United States
Environmental Protection
Agency
Municipal Environmental Research *•
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-82-044 August 1982
Project Summary
Evaluation and
Documentation of Mechanical
Reliability of Conventional
Wastewater Treatment Plant
Components
David W. Shultz and Van B. Parr
The study summarized here was
initiated to determine the in-service
reliability, maintainability, and availa-
bility of selected critical wastewater
treatment plant components. Compo-
nents were considered critical if their
failure caused an immediate impact
on effluent quality.
Eight treatment plant components
were selected for study: pumps,
power transmissions, motors, com-
pressors, diffusers, valves, controls,
and conveyors. These components
were from four types of wastewater
treatment plants: air activated sludge,
oxygen activated sludge, trickling fil-
ter, and rotating biological contactor
(RBC). Operation and maintenance
data obtained from nine operating
plants were used to estimate failure
rates, mean time between failures
(MTBF), mean time to repair (MTTR),
and availability for each component.
These performance statistics are by
component type, size range, and
application. Performance statistics
were calculated from operating data
for 119 pumps, 249 power transmis-
sions, 285 motors, 17 compressors,
13 valves, and 269 controls.
Design engineers and plant opera-
tors can use these calculated perfor-
mance statistics to assist in selecting
new equipment. The development of a
performance data base and subse-
quent proper use of the data should
improve treatment plant perfor-
mance. These data allow the perfor-
mance expected from certain
equipment in certain applications to
be estimated.
This Project Summary was devel-
oped by EPA's Municipal Environ-
mental Research Laboratory,
Cincinnati. OH. to announce key find-
ings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
Approximately 21,000 publicly
owned municipal treatment plants
(POTW's) operate in the United States
at the present time, with 1,000 to 1,200
new plants being constructed each
year Since 1957, federal grants
exceeding $20 billion have been
awarded to help state and local govern-
ments construct these treatment
plants. Once constructed, local govern-
ments become responsible for plant
operation in accordance with effluent
discharge permit requirements estab-
lished by the National Pollutant Dis-
charge Elimination System (NPDES).
-------
Many POTW's are apparently not
meeting NPDES permit requirements.
For many years, widespread operation
and maintenance problems with
POTW's have resulted in inefficient
plant operations. A 1975 EPA analysis
of 954 POTW inspections showed 386
plants had sufficient design and opera-
tional performance data to determine
whether the plant was meeting design
criteria for biochemical oxygen demand
(BOD) removal. Forty percent of the 386
plants failed to meet design BOD re-
moval. Forty-nine percent of 305 of
these plants were operating below
design criteria for suspended solids
removal. Other studies have indicated
significant problems of noncompliance
by POTW's with NPDES permits.
In recognition of these operational
and maintenance problems, EPA
initiated a national research program
dealing with performance and reliability
of POTW's. A significant part of the
effort involves determining the reliabil-
ity of various mechanical components
used at POTW's.
Technical Approach
Criticality Analysis
To determine the in-service reliability
of critical mechanical components,
eight components (listed above) were
identified as critical. After a criticality
analysis was conducted to determine
the component-application combina-
tions to be included in the study, a fail-
ure mode, effects, and criticality
analysis was performed for each of the
four types of wastewater treatment
plants. In performing the analysis, it
was assumed that there was no equip-
ment duplication and that the plant was
operating at the design condition. The
resulting criticality rating for each
component-application combination
reflected the degree of impact on the
effluent quality (significant, minimal, or
no impact) as a function of time after
failure (0-4 hours, 4-12 hours, or 24
hours).
From this analysis, the components
included in the data collection effort
were:
• Raw and intermediate wastewater
pumps, power transmission, and
motors for all plant types
• Return activated sludge pumps,
power transmission, and motors
for air and oxygen activated sludge
• Recirculation pump, power trans-
mission, and motors for trickling fil-
ter and rotating biological
contactors
• Motors and power transmission for
final clanfiers for all plant types
• Motors, power transmission, com-
pressors, valves, controls, and dif-
fusers used in dissolved air
production application and
mechanical aerators in air acti-
vated sludge plants
• Motors, power transmission, recir-
culation pumps, controls, diff users,
and valves used in oxygen genera-
tion application and recirculation in
oxygen activated sludge plants
• Liquid application systems for trick-
ling filter plants
• Compressors, motors, power trans-
mission, and controls used in prim-
ary clarifiers and primary sludge
pumping in trickling filter and RBC
plants
• Recirculation pumps controls used
in secondary treatment in RBC
plants
• Raw and intermediate wastewater
pump controls used in all plant
types and valves used in raw and
intermediate pumps in trickling fil-
ter plants
• Controls used in dosing siphon in
secondary treatment in trickling fil-
ter plants
• Controls used in final clarifiers for
all plant types
• Pumps, motors, power transmis-
sion, and pressure vessels used in
disinfection for all plant types.
Data Source Selection
Having defined the data collection
requirements, treatment plants having
adequate records from which these
data could be collected were selected by
developing a list of candidate plants,
selecting plants for 1-day screening vis-
its, conducting the 1-day visits, and,
finally, selecting the study plants.
From a candidate list of approxi-
mately 200 treatment plants, the 9
plants to be included in the data acquisi-
tion effort were selected. The criteria
used to make the final selection were:
• Existence of adequate and com-
plete preventive and corrective
maintenance records for all major
equipment components at each
plant;
• Existence of plant equipment typi-
cal of the generic equipment found
in the four types of treatment plants
under consideration; and
• An indication that plant operations
and maintenance personnel were
willing to cooperate in the data col-
lection effort.
Data Acquisition
Maintenance records for each critical
component were reviewed and the data
were encoded on data collection forms
(Figure 1). Required data elements are
shown at the top of this form. All data
collected in the field were encoded
using a numbering system developed to
allow a computer program analysis of
the data. These codes are shown in the
appropriate columns in Figure 1. For
example, column one contains the code
number 04-8731. The first two digits,
04, identify the plant location. The
second four digit number represents a
record number used to identify that spe-
cific piece of equipment. Column two,
plant type, shows the number 05, which
identified this plant as a combination
air/oxygen activated sludge plant.
Codes for each of the remaining 23
columns are shown, with a description
of each given below the code. Column
23 was for recording actual down time
of sludge processing systems. Data for
the sludge processing systems at each
facility were recorded as shown at the
bottom of Figure 1. These data allowed
calculation of the MTBF and availability
of the system.
The data collection from plant records
often required a degree of judgment and
interpretation. This was especially true
when deciding when equipment had
actually failed. A failure was defined as
when the mechanical component no
longer performed its intended function
because of a fa ilure of that component's
subcomponent(s). In contrast, an equip-
ment malfunction was defined as situa-
tions where the equipment could be
placed back into service by minor clean-
ing, or adjustments, or both. Only data
for down time considered to be caused
by failure were recorded. Malfunctions
including pump blockages, reduction in
blower capacities due to dirty air filters,
and plugging of air diffusers were not
recorded. If interpretation of mainte-
nance records was questionable, the
data collection team conferred with
plant maintenance staff to decide
whether a failure had occurred.
Data Analysis
A computer program entitled
"Wastewater Reliability Analysis Pro-
gram — WRAP" was written to handle
the sorting and analysis of the data
base. Reliability, maintainability, main-
tenance, and availability statistics were
-------
Plant No. 4
Unit Number
2
K)
Combination
Air/Oxygen Act. SI.
Incinerator
July. 1976
100%
II Failures
Down time
1.584 hours
August. 1979
§
o
o
SI
N
o>
CO
CO
-^
1
ID
SI
CO
Sludge Processing Equipment
Plant No. 4
Unit Number
2
Combination
Air/Oxygen Act. SI.
Secondary treatment
Pumps
Pump propeller
axial
Intermediate wastewater
Pump
100.000 gpm
Bearing, ball
Double row
Effluent impact
within 4 hours
July, 1977
100%
1 Failure
Not known
Normal wear
24-168 hours
6 hrs - 24 hours
Dry basement
Replaced
Not known
A His Chalmers
Not known
Corrective maintenance
August. 1979
8 hours
§
8
o
0
fe
8
2
S
Co
o
0
SI
s
s
CO
CO
2
§
§
2
o
Co
2
o
1*0
§
n
8
o
Kj
O
OB
S
|
>
Wastewater Treatment Equipment
Plant
location
Type
Unit
operation
Co
Broad
Classification
Generic
groups
Ol
Application
Size
Sub-
component
CD
Criticality
co
Date
installed
% Operating
time
Failures
Failure
mode
co
Cause of
Failure
Down
time
Time to
repair
Environmental
factor
Part
disposition
Subcomponent j*
manufacturer
Component
manufacturer
Part Inv.
information
Type of
maint.
o
Down time of
unit op.
fSL only)
Collection
date
Clock time
repair
Figure 1. Sample of a Field Data Collection Form.
calculated from the data base across all
plants and have been grouped by:
• Components
• Components by application
• Components by size/capacity
• Components by generic group
• Components by generic group by
application
• Components by generic group by
size/capacity
Further calculations included failure
distribution information by subcompo-
nent type for the first four categories
listed above and reliability and availabil-
ity statistics for sludge processes. An
example of the type of reliability, main-
tainability, maintenance, and availabil-
ity statistics calculated from the data is
shown in Table 1. For each application,
the following data elements/statistics
are given:
1. Number of units (pumps) in the
data base.
2 Number of failures reported.
3. Total operating hours, in millions
of hours.
Reliability Statistics
4. Mean time between failures
(Overall MTBF) for the pumps in
millions of hours.
5. Minimum MTBF (Min MTBF) for
any subclassification in the cate-
gory.
6. Maximum MBTF (Max MTBF) for
any subclassification in the cate-
gory.
7. Lower 90% confidence limit on
the point MTBF estimate (overall
90% CL)
Maintainability and Maintenance
Statistics
8. Preventive maintenance man
hours per unit per year (PM-
hr/Umt/yr)
9. Corrective maintenance man
hours per unit per year (CM-
hr/Unit/yr)
10. Mean time to repair (MTTR)
Availability Statistics
11. Inherent availability (AVI)
12 Operating availability of the com-
ponent (AVO).
Table 2 is an example of data on sub-
component failures by component type
Results
The performance statistics calculated
from the data serve as the best sum-
mary of the data base in terms of the
reliability, maintainability, mainte-
nance, availability, and subcomponent
failure distribution parameters for the
nine wastewater treatment plants.
-------
Table 1 . Reliability and Maintenance Data for Pumps — Open Impeller Centrifugal by Application
Operating Overall Mm Max Overall PM-Hrs/ CM-Hrs/
Application
Raw wastewater
pumping
Intermediate
wastewater
pumping
Return activated
sludge
handling
TOTAL
No of
Units
34
9
27
70
No of
Failures
60
9
28
97
Hours
fx 10e Hr)
.9277
.1669
1.023
2.1176
MTBF
Ix 706 Hr)
.01530
.01727
.03569
.02166
MTBF MTBF 90% CL
Ix 10e Hr) fx 10e Hr) Ix TO6 Hr)
— — .07303
— — .07775
— - .02535
.07530 .03563 .07977
Unit/
Year
.009442
1079
.001745
.003428
Unit/
Year
.3179
.7871
.1028
.094773
MTJR
(Hr)
8483
6.667
6.786
7.825
AVI
.994
.9996
.9998
.99962
AVO
.9845
.9548
.9998
.98955
Table 3 is an example of the "by com-
ponent" grouping, showing reliability
statistics for the eight critical compo-
nents included in the data base, across
all plants. Table 4 presents the main-
tainability, maintenance, and availabil-
ity statistics for the eight components
across all plants. As previously indi-
cated, the failure data for sludge pro-
cesses were also collected at the nine
treatment plants, and the results of the
data analysis for five processes are
given in Table 5.
Conclusions
Based on the results of the perform-
ance data collection and analysis effort
conducted during this project, the fol-
lowing general conclusions are made:
1. Three types of performance
statistics (reliability, maintain-
ability, and availability) have been
determined for selected critical
mechanical components of
wastewater treatment plants
2 These statistics were not avail-
able elsewhere in the literature,
from manufacturers, or from
owners of equipment.
3. Design engineers can use these
performance statistics as a tool to
compare and predict performance
of generic equipment in various
applications and to design relia-
bility into new treatment plants.
4. Wastewater treatment plant
operators can use these data to
develop spare parts inventories
for new equipment and to estab-
lish preventive maintenance pro-
grams.
5. Despite the significant number of
operating hours included in this
study, this data base does not
cover all the size ranges or types
of equipment found in the 21,000
treatment plants operating in the
United States. Lacking are data
Table 2. Pumps — Open Impeller Centrifugal
Rank
1
2
3
4
5
Subcomponent Type
Impeller wear ring or plate
Seal, packed, water, oil, grease lubricated
Solenoid valve
Bearing, ball, double row
Bearing cast pillow
No. Of
Failures
15
14
14
6
5
Relative
Frequency
.197
.184
.184
.079
.066
Table 3. Mean Time Between Failure I MTBFj for Components
Component
Pumps
Power transmissions
Motors
Compressors
Diffusers, air/water
Valves
Controls
Conveyor (unconfined
materials handling)
Overall
MTBF
Ix 106 Hr)
.032066
.03562
.06670
.00714
.01813
.01444
.08358
.14848
Mm
MTBF
(x 70s Hr)
.021662
.01785
.01088
.00562
.01263
.00893
.00393
.06175
Max
MTBF
(x 106 Hr)
.074191
.71091
.11482
.08392
1.834
.03259
.10064
.35856
Overall
90% CL
(x 106 Hr)
.028630
.03317
.06122
.00631
.O1667
.01040
.07569
.11690
Table 4. Maintenance Statistics for Components
Component
Pumps
Power transmissions
Motors
Compressors
Diffusers, air/water
Valves
Controls
Conveyors (unconfined
materials handling)
PM-Hrs/
Unit/Yr
.00227
.00032
.00098
.03399
.00065
00684
.00025
.00055
CM-Hrs/
Unit/Yr
.05177
.00469
.00660
. 1 7843
.04664
.73845
.00261
.10086
MTTR
Hr
9.541
2.273
6.854
0.960
8.305
11.615
3.696
4.768
AVI
.99968
.99994
.99989
.99987
.99951
.99879
.99996
.99996
AVO
.99116
.99898
.99816
.99306
.99875
.96446
.99870
.99980
-------
Table 5.
Reliability of Sludge Processes
Process
Sludge process external to
WW treat, process
Anaerobic digestion
incineration
Sludge thickening DAF
Vacuum filter
Operating
Hours
(x 106 Hr)
.780
.678
.374
406
.697
No. of
Failures
400
40
454
702
1150
MTBF
(x TO6 Hr)
.00195
.0167
.000824
.000577
000606
90% CL
(x 10s Hr)
.00183
.0137
000776
.000583
.000583
(x W6 Hr)
A vail.
.997
1.000
.723
.994
.931
for many small size range com-
ponents; these records are not
routinely kept by the smaller
plants surveyed during this study.
6. The quality of data obtained from
the nine treatment plants was
good. The form of record keeping
sometimes required judgments
as to the appropriateness with re-
spect to project data require-
ments.
7. Effluent quality from wastewater
treatment plants can be expected
to improve (to the extent that
equipment failure causes de-
creases in quality) over the long
term by use of performance sta-
tistics by design engineers and
operators. For example, design
engineers can use these data as
an additional tool when selecting
equipment, in determining the re-
liability and availability of unit
operations as well as entire treat-
ment systems, and in predicting
operational performance of
equipment in various applica-
tions. Plant owners/operators
can use preventive and corrective
maintenance data to help refine
overall future maintenance bud-
get and staffing projections.
Knowledge of equipment failure
rates and subcomponent failures
can provide input into determin-
ing spare parts inventories.
The full report was submitted in ful-
fillment of Contract No. 68-03-2712 by
Southwest Research Institute under
the sponsorship of the U S. Environ-
mental Protection Agency
David W. Shultz and Van B. Parr are with Southwest Research Institute, San
Antonio, TX 78284.
Jon Bender is the EPA Project Officer (see below}.
The complete report, entitled "Evaluation and Documentation of Mechanical
Reliability of Conventional Wastewater Treatment Plant Components," (Order
No. PB 82-227 539; Cost: $19.50. subject to change) will be available only
from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
•&U. S. GOVERNMENT PRINTING OFFICE: 1982/559-092/0464
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
RETURN POSTAGE GUARANTEED
Third-Class
Bulk Rate
LOU hi TILLtY
HEGIUN V EHA
LIBRARIAN
CHICAGO LL 60604
ST
------- |