[Code of Federal Regulations]
[Title 40, Volume 30]
[Revised as of July 1, 2004]
From the U.S. Government Printing Office via GPO Access
[CITE: 40CFR1065.110]

[Page 654-657]
 
                   TITLE 40--PROTECTION OF ENVIRONMENT
 
         CHAPTER I--ENVIRONMENTAL PROTECTION AGENCY (CONTINUED)
 
PART 1065_TEST PROCEDURES AND EQUIPMENT--Table of Contents
 
                    Subpart B_Equipment and Analyzers
 
Sec. 1065.110  Exhaust gas sampling system; spark-ignition (SI) engines.

    (a) General. The exhaust gas sampling system described in this 
section is designed to measure the true mass of gaseous emissions in the 
exhaust of SI engines. (If the standard-setting part requires 
determination of THCE or NMHCE for your engine, then see subpart I of 
this part for additional requirements.) Under the constant-volume 
sampler (CVS) concept, you must measure the total volume of the mixture 
of exhaust and dilution air and collect a continuously proportioned 
volume of sample for analysis. You must control flow rates so that the 
ratio of sample flow to CVS flow remains constant. You then determine 
the mass emissions from the sample concentration and total flow over the 
test period.
    (1) Do not let the CVS or dilution air inlet system artificially 
lower exhaust system backpressure. To verify proper backpressures, 
measure pressure in the raw exhaust immediately upstream of the inlet to 
the CVS. Continuously measure and compare the static pressure of the raw 
exhaust observed during a transient cycle--with and without the CVS 
operating. Static pressure measured with the CVS system operating must 
remain within 5 inches of water (1.2 kPa) of the 
static pressure measured when disconnected from the CVS, at identical 
moments in the test cycle. (Note: We will use sampling systems that can 
maintain the static pressure to within 1 inch of 
water (0.25 kPa) if your written request shows that this closer 
tolerance is necessary.) This requirement serves as a design 
specification for the CVS/dilution air inlet system, and should be 
performed as often as good engineering practice dictates (for example, 
after installing an uncharacterized CVS, adding an unknown inlet 
restriction on the dilution air, or otherwise altering the system).
    (2) The system for measuring temperature (sensors and readout) must 
have an accuracy and precision of 3.4[deg] F 
(1.9[deg] C). The temperature measuring system for 
a CVS without a heat exchanger must respond within 1.50 seconds to 62.5 
percent of a temperature change (as measured in hot silicone oil). For a 
CVS with a heat exchanger, there is no specific requirement for response 
time.
    (3) The system for measuring pressure (sensors and readout) must 
have an accuracy and precision of 3 mm Hg (0.4 
kPa).
    (4) The flow capacity of the CVS must be large enough to keep water 
from condensing in the system. You may dehumidify the dilution air 
before it enters the CVS. You also may heat or cool the air if three 
conditions exist:
    (i) The air (or air plus exhaust gas) temperature does not exceed 
250[deg] F (121[deg] C).

[[Page 655]]

    (ii) You calculate the CVS flow rate necessary to prevent water 
condensation based on the lowest temperature in the CVS before sampling. 
(We recommend insulating the CVS system when you use heated dilution 
air.)
    (iii) The dilution ratio is high enough to prevent condensation in 
bag samples as they cool to room temperature.
    (5) Bags for collecting dilution air and exhaust samples must be big 
enough for samples to flow freely.
    (6) The general CVS sample system consists of a dilution air filter 
(optional) and mixing assembly, cyclone particulate separator 
(optional), a sample line for the bag sample or other sample lines a 
dilution tunnel, and associated valves and sensors for pressure and 
temperature. Except for the system to sample hydrocarbons from two-
stroke engines, the temperature of the sample lines must be more than 
3[deg] C above the mixture's maximum dew point and less than 121[deg] C. 
We recommend maintaining them at 113  8[deg] C. 
For the hydrocarbon sampling system with two-stroke engines, the 
temperature of the sample lines should be maintained at 191  11[deg] C. A general schematic of the SI sampling 
system is shown in Figure 1065.110-1, which follows:

[[Page 656]]

[GRAPHIC] [TIFF OMITTED] TR08NO02.017

    (b) Steady-state testing. Constant proportional sampling is required 
throughout transient testing, but is not required throughout steady-
state testing.

[[Page 657]]

Steady-state testing requires that you draw a proportional sample for 
each test mode, but you may sample in different proportions for 
different test modes, as long as you know the ratio of the sample flow 
to total flow during each test mode. This allowance means that you may 
use simpler flow control systems for steady-state testing than are shown 
in Figure 1065.110-1 of this section.
    (c) Configuration variations. Since various configurations can 
produce equivalent results, you need not conform exactly to the drawings 
in this subpart. You may use other components--such as instruments, 
valves, solenoids, pumps and switches--to provide more information and 
coordinate the components' functions. Based on good engineering 
judgment, you may exclude other components that are not needed to 
maintain accuracy on some systems.
    (d) CFV-CVS component description. The flow characteristics of a 
Critical-Flow Venturi, Constant-Volume Sampler (CFV-CVS) are governed by 
the principles of fluid dynamics associated with critical flow. The CFV 
system is commonly called a constant-volume system (CVS) even though the 
mass flow varies. More properly, they are constant-proportion sampling 
systems, because small CFVs in each of the sample lines maintains 
proportional sampling while temperatures vary. This CFV maintains the 
mixture's flow rate at choked flow, which is inversely proportional to 
the square root of the gas temperature, and the system computes the 
actual flow rate continuously. Because pressures and temperatures are 
the same at all venturi inlets, the sample volume is proportional to the 
total volume. The CFV-CVS sample system uses critical flow venturis for 
the bag sample or other sample lines (these are shown in the figure as 
flow control valves) and a critical flow venturi for the dilution 
tunnel. All venturis must be maintained at the same temperature.
    (e) EFC-CVS component description. The electronic flow control-CVS 
(EFC-CVS) system for sampling is identical to the CFV system described 
in paragraph (b) of this section, except that it adds electronic flow 
controllers (instead of sampling venturis), a subsonic venturi and an 
electronic flow controller for the CVS (instead of the critical flow 
venturi), metering valves, and separate flow meters (optional) to 
totalize sample flow volumes. The EFC sample system must conform to the 
following requirements:
    (1) The system must meet all the requirements in paragraph (b) of 
this section.
    (2) The ratio of sample flow to CVS flow must not vary by more than 
5 percent from the test's setpoint.
    (3) Sample flow totalizers must meet the accuracy specifications in 
Sec. 1065.150. You may obtain total volumes from the flow controllers, 
with our advance approval, if you can show they meet these accuracies.
    (f) Component description, PDP-CVS. The positive-displacement pump-
CVS (PDP-CVS) system for sampling is identical to the CFV system 
described in paragraph (b) of this section, except for the following 
changes:
    (1) Include a heat exchanger.
    (2) Use positive-displacement pumps for the CVS flow and sampling-
system flow. You do not need sampling venturis or a venturi for the 
dilution tunnel. All pumps must operate at a constant flow rate.
    (3) All pumps must operate at a nominally constant temperature. 
Maintain the gas mixture's temperature--measured at a point just ahead 
of the positive-displacement pump (and after the heat exchanger for the 
main CVS pump)--within 10[deg] F (5.6[deg] C) of the average operating temperature 
observed during the test. (You may estimate the average operating 
temperature from the temperatures observed during similar tests.) The 
system for measuring temperature (sensors and readout) must have an 
accuracy and precision of 3.4[deg] F (1.9[deg] C), 
and response time consistent with good engineering judgment.
    (g) Mixed systems. You may combine elements of paragraphs (d), (e), 
and (f) consistent with good engineering judgment. For example, you may 
control the CVS flow rate using a CFV, and control sample flow rates 
using electronic flow controllers.

[[Page 658]]