 300
TIME IN HOURS
400
500
Figure 134. Pile 11A, temperature at position 7 (upper plot)
and position 8 (lower plot).
221
-------
80
60
40
UJ
t-
20
0
100
200 300
TIME IN HOURS
400
500
80
60
o
o
40
UJ
h-
20
100
200 300 400
TIME IN HOURS
300
Figure 135. Pile 11A, temperature at position 9 (upper plot)
and position 10 (lower plot).
222
-------
80
60
o
o
•i™.'
I 4°
UJ
20
100 200 300
TIME IN HOURS
400
500
80
60
o
1 40
UJ
20
100
__J=
200 300
TIME IN HOURS
400
500
Figure 136. Pile 11 A, temperature at position 11 (upper plot)
and position 12 (lower plot).
223
-------
100
200 300
TIME IN HOURS
400
500
200 300
TIME >N HOURS
400
300
Figure 137. Pile 11B, temperature at position 1 (upper plot)
and position 2 (lower plot) „
224
-------
80
o
o
UJ
40
20
0 100 200 300
TIME IN HOURS
400
500
100
200 300
TIME IN HOURS
400
500
Figure 138. Pile 11B, temperature at position 3 (upper plot)
and position 4 (lower plot).
225
-------
6O
o
o
40
UJ
j-
100
200 300
TIME IN HOURS
4OO
SOO
80
a.' 40
LU
H
20
IOO
200 300
TIME IN HOURS
400
SOO
Figure 139. Pile 11B, temperature at position 5 (upper plot)
and position 6 (lower plot).
226
-------
60
40
UJ
s-
20
100 200 300
TIME IN HOURS
400
500
80
SO
u
tu
100 200 300
TIME IN HOURS
400
BOO
Figure 140. Pile 11B, temperature at position 7 (upper plot)
and position 8 (lower plot).
227
-------
100
200 300
TIME IN HOURS
500
100
200
TIME IN HOURS
400
300
Figure 141. Pile 11B, temperature at position 9 (upper plot)
and position 10 (lower plot).
228
-------
60
o
e
40
UJ
H
20
100 200 300
TIME IN HOURS
400
500
80
60
o
SL,
40
UJ
20
100 200 300
TIME IN HOURS
400
soo
Figure 142. Pile 11B, temperature at position 11 (upper plot)
and position 12 (lower plot).
229
-------
80
o
o
UJ
H
40
20
100
200 300
TIME IN HOURS
400
500
80
o
,
40
t-
20
0
100 200 300
TIME IN HOURS
400
500
Figure 143. Pile 11C, temperature at position 1 (upper plot)
and position 2 (lower plot).
230
-------
80
SO
o
o^
S! 40
a
LU
t-
20
100 200 300 400
TIME IN HOURS
500
80
60
a: 40
UJ
20
100
200 300
TIME IN HOURS
400
500
Figure 144. Pile 11C, temperature at position 3 (upper plot)
and position 4 (lower plot)„
231
-------
80
60
40
UJ
I-
100 ZOO 300
TIME IN HOURS
400
500
a:
20
0
100
200 300
TIME IN HOURS
400
iOO
Figure 145. Pile 11C, temperature at position 5 (upper plot)
and position 6 (lower plot).
232
-------
o
80
60
40
20
0
100
200 300
TIME IN HOURS
400
500
60
cj
o
a: 40
UJ
t-
20
100
200 300
TIME IN HOURS
400
300
Figure 146. Pile 11C, temperature at position 7 (upper plot)
and position 8 (lower plot).
233
-------
90
60
o
o
40
20
100 200 300
TIME IN HOURS
400
500
80
60
u
o_
a:
§
UJ
h-
20
100 200 300
TIME IN HOURS
400
300
Figure 147. Pile 11C, temperature at position 9 (upper plot)
and position 10 (lower plot).
!34
-------
60
o
o
40
20
0
100
200 300
TIME IN HOURS
400
300
80
60
o
o
40
I-
20
0
100
200 300
TIME IN HOURS
400
500
Figure 148. Pile lie, temperature at position 11 (upper plot)
and position 12 (lower plot).
235
-------
APPENDIX D-l
AIR NE^JED TO REMOVE HEAT AND SUPPLY OXYGEN
A wide range of organic compounds shows little variation in the
energy released per mass of 0« used for complete oxidation to carbon
dioxide and water. The mean value is approximately 14,000 kJ released/
kg 0, utilized. Using equation (iv) and the values of h t from
TABLE 1 (for approach R) : Q^ *• 362 kJ/kg dry air; thus Ttutakes
m™
14000 = 38.7 kg dry air to remove 14,000 kJ o£ released energy. Since
there is 0.232 kg 0,,/kg dry air, it takes 4.31 kg of dry air to re-
plenish 1 kg of 02. The ratio is 38.7 - 8.98.
APPENDIX D-2
UNSUCCESSFUL ATTEMPT BY THE BELTSVILLE GROUP TO IMPROVE DRYING, IN ISO-
LATION FROM CONSIDERATIONS OF PROCESS DYNAMICS
In an attempt to improve drying, the "active stage" of the Belts-
ville Process was extended from the standard 21 day period to 28 days,
and on day 14 ventilation was switched from the vacuum-induced to the
forced pressure direction and increased by approximately 4-fold (10).
The modified process behaved as follows. During the first 14 days the
average peak interior temperature was 70°C. The increase in ventilation
on day 14 marked the onset of a cooling trend, such that on day 21 the
average temperature was 42 C. (In conventional Beltsville operation
the temperature remains at peak, or near pejak, values throughout the
21 day period,) This modification yielded only a modest improvement in
v/ater removal. The initial moisture content of 63.6% decreased to
52.1% in the 28 day period, compared to a terminal value of 56.3%
obtained in the usual 21 day period through conventional operation.
This behavior is predictable based on expression iv, as con-
strained by the interaction between heat generation and temperature.
As is characteristic of the Beltsville Process, a sma31 jn resulted in
a temperature ascent to biologically inhibitive levels. Once peak
temperatures were reached (-vday S) h £ was large but the resultant
0 was small, as evidsnced by the small moisture content decrease. In-
creasing m on day 14 cooled the pile because the rate of biological
activity ~t=heat generation) could not sustain the peak temperature
against the increased rate of heat removal. The decreasing outlet
temperatures were accompanied by decreasing values of h t- As pre=
dieted the resultant .Q* was small, judged by-the small Moisture
content decrement.
236
-------
70
50
o
o
I 3°
UJ
-10
70
50
I 3°
UJ
I-
10
APPENDIX E
Temperature Observations
Pile 12
100 200 300
TIME IN HOURS
400
500
_L
100
200 300
TIME IN HOURS
400
500
Figure 149. Pile 12, temperature at position 1 (upper plot)
and position 2 (lower plot).
237
-------
70
50
O
I 30
UJ
t-
10
-10
70
50
30
10
-10
100
100
200 300
TIME IN HOURS
200 300
TIME IN HOURS
400
400
500
500
Figure ISO. Pile 12, temperature at position 3 (upper plot)
and position 4 (lower plot).
238
-------
to
80
CJ
I
UJ
10
-10
100 200 300
TIME IN HOURS
400
500
70
50
o
o
LU
h-
10
-10
100
200 300
TIME IN HOURS
400
SOO
Figure 151. Pile 12, temperature at position 5 (upper plot)
and position 6 (lower plot).
239
-------
70
50
30
10
-10
100
200 300
TIME IN HOURS
400
500
Figure 152.
Ambient temperature during the experimental period
for pile 12.
240
-------
APPENDIX F
Observations on Blower Operation,
Temperature, 02, C0?s pH, and
Moisture Content for Piles 6A, 6B,
and 6C
Figure 153.Piles 6A, 6B, and 6C, cross-sectional representation: textured area, wood-
chip cover and base; clear area, sludge-woodchip mixture; circle, perforated flex-
hose. The numbers indicate monitoring and control positions: thermocouples,
positions 1 through 15; 7as sampling probes, adjacent to position 13 (also adjacent
to position 6 in pile 6A); control thermistor, between positions 1 and 6. The
blower was operated in the forced-pressure mode.
-------
100-ff
o
to)
2
LJ
o
-------
100-f
80-
o
i-SO-ji
o
Q
£C
o
_!
IS.
01.
JOO
200
3OO
HOURS
400
Figure 155. Pile 6B, blower operation. The baseline represents operation as scheduled
by rimer, and the area above the baseline represents operation through the temp-
erature feedback control system. The blower was operated in the forced pressure
mode.
-------
\oo-r
80-!
2
O
KSO-
a
a.
O
t- 40-
«a
20-
O
500
200 3(
TIME IN HOURS
Figure 156. Pile 6C, blower operation. The baseline represents operation as scheduled
by timer, and the area above the baseline represents operation through the temp-
erature feedback control system. The blower was operated in the forced pressure
mode.
-------
80
60
u
o
tu
h-
20
0
100 200 300
TIME IN HOURS
400
500
80
60
O
o^
a: 40
h-
20
100
200 300
TIME IN HOURS
400
Figure 157. Pile 6A, temperature at position 1 (upper plot) and
position 2 (lower plot)„
245
-------
00
60
u
o
40
UJ
20
0
100 200 300 400
TIME IN HOURS
500
60
u
0
a: 40
UJ
H
20
0
100
200 500
TIME IN HOURS
400
SOO
Figure 158, Pile 6A, temperature at position 3 (upper plot) and
position 4 (lower plot).
246
-------
o
o
UJ
60
40
20
0
0
100
200 300
TIME IN HOURS
400
500
80
60
o
o
Q.' 40
UJ
t-
20
0
0
100
200 300
TIME IN HOURS
400
500
Figure 159. Pile 6A, temperature at position 5 (upper plot) and
position 6 (lower plot).
247
-------
80
40
UJ
H-
20
100 200 300
TIME IN HOURS
400
500
80
60
Q-' 40
20
100
200 300
TIME IN HOURS
400
500
Figure 160. Pile 6A, temperature at position 7 (upper plot) and
position 8 (lower plot).
248
-------
QO
60
o
o
40
UJ
t-
20
0
80
60
JL
a: 40
in
20
100
100
200 300
TIME IN HOURS
400
500
200 300
TIME IN HOURS
400
300
Figure 161. pile 6A, temperature at position 9 (upper plot) and
position 12 (lower plot).
249
-------
80
80
o
o
40
UJ
t-
20
100 200 300
TIME IN HOURS
400
80
60
o
o
Q-' 40
UJ
t-
20
.00 200 300
TIME IN HOURS
400
500
Figure 162. Pile 6A, temperature at position 13 (upper plot)
and position 14 (lower plot).
250
-------
80
60
UJ
t-
20
0
100
200 300
TIME IN HOURS
400
500
Figure 163. Pile 6A, temperature at position 15,
251
-------
o
o
UJ
t-
80
60
40
20
100 200 300
TIME IN HOURS
400
500
80
60 -
o
p^
a: 40
S
UJ
20
0
100
200 300
TIME IN HOURS
400
500
Figure 164. Pile fB, temperature at position 1 (upper plot)
and position 2 (lower plot).
152
-------
60
I 40
UJ
20
100
200 300
TIME IN HOURS
400
500
80
60
u
o
Q,'
s
UJ
20
0
100
200 300
TIME IN HOURS
400
500
Figure 365. Pile 6B, temperature at position 5 (upper plot)
and position 7 (lower plot).
253
-------
80
60
o
o
UJ
I—
20
100 200 300
TIME IN HOURS
400
500
80
60
cj
o__
o; 40
UJ
(-
20
100
200 300
TIME IN HOURS
400
500
Figure 166. Pile 6B, temperature at position 9 (upper plot)
and position 12 (lower plot).
254
-------
u
o
UJ
H
80
SO
40
20
100 200 300
TIME IN HOURS
400
500
Figure 167. Pile 6B, temperature at position 13,
255
-------
80
60
40
20
o t=~
100 200 300
TIME IN HOURS
400
500
o
8-
I 4°
20
100
200 300
TIME IN HOURS
400
500
Figure 168. Pile 6C, temperature at position 2 (upper plot)
and position 5 (lower plot).
256
-------
00
60
o
0
40
10
100 200 300 400
TIME IN HOURS
300
80
60
UJ
t-
40
20
100
200 300
TIME IN HOURS
400
500
Figure 169. Pile 6C, temperature at position 7 (upper plot)
and position 9 (lower plot).
257
-------
80
60
40
UJ
t-
20
100 200
TIME IN HOURS
400
100
200 300
TIME IN HOURS
400
500
Figure 170. Pile 6C, temperature at position 12 (lower plot)
and position 13 (upper plot).
258
-------
100
200 300
TIME IN HOURS
400
500
Figure 171. Pile 6A, concentrations of 02 (upper plot) and C02
(lower plot).
position 6.
The gas sampling probe was adjacent to
259
-------
100
200 300
TIME IN HOURS
400
500
Figure 172. Pile 6A, concentrations of CU (upper plot) and
CO 2 (lower plot).
position 13.
The gas sampling probe was adjacent to
260
-------
20
UJ
3.
:D
_j
§16
z
UJ
UJ
CL
I-
-------
20
§16
UJ
o
UJ
Q.
I-
Z
UJ
o
o
O
K
UJ
100
200 300
TIME IN HOURS
400
500
Figure 174. Pile 6C, concentrations of 02 (upper plot) and C02
(lower plot).
position 13.
The gas sampling probe was adjacent to
262
-------
PILE 6A —
PILE 68
PILE SC —
o....
••©
100
200 300
TIME IW HOURS
400
500
Figure 175. Piles 6Af 63, and 6C, pH. The samples were from interior locations.
-------
70
SO
I-
OT
o
50
o
£E
PILE SA
PILE SB
PILE SC
30
20 U
100
200 300
TIME IN HOURS
soo
Figure 376. Piles 6A, 6B, and 6C, moisture content. The samples were from
interior locations.
-------
19.5 29.6 --- 64.3 62.9
27.2 31.6 32.7 50.4 (
-4.7 -1.8 +().()
-0.8 +0.2 + 0.5 + 0.2
1.1 +3.1
10.3 +18.0 +17.6 +53.8 (
Figure 177. Pile 6B, special terminal test of air penetration. Temperature { C)
immediately prior to the introduction of heated air, left hand cross-section;
temperature 30 min after start of introduction of heated air, middle cross-
section; temperature differential, right hand cross-section.
-------
80
a: 40 -
UJ
H-
20 ~
APPENDIX G
Temperature Observations
Pile 13
200 300
TIME IK) HOURS
200 300
TIME IN HOURS
L
400
eoo
Figure 178. Pile 13, temperature at position 1 (upper plot]
and position 2 (lower plot).
266
-------
80
SO
o
o
UJ
(-
40
2C
100
200 300
TIME IN HOURS
400
500
100
200 300
TIME IN HOURS
400
500
Figure 179. Pile 13, temperature at position 3 (upper plot)
and position 4 (lower plot).
-------
80
60
a:
UJ
H-
40
20
100
200 300
TIME IN HOURS
400
500
80
60
o"
o__
a: 40
UJ
(-
20
0
100 200 300
TIME IN HOURS
400
500
Figure 180. Pile 13, temperature at position 5 (upper plot)
and position 6 (lower plot)„
268
-------
100
200 300
TIME IN HOURS
400
500
0 *•=
100
200
TIME IN HOURS
400
500
Figure 181. Pile 13, temperature at position 7 (upper plot)
and position 8 (lower plot;.
269
-------
60
| 40
UJ
20
80
e_^
1 4°
UJ
20
100
200
TIME IN HOURS
400
100 200 300
TIME IN HOURS
400
500
500
Figure 182. Pile 13, temperature at position 9 (upper plot)
and position 10 (lower plot).
270
-------
80 ~
60
o
e
a:
4O
20
100 200 300
TIME IN HOURS
400
500
8O
60
o
20
100
2OQ 300
TIME IN HOURS
400
500
Figure 183. Pile 13, temperature at position 11 (upper plot)
and position 12 (lower plot).
1.71
-------
80 -
o
o
; 40 -
0 L
80
a: 40 -
S
UJ
20 L.
0
100
200 300
TIME IN HOURS
200 300
TIME IN HOURS
400
500
300
Figure 184. Pile 13, temperature at position 13 (upper plot)
and position 14 (lower plot).
272
-------
§0
u
o
a:
LU
{=
100
200 300
TIME IN HOURS
400
500
80
60
o
o
a, 40
20
100
200 300
TIME IN HOURS
400
Figure 185. Pile 13, temperature at position 15 (upper plot)
and position 16 (lower plot).
273
-------
80
60
2_
a:
UJ
t-
20
IOO 200 300
TIME IN HOURS
400
500
80
60
20
IOO
200 300
TIME IN HOURS
400
500
Figure 186. Pile 13, temperature at position 17 (upper plot)
and position 18 (lower plot).
274
-------
80
60 -
40
20
100 200 300
TIME IN HOURS
400 500
Figure 187. Pile 13, temperature at position 19,
275
------- |