EPA-AA-TSS-83-8-A
Technical Report
EPA RECOMMENDED PRACTICE
FOR NAMING I/M CALIBRATION GAS:
A DISCUSSION FOR I/M PROGRAMS
September 1983
Technical Support Staff
Emission Control Technology Division
Office of Mobile Sources
Office of Air and Radiation
U.S. Environmental Protection Agency
Ann Arbor, Michigan 48105
-------
Abstract
This report explains how calibration gases are used in I/M
programs, and introduces a Recommended Practice for gas
manufacturers to use when naming I/M calibration gas
cylinders. Details of the Recommended Practice are presented
in a separate report entitled "RECOMMENDED PRACTICE FOR
NAMING I/M CALIBRATION GASES" (EPA-AA-TSS-83-8-B). States
are encouraged to procure gases named according to this
Recommended Practice for their own use, and to require
licensed inspection stations to procure them to ensure that
they are obtaining accurate calibration gases which meet the
terms of the Emission Control System Performance Warranty,
and to improve the general quality of their I/M programs.
-------
Table of Contents
Page
1.0 BACKGROUND 1
1.1 EPA Regulations Concerning Calibration
Gas Accuracy 1
1.2 What are NBS Standards, and How Are They
Used to Make Accurate Gases? 2
1.2.1 NBS Weight Standards 2
1.2.2 NBS Gas Standards 2
1.3 EPA Efforts to Help States Obtain
Accurate Gases . 4
2.0 DISCUSSION 5
2.1 Selecting Components 5
2.1.1 Propane and Hexane .6
2.1.2 Diluents 8
2.2 Concentration of I/M Calibration Gases 9
2.2.1 Warranty Requirements 9
2.2.2 State Requirements - C02 Concentration 10
2.2.3 State Requirements - Carbon Monoxide -
Propane Interference 10
2.3 Cylinders and Hardware 11
2.4 Cylinder Labels and State Audits 12
2.5 EPA Audits 13
3.0 RECOMMENDATIONS 13
Appendices
Appendix 1 Emission Performance Warranty Regulations
Pertaining to Calibration Gases
Appendix 2 Carbon Monoxide - Propane Interference
-------
1.0 BACKGROUND
Inspection and Maintenance (I/M) programs are now fully
operating in seventeen States, and other States are expected
to implement I/M programs, in the coming months. Many I/M
managers are concentrating on implementing and refining
program quality assurance procedures so that motorists .can
obtain fair and accurate tests of their cars' emissions.
One of the benchmark quality assurance procedures which will
be used by every I/M program is the periodic check of every
inspection analyzer with a calibration gas of known
concentration. This is the only method of checking how
accurately a given analyzer is reading vehicle exhaust
emissions. The calibration check is performed by flowing the
calibration gas into the analyzer and determining whether the
analyzer is reading this calibration gas correctly. If the
analyzer is not reading this calibration gas accurately, then
a simple adjustment of the analyzer can usually be performed
which will result in the analyzer being accurate again.
For gas calibration checks and related adjustments to be done
properly, the accuracy of the calibration gas used is
critical. However, because of a lack of standard procedures
in the gas blending industry, it is not safe for the I/M
manager to assume that calibration gases have been named
properly and are traceable to known standards (NBS). If the
labeled concentration of the calibration gas is significantly
different from the true concentration in the cylinder, the
analyzer could become significantly misadjusted.
Furthermore, an operator who had performed the calibration
would be unaware that he/she had actually misadjusted the
analyzer, since he/she trusted the label on the calibration
cylinder to be correct.
In light of its potential significance, EPA has been working
with the gas industry over the past several years to identify
a way to resolve this problem. The result of these
activities is the Recommended Practice which is referred to
in this report and discussed in detail in the parallel report
("EPA Recommended Practice for Naming I/M Calibration Gases,"
EPA-AA-TSS-83-8-B). .
1.1 EPA Regulations Concerning Calibration Gas Accuracy
In 1980 the EPA promulgated Emission System Performance
Warranty Regulations for 1981 and later vehicles which
entitle a vehicle owner to emission-related repairs at the
manufacturer's expense if, among other things, the vehicle
fails an "approved" emission short test. One condition is
that the analyzer used to conduct the emission short test
-------
— e.
must be checked with gases "traceable to NBS [National Bureau
of Standards] standards +2%" at least once per week. (See
Section 85.2217, 45 FR 34808, May 22, 1980. The relevant
portion of this page is provided in Appendix 1 to this
document.) Many States implementing I/M programs will want
to meet the quality assurance requirements of the Warranty
regulations so that eligible consumers participating in their
I/M programs will have the benefit of receiving warranty
repairs from the automobile manufacturers. These States must
make sure that the gases used to periodically calibrate
analyzers used in their I/M programs meet the regulatory
requirement of having an accuracy of +2% to NBS standards.
1.2 What Are NBS Standards and How Are They Used to Make
Accurate Calibration Gases?
A standard is an object to which other similar objects are
compared. The National Bureau of Standards takes the utmost
care in maintaining in-house standards of many different
kinds. The standards of concern in obtaining accurate
calibration gases are weight standards and gas standards.
These standards can be used in a variety of ways in obtaining
accurate calibration gases. Some of these ways are better at
obtaining accurate calibration gases than others.
1.2.1 NBS Weight Standards
NBS weight standards can be used in making accurate
calibration gases by weighing in certain masses of
calibration gases into a cylinder. For example, a cylinder
with a desired concentration of 1000 ppm carbon monoxide (CO)
in nitrogen might require 300 grams of CO to achieve that
concentration. The cylinder is first emptied and then
weighed with a scale calibrated with the NBS weights. The
300 grams of CO are then pumped into the cylinder. Then
nitrogen is added to the cylinder in a certain weight to
yield the 1000 ppm CO with an appropriate pressure. The CO
mixture might then be considered partly "traceable to NBS"
via the NBS weight standards. At least some members of the
gas manufacturing industry do not consider a mixture made
with this method to be entirely "traceable to NBS". They
feel that the mixture should be compared to some NBS gas
standards using a gas analyzer before the mixture is
considered "traceable to NBS". EPA agrees that a mixture
made solely using the weight-based method is not "traceable
to NBS", particularly as that phrase was intended in the
Warranty regulations.
1.2.2 NBS Gas Standards
The National Bureau of Standards tries to make available
standard gases in commonly requested concentrations. Until
-------
-3-
1981, the only NBS standards available were called Standard
Reference Materials (SRM's). SRM's have a maximum analytical
uncertainty of +1% to the true value of the concentration in
the cylinder. The analytical uncertainty is usually less
than j+0.5% of true value.
EPA has required SRM's as the standards to which calibration
gases used in many mobile and stationary source programs have
to be traced. This created a large demand for SRM's since
the tracing process consumes the SRM's. Also, the analysis
procedures for SRM's are very detailed and lengthy. These
two factors have resulted in SRM's being in short supply. To
alleviate this problem, the EPA Emission Monitoring and
Support Laboratory at Research Triangle Park (EMSL-RTP) and
NBS jointly developed procedures that would allow gas
manufacturers to make a new kind of standard, called a
Certified Reference Material (CRM), which if made properly
has a maximum uncertainty of about +1.3% to true value.
These CRM's can be made in batch quantities, but a limitation
of these CRM's is that their concentrations cannot be more
than about +1% different from current SRM concentrations.
These CRM's are themselves NBS standards, and a gas mixture
which is traceable to a CRM is "traceable to NBS".
NBS gas standards can be used in making I/M calibration gas
by comparing the standards to an I/M gas mixture after the
I/M gas mixture has been blended into a cylinder. Usually
the instrument being used to analyze a mixture is calibrated
or checked with NBS gas standards first, then .the instrument
is used to determine the concentration of the subject
mixture. The mixture is then considered "traceable to NBS"
since NBS standards were used to calibrate or check the
instrument used in analyzing the I/M gas.
Generally, gas manufacturers agree that the method of
demonstrating traceability between a gas mixture and NBS gas
standards is a better method than the weighing technique,
when NBS gas standards exist for a given gas mixture,*
because of several possible sources of errors in the latter.
(For instance, in the weighing method improper evacuation of
the cylinder prior to filling can cause the concentration of
the cylinder to be significantly different from its intended
concentration. Errors can also occur in the filling
process.) However, even though there is agreement that the
gas standard comparison procedure is probably best, there is
* There are no NBS gas standards for hexane, therefore
traceability to NBS within the Recommended Practice can only
be obtained with the weighing technique. The Recommended
Practice requires hexane gases to be checked with a gas
analyzer against the manufacturers' gravimetric (i.e.,
weighed-in) primary hexane standards.
-------
-4-
little agreement among manufacturers about what analytical
techniques are best when comparing NBS gas standards with
other gas mixtures. These different analytical procedures
and methods of handling data may cause differences in the
concentrations of identically labeled calibration gases made
by different manufacturers or even by the same manufacturer.
I/M managers would find it difficult to detect these
differences and identify which gas cylinders were correctly
labeled, unless they maintain a sophisticated gas analysis
program of their own.
1.3 EPA Efforts to Help States Obtain Accurate Calibration
Gases
In order to improve the general quality of I/M programs and
to help States in meeting the calibration gas accuracy
requirement of the Warranty regulations, EPA has developed a
standard procedure which can be used to accurately name
calibration gas traceable to NBS. In a separate report,
entitled."EPA RECOMMENDED PRACTICE FOR NAMING I/M CALIBRATION
GASES" (EPA-AA-TSS-83-8-B) , the "Recommended Practice," which
individual States can require gas manufacturers to follow in
naming and labeling calibration gas for their I/M programs,
is outlined in detail. A State requirement could be imposed
by the State in its own purchases of gas and by establishing
rules and regulations which require other I/M gas users
(contractors or licensed inspection stations) to buy only gas
that the manufacturer certifies to be named and labeled
according to the Recommended Practice. Such a requirement
would assure consistent gas quality among both manufacturers
and inspection stations.
Because of the lack of agreement among gas manufacturers as
to the analytical procedures for tracing gas mixtures to NBS
standards, EPA has taken the initiative in selecting a single
set of procedures which it believes represents a good
compromise between cost and quality. A special effort has
been made to make these procedures workable from a gas
manufacturer's viewpoint. A meeting of representatives of
the gas manufacturing industry, the National Bureau of
Standards and the EPA was held to decide on the original
framework of the procedures. Manufacturers have also been
allowed to comment on the draft forms of the Recommended
Practice, and many of their comments have been incorporated
into the procedures.
-------
-5-
2.0 DISCUSSION
2.1 Selecting Components of I/M Calibration Gases
The primary criterion for the user to consider in deciding
which components to include in the calibration gas is the
kinds of pollutants measured in the I/M program. Some I/M
programs measure only for CO emissions; most measure both HC
and CO; and a few measure HC, and/or CO, and C02« Whenever
a particular pollutant is measured, then the analyzer
accuracy for that pollutant should be periodically verified
with a calibration gas.
In those I/M programs which measure only CO, the State may
feel that it is necessary only to calibrate analyzers with a
CO calibration gas. However, because inspection analyzers
also measure the HC in the exhaust, and because mechanics use
these HC measurements along with CO measurements for tune-up
adjustments and for diagnosing problems causing high
emissions and/or poor fuel economy, EPA recommends that the
State also require periodic calibration checks of the HC
channel of the analyzer. EPA feels that the extra cost of
calibrating for both HC and CO is not significantly greater
than the the cost of calibrating for only CO and that this
cost is greatly outweighed by the benefits of better quality
of service to the public.
Some States are also requiring the measurement of C02 in
order to ensure that emission measurements are not biased by
improper probe insertion depth or exhaust system leakage,
both of which may cause dilution of exhaust gases and, thus,
lower emission readings. Whenever C02 measurements are
required, the State should also require periodic calibration
checks of the C02 channel with a C02 calibration gas. In
this case, the issue arises as to whether the C02 gas
should be included along with HC and CO (and diluent) in the
same cylinder or whether the C02 gas should be obtained in
a separate cylinder. At the heart of this issue is the basic
traceability issue. Introducing C02 into the HC/CO/diluent
blend greatly changes the viscosity of the mixture thereby
confounding analysis of the HC component by the flame
ionization technique (which is currently the best technique
for traceability for HC) . There is also an. interference
effect between measurements of CO and C02 using the
non-dispersive infrared technique, thus confounding analysis
of these two components. These problems can be overcome by
performing gas analysis using gas chromatography.
Some proponents of C02 inclusion in the calibration gas
mixture argue that because C02 is a normal constituent of
auto exhaust, there is some degree of C02 interference in
the I/M analyzer's CO readings. They conclude, therefore,
that it would be beneficial to have C02 in the calibration
-------
-6-
gas so that the calibration gas would more closely simulate
auto exhaust. Again, establishing traceability, as discussed
above, is a basic problem. Another consideration here is
that the amount of C02 in auto exhaust varies between 6 and
14%, depending on the condition of the vehicle and whether an
air pump and/or catalytic converter is used on the vehicle.
Because this variability is unpredictable among vehicles,
including CC-2 (one specific concentration) in the
calibration gas is just as likely to introduce new errors as
it is to correct existing ones.
In light of the above, EPA recommends that all States use
calibration gases which include HC and CO. For those States
which require CC>2 measurements, EPA recommends that the
C02 calibration gases be obtained in separate cylinders
unless the traceability issues discussed above can be
adequately addressed.
Two side issues related to selecting calibration gas
components are discussed below. Section 2.1.1 discusses the
hexane versus propane issue, and Section 2.1.2 discusses the
issue surrounding the choice of diluents.
2.1.]. Propane versus Hexane
Even though analyzers measure HC emissions in terms of
hexane, analyzers have traditionally been calibrated with
propane. This is because propane is an easier gas to work
with and is more commercially available.
Because propane is the traditional calibration gas and the
analyzer actually measures only the "hexane equivalent" of
all non-hexane gases, analyzer manufacturers must establish,
during the manufacturing process, a ratio called the propane
equivalency factor (PEF) for each analyzer they build. An
analyzer's PEF defines the optical relationship between
propane and hexane for that particular analyzer's optical
bench. PEF values generally range from 0.48.to 0.56 for good
quality I/M analyzers. An analyzer which has a PEF of 0.52
will provide a reading of 520 ppm HC (measured as hexane)
when calibrated properly and checked with a calibration gas
with a concentration of 1000 ppm propane. (The general
formula is: Analyzer Response (hexane) = PEF x Calibration
Gas Concentration (propane).) This kind of calculation has
to be performed each time an analyzer is checked for
calibration.
Some I/M program officials feel that it is more appropriate
to use hexane as the calibration gas and, thus, to avoid
having to perform the PEF calculations and the possible
-------
— 7 —
arithmetic errors associated with them. In addition,there is
concern that the PEF value for a particular analyzer may
change during the analyzer's useful life.
There are two constraints which make the use of hexane as a
calibration gas more difficult than using propane. First,
hexane has a low vapor pressure which causes it to condense
on the cylinder walls at certain combinations of temperature
and concentration levels. Table 1 illustrates the
relationship of temperature and concentration at a normal
cylinder pressure (for a class 1-A cylinder) of 2000 psi.
Table 1
Theoretical Dew Point Temperatures of
Certain Hexane Concentrations at 2000 psi
Dew Point Temperature Hexane Concentration
81°F 1600 ppm
61°F 1000 ppm
43°F 600 ppm
28°F 400 ppm
7°F 200 ppm
-15°F 100 ppm
The proper interpretation of the information in Table 1 is
that, theoretically, a mixture of 600 ppm will begin to
condense if the temperatures falls below 43°F. Conversely,
at a temperature of 43°F, mixtures of hexane with
concentrations greater than 600 ppm will condense to some
extent. The cylinder pressure is also a factor in this
relationship. If the cylinder pressure is less than 2000
psi, as assumed to develop Table 1, the dew point
temperatures would be lower for the hexane concentrations
listed in Table 1, and the concentrations would be higher for
the temperatures listed.
For instance, a mixture of 1000 ppm hexane at 1000 psi would
have a theoretical dew point of about 34°F. A mixture of
1600 ppm hexane at 500 psi would have theoretical dew point
of about 28°F. Thus, by limiting cylinder pressure to these
values and maintaining constant temperatures above the
freezing point of water, gas manufacturers can blend stable
mixtures of hexane at these higher concentrations. The
trade-off, however, is a reduced amount of calibration gas in
each cylinder purchased, since the volume of gas in a
cylinder is directly proportional to the cylinder pressure.
At 500 psi, for example, only one-quarter as much calibration
gas can be stored in a cylinder as in the same cylinder at
2000 psi.
-------
-8-
The condensation effects of hexane are not as troublesome for
the smaller, disposable cylinders, which have a maximum
cylinder pressure of 300 psi or less. At this pressure, a
mixture of 1600 ppm hexane has a theoretical dew point of
about 10°F, and lower concentration mixtures would have even
lower dew points. Of course, the cost per volume of gas is
higher for these smaller cylinders.
If hexane calibration gases are used, it is necessary to take
adequate precautions during gas shipments, storage and use to
make sure that the cylinders are never exposed to
temperatures below their dew point level. If a lower
temperature limit of 43°F were set, the working range of
hexane calibration gases in the field would be about 200 ppm
to 500 ppm hexane (in 1-A cylinders) . However, many areas
around the country typically have temperatures below 43°F.
For such areas the lower temperature limit would need to be
less than 43°F, thus restricting the working range of hexane
concentration even more. The exact hexane value for a
particular State would depend on the State's climate during
the season of interest. In general, very few problems would
be encountered during the summer months in any place in the
country. However, during winter months, the use of hexane
calibration gases in many States may be very risky.
The second constraint regarding the use of hexane is that,
unlike propane, there are no NBS gas standards for hexane.
Consequently, hexane traceability must be established through
the use of weight standards (sometimes referred to as
gravimetrics) and the manufacturer's in-house hexane gas
standards which are developed from the weight standards.
Traceability to NBS can be established in this manner through
careful analysis. (At least one gas manufacturer has already
developed in-house gas standards for hexane, but others have
not.) Because of this more complex analysis, hexane
calibration gases may be slightly more expensive in the short
run (for those manufacturers which not yet developed in-house
hexane gas standards) . However, over the long run, cost
differences should disappear as the gas manufacturers perfect
and streamline the hexane analysis process.
2.1.2 Diluents (Balance Gases)
The diluent is by far the largest component of the gas
mixture. It is the carrier or balance gas in which the gases
of interest are mixed to get the desired concentrations. The
diluent must be' chosen with care to avoid chemical reactions
with the other gas components and contamination due to
impurities which may confound gas analysis.
Nitrogen is the easiest gas to use as a diluent for I/M
calibration gases since it is readily available in very pure
-------
— 9 —
forms. Nitrogen is also the best diluent to use for two
reasons. First, nitrogen does not react with HC, CO, or
CC>2. Second, it creates no interference problems in the
gas analysis (naming) process. Purified air, the other
possible diluent, on the other hand, is not recommended as a
diluent for CO because the oxygen in air causes biases in
naming the HC component with a process type total hydrocarbon
analyzer. This type of analyzer is commonly used by gas
manufacturer to name HC gases. Because of this problem,
purified air is not recommended as a diluent for I/M
calibration gases.
The I/M gas user needs to be aware of the problems with air
as a diluent for one other reason. NBS has available propane
gas standards in either nitrogen or purified air as the
diluent. In ordering calibration gases, I/M users need to be
careful to stipulate that NBS propane gas standards in
nitrogen are to be used in the analysis of the HC component
of the I/M calibration gas in order to prevent errors in
analysis.
2.2 Concentrations of I/M Calibration Gases
2.2.1 Emission Performance Warranty Requirements
In addition to the accuracy requirement for gases used .to
check emission analyzers (i.e., +2% to NBS standards), the
Warranty regulations also require the calibration gas to be
within certain concentration ranges. Section 85.2217 states
that "span gases shall have concentrations either:
(i) Between the standards specified in this subpart and
the jurisdiction's inspection standards for 1981
model year light duty vehicles, or
(ii) Be within -50% to +100% of the standards in this
subpart."
The standards specified in the Warranty regulations are 1.0%
CO and 200 ppm HC (measured as hexane) for the two-speed idle
test and 1.2% CO and 220 ppm HC for the idle test or the
two-mode loaded test. Some States are using these cutpoints,
although many States are choosing standards for 1981 and
later vehicles which are less stringent. At this time, the
most lenient standards likely to be chosen for 1981+ vehicles
are 4.0% for CO and 400 ppm for HC (measured as hexane).
Table 2 indicates the required calibration gas concentrations
for each of these cases.
-------
-10-
Table 2
Required Calibration Gas
Concentrations for Selected I/M Cutpoints
Outpoints
(CO/HC*)
1.0%/200 ppm
1.2%/220 ppm
4%/400 ppm
Cal. Gas Concentration
CO (%)
0.5-2.0
0.6-2.4
0.5-4.0
HC (ppm)
100-400 (if hexane)
200-800 (if propane)
110-440 (if hexane)
220-880 (if propane)
100-400 (if hexane)
200-800 (if propane)
*HC outpoints are always expressed as hexane, and vehicular
HC emissions are always measured as hexane in I/M programs.
Based on the information in Table 2, it is clear that users
can meet the gas concentration requirement in the Warranty
regulations by limiting their gas orders to concentrations in
the ranges of 1.0-2.0% CO and 200-800 ppm propane (or 100-400
ppm hexane).
2.2.2 State Requirements - CO? Concentration
In those cases where a CC>2 calibration gas is required, the
C02 concentration of the calibration gas should approximate
the C02 cutpoint used in the I/M program. Unlike HC and CO
where cutpoints generally vary by model year, there is often
only one C02 cutpoint applied to all model years.
(Sometimes a State may choose different C02 cutpoints for
vehicles with and without air pumps.) Typical State
cutpoints for C02 are from 4 to 6%.
2.2.3 State Requirements - CO/Propane Interference
In addition to meeting the Warranty requirements for HC and
CO calibration gas concentrations, some States may wish to
use other concentrations of HC and CO (and C02) in
performing multipoint calibrations on analyzers used in the
I/M programs. In these cases, the State would need several
cylinders of calibration gases at varying concentrations, all
with accuracies of +2% to NBS standards for best results.
However, States should be aware of certain aspects of
ordering high concentration (i.e., propane greater than 500
ppm and CO greater than 4.0%) blends of calibration gases.
There is a small interference effect between CO and propane
at high concentration levels on certain instruments that are
-------
-11-
used to name these components. This interference effect is
described in detail in Appendix 2. To avoid this
interference, when high concentration calibration gases are
being blended, gas chromatographic analysis of the components
should be used by the gas manufacturer to label the gas after
mixing. This is required by the Recommended Practice for
cylinders containing greater than 500 ppm propane (or 250 ppm
hexane) and 4.0% CO. Failure of the gas manufacturer to do
this could result in gases not meeting the accuracy limits of
+2% traceable to NBS. Although there have been no known
studies of the degree of interference between CO and hexane,
it would be advisable to require gas chromatographic analysis
of hexane/CO mixtures of similar concentrations as a
precaution. Of course, such mixtures will often not exceed
500 ppm hexane because of the condensation problems discussed
earlier.
2.3 Cylinders and Hardware
Gas manufacturers have many different sizes of cylinders they
can use for I/M calibration gas mixtures, ranging from very
large size 9 1/4" by 60" high pressure cylinders* to the
low-pressure disposable cylinders. The major cost of a
mixture of calibration gas is not for the raw gases that go
into making it, but for the analyses required to make sure it
is labeled correctly. Consequently, the large cylinders are
much more economical per cubic foot of calibration gas
mixture than the small disposables, although the disposables
have a lower cost per cylinder.
The Recommended Practice requires that the large 9 1/4" by
60" size cylinders be equipped with CGA-350 valves, and that
the disposable cylinders be equipped with 1/4-inch
flare-fitting valves. The Recommended Practice does not
specify valves for other size cylinders because there are so
many different sizes. Furthermore, the 9 1/4" by 60" and the
disposables will probably be most commonly used. The
Compressed Gas Association (CGA) makes recommendations for
valves for other cylinders, consequently, there is uniformity
between manufacturers on valve usage in other size cylinders.
Users of the I/M mixtures will need regulators to adjust the
delivery pressure of the gas mixture to the analyzer. The
uniform specifications for valve sizes for the large and
disposable cylinders will allow users to change gas
manufacturer suppliers without having to purchase new
* The Department of Transportation (DOT) has specifications
for all cylinders used to hold pressurized gases. The DOT
designation for this cylinder is 3AA2400. Some manufacturers
call this a "K" cylinder; others call it a "1-A" cylinder.
-------
-12-
regulators, as long as the same size cylinder is ordered from
the new supplier. Regulators may or may not be
interchangeable among different size cylinders.
EPA highly recommends the use of dual-stage regulators with
all calibration gas mixtures and cylinders instead of single
stage regulators. The dual stage regulators allow gas to be
delivered to the analyzer at a constant pressure, even though
the pressure in the gas cylinder declines as gas is used.
The single stage regulators do not have this feature. This
feature is important because it helps ensure that the
instrument is neither starved nor over-pressurized with
calibration gas, when gas is introduced through the probe.
Users may also need additional hardware to connect the gas
cylinder to the instrument through the sample cell port
and/or probe. However, different analyzer manufacturers have
different ways of handling this. Users should discuss with
their analyzer suppliers what additional hardware is needed
to perform these hook-ups.
To conform to the Recommended Practice, the inside surfaces
of all cylinders must conform to the NBS CRM or SRM
requirements for preparation, cleanliness, trace materials,
composition, coatings, etc. for the gas composition and
concentrations used. Also, cylinder valves must conform to
the NBS CRM or SRM requirements for preparation, packing
materials, cleanliness, composition, etc. for the gas
composition and concentrations used.
2.4 Cylinder Labels and State Audits
The Recommended Practice requires gas manufacturers to label
gases with the following information:
(i) Cylinder number (except in the use of disposables,
where the batch number is required)
(ii) Concentration of propane or hexane (in ppm), CO (in
mol%) , and C02 (in mol%) and accuracy
specification (i.e., +2%, +1.0%, etc.)
(iii) Balance gas
(iv) Analysis date
(v) Cylinder numbers of NBS standards (and primary
standards if hexane is present) used in determining
instrument calibration curves
(vi) Vendor name
-------
-13-
(vii) The statement that "This gas has been named in
accordance with the EPA Recommended Practice of
September 1983 for Naming I/M Calibration Gas".
In most decentralized programs, State auditors will be
visiting all inspection facilities once per month in order to
check the facility's inspection analyzer, calibration gas and
records. The check of the accuracy of the facility's
calibration gas will probably be conducted as follows. The
auditor will check the facility's analyzer with the auditor's
calibration gas. If the analyzer cannot read the auditor's
calibration gas within about +5% of its labeled
concentration, the auditor will check the analyzer with the
facility's calibration gas. If the analyzer reads that gas
within about +3%, it may be concluded that either the
facility's gas or the auditor's gas has been improperly
labeled. If the former, then the information present on each
cylinder of gas named according to the Recommended Practice
will allow auditors to trace that cylinder back to its
manufacturer. Gas manufacturers are required to keep the
records of the naming process for each batch of gas named
according to the Recommended Practice for at least two
years. Through this process, it is possible that other
cylinders that had been improperly labeled could be
identified and recalled.
2.5 EPA Audits
EPA does not think it is necessary to check samples of a
batch of calibration gas named according to the Recommended
Practice prior to the sale and delivery of cylinders of the
batch. However, EPA may from time-to-time order gas mixtures
named according to the Recommended Practice from
manufacturers through a third party for audit purposes. The
results of such audits are likely to be published for public
inspection.
3.0 SUMMARY
EPA recommends that States should require .all gases to be
named with the EPA Recommended Practice in order to ensure
consistent, traceable calibration gases among gas
manufacturers and inspection stations. This should be done
by specifying the EPA Recommended Practice by name in
licensing regulations, contract provisions, and/or purchase
orders.
As pointed out in previous sections of this report, EPA has
the following' additional recommendations on I/M calibration
gases:
-------
-14-
1. Gas components should include CO and HC. CC>2
calibration gas, where needed, should be purchased in
separate cylinders unless the related traceability issues
are adequately addressed, in which case the C02 can be
obtained in the same cylinder as the CO and HC.
2. The HC component of the gas should be propane, unless
adequate precautions are identified and taken to overcome
the climatological factors related to hexane condensation
and the traceability issues related to the lack of NBS
gas standards for hexane.
3. The diluent for I/M calibration gases should be
nitrogen. The diluent for the NBS gas standards used in
analysis should also be specified as nitrogen.
4. Gas concentrations should be as follows:
a. CO: 1.0-2.0%.
b. HC: 200-800 ppm propane or 100-400 ppm hexane.
c. C02: about the same as C02 cutpoint.
5. For calibration gases with gas concentrations of greater
than 500 ppm propane (or 250 ppm hexane) and 4.0% CO, gas
chromatographic analysis should be used in the naming
process to avoid interference problems.
6. Users may purchase I/M calibration gases in any size
container they choose. However, corresponding gauges and
other hardware should be selected accordingly.
Dual-stage regulators are recommended. All inside
cylinder surfaces and valves must conform to the NBS CRM
or SRM requirements for preparation, cleanliness,
composition, etc. for the gas composition and
concentrations used.
-------
Appendices
-------
Appendix 1
EXCERPT FROM: 45 FR 34808, MAY 22, 1980,
EMISSION PERFORMANCE WARRANTY REGULATIONS
. § B5-2217 Calibrations, adjustments.
(a) Equipment shall be calibrated in
accordance with the manufacturers'
. instructions...- -
(b) Within one hour prior to a test, the
analyzers shall be zeroed and spanned.
: Ambient air is acceptable as a zero gas; .
• an electrical span check is acceptable. .
Zero and span checks shall be made on
the lowest range capable of reading the.
-. short test standard.
(c) Within eight hours prior to a • -..••
loaded test the dynamometer shall be -'
checked for proper power absorber
. settings. •
(d)(l) The analyzers shall have been "'•
• spanned and adjusted, if necessary.- V'
using gas traceable to NBS standards ±
2% within one week of the test These
span gases shall have concentrations- .:'"'"-;
. either. . - .: ••'•.* . • : -
(i) Between the standards specified tn~ '
this subpart and the jurisdictions
inspection standards for 1981 model
• year light duty vehicles, or
• • (ii) Be within -50% to +100% of the '
standards in this subpart
(2) For analyzers with a separate
calibration or span port CO readings
using calibration gas through the probe
and through the calibration port shall be
made: discrepancies of over 3% shall
require repair of leaks. No analyzer
adjustments shall be permitted during
this check.
-------
Appendix 2
CO-HC Interference Discussion
The EPA and the National Bureau of Standards (NBS) have found
that a small amount of interference exists in infrared and
total hydrocarbon analyzers between CO and propane at high
concentration levels.* NBS conducted a test program in which
two tri-blend calibration gases, one at a high concentration
and one at a lower concentration, were checked for
interference. The concentrations ordered for the high
concentration cylinder were 8.0% CO, 2500 ppm propane in
nitrogen. The concentrations ordered in the lower
concentration cylinder were 2.0% CO, 500 ppm propane.
NBS used a gas chromatograph (GC) to check the concentration
of. each component in both cylinders. Then a total
hydrocarbon (THC) analyzer was used to check the propane
values, and a non-dispersive . infrared analyzer was used to
determine the CO values. The assumption in this technique
was that the GC values were closest to the true values.
Interference between CO and propane would be seen on the THC
and NDIR analyzers, if the values they gave were different
from the GC values.
The results of the study are presented in the following
table, which shows the differences between methods of
analyses.
* "A Study of Interference in Trinary Span Gases Calibrated
With Binary SRM's", available by writing the Technical
Support Staff, U.S. EPA, 2565 Plymouth Road, Ann Arbor,
Michigan, 48105.
-------
Levels of Interference
Between CO and Propane
Hydrocarbon Concentration
As Determined By
Sample G.C. THA % Diff
XA4450 496 +lppm 501 +lppm 1%
A-013329 2474 +lppm 2541 +lppm 2.6%
Carbon Monoxide Concentration
As Determined By
Sample G.C. NDIR % Diff
XA4450 1.997 +.004% 2.002 +.004% .25%
A-013329 7.91 +.01% 8.04 +.02% 1.6%
G.C. = Gas Chromatograph
NDIR = Non-Dispersive Infrared Analyzer
THA = Total Hydrocarbon Analyzer
The data in the table indicate that, for the HC readings,
there is a substantial amount of interference (2.6%) caused
by the high concentration (8.0%) of CO. For the CO readings
the degree of interference caused by the high concentration
(2500 ppm) of propane is less (1.6%) but still significant.
The Recommended Practice assumes that there is a +1%
interference level between propane and CO for gases intended
to be traceable to NBS +2%, and controls other sources of
errors so that overall accuracy remains within +2%. From the
information in the table, it appears that this level of
interference can only be assumed at propane levels equal to
500 ppm or less, and CO levels equal to about 4.0% CO or
less. If States order higher concentration calibration gases
with intended accuracies of +2% NBS, gas manufacturers must
use gas chromatographs to overcome interference in their gas
analysis in order to be in accordance with the EPA
Recommended Practice.
------- |