72-9
EVALUATION OF THE BERG EMISSION
CONTROL SYSTEM
January 1972
Henry L. Gorapf
Test and Evaluation Branch
Environmental Protection Agency
-------
Background
The Berg Emission Control System was initially evaluated in
Cincinnati, Ohio, by the National Air Pollution Control
Administration. That testing indicated substantial reductions
in hydrocarbon, and carbon monoxide. The validity of the
results was questioned, however, due to the use of smaller
metering jets and adjustment-of idle carbon monoxide to a
lean level of .5%. Gordon C. Berg, the developer, approached
the Environmental Protection Agency with a modified system
in September, 1971. It was decided, based upon the previous
testing and the system's potential applicability to used,
uncontrolled motor vehicles, to initiate an evaluation of
the modified system.
System
The Berg System performs two functions. The primary effect is
that of bleeding additional air to the intake manifold thus
leaning combustion. The second function is to supply fuel
vapors from the gasoline tank to the intake manifold. The
hardware required to accomplish this two-fold approach is
minimal. The PCV valve is removed leaving the line from the
crankcase to the carburetor in place. A plastic fitting topped
with a tee is inserted in this line. One side of the tee is
connected to a hose running to the fuel tank cap which has been
fitted with a small bore copper tube to allow flow-through of
gasoline vapors. The other side of the tee is left open to
atmosphere as a source of bleed air. The size of this orifice
can be varied depending on the desired bleed-air rate. The
effect of this system is to add a mixture of air and fuel
vapors to the PCV system. The flow through the PCV system
is no longer metered by the PCV valve.
Test Program
A 1963 Ford Galaxie was selected as the vehicle to be used in
the evaluation program. This vehicle was equipped with a 289
cubic inch displacement engine and had an automatic transmission.
Three different configurations of the Berg System were evaluated.
They differed each time through increasing the bleed-air rate
accomplished by enlarging the open side of the tee or adding
additional bleed holes to the base of the tee.
Baseline or stock vehicle testing was conducted before, during
and after the device testing. All tests were performed according
to the 1972 Federal Emission Test Procedure as described in the
November 10, 1970, Federal Register. This testing employed the
-------
2.
LA4-S4 driving cycle and the constant volume sampling technique
for obtaining bag samples. These samples were analyzed using
NDIR analysis for carbon monoxide and carbon dioxide, FID analysis
for hydrocarbons, and chemiluminescence analysis for oxides of
nitrogen.
Results
The actual test by test emission results are presented in the
Appendix of the report. The following table indicates the
percent reductions from baseline for each of the three con-
figurations of the Berg system that were tested.
Percent Reduction From Baseline .
1st Berg
2nd Berg
3rd Berg
This data reveals the effect of increasing bleed-air rates
on the emissions. The major effect of leaner combustion
is seen in the reduction of carbon monoxide with a smaller yet
progressively increasing reduction in hydrocarbon levels. The
increased availability of oxygen during the combustion process
result in higher temperatures and increased levels of oxides
of nitrogen. The vapor line system was not modified once it
was installed. It can be predicted that this added fuel effect
was offset by add-ition of larger quantities of bleed air.
Because of the vapor line connection, analysis of potential
safety hazards should be made. Simple provisions for filtering
bleed air would eliminate the harmful effects of air-born dirt
to the engine.
Conclusions
HC
0%
13%
29%
CO
9%
. 34%
64%
C07
— — ..£i
-3%
-6%
-10%
inc
inc
inc
NOx
-23%
inc
The Berg System, in essence is a simple air bleed device.
Depending on the quantity of bleed air, the system reduces
hydrocarbon and carbon monoxide levels. Maximum reductions
of 29% and 64% respectively, were obtained during the evaluation,
The leaner combustion resulted in increased emissions of oxides
of nitrogen.
Since the system works on the principal of increasing the air
fuel ratio regardless of what that ratio was initially, it is
predictable that certain vehicles already operating near the
lean limit air-fuel ratio could suffer misfire, stumble and
-------
3.
surge. There is no estimate as to the proportion of the
in-use vehicles which would be affected in this manner.
It is, however, a consideration which deserves further
investigation.
-------
APPENDIX
-------
Baseline
1972 FEDERAL EMISSION TEST RESULTS
ON 1963 FORD
(All icesults are in grams per mile)
Test# Date HC CO
Average Initial
Baseline
1st Berg Approach
18-0059
18-0064
Average First
Approach
2nd Berg Approach
18-0067
18-0068
Average Second
Approach
Interum Baseline
18-0100
18-0100
18-0105
18-0106
Average Interum
Baseline
Average First and
Interum Baseline
7.7
7.5
91.8
91.9
CO-
11/3/71-
11/2/71
11/4/71
11/8/71
11/24/71
11/26/71
11/29/71
11/30/71
12/30/71
1/3/72
1/4/72
1/5/72
7.1
7.3
7.1
7.6
7.3
6.7
8.2
7.5
7.0
5.9
6.5
8.2
7.9
7.3
7.2
93.9
83.4
94.9
95.8
92.0
80.7
86.1
83.4
64.2
56.6
60.4
114.1
82.5
83.5
87.1
450.5
423.7
455.7
439.0
442.2
455.3
476.5
465.9
478.5
479.11
478.8
413.0
484.3
480.6
460.6
459.6
450.9
NOx
-------
This series of testing is treated separately as a different
analysis train was employed with inherently different test
errors.
Test# Date HC CO C09 NOx
12-2084
12-2089
12-2096
12-2094
1/7/72
1/11/72
1/13/72
1/12/72
5.9
5.7
5.6
5.8
37.9
36.8
36.8
42.8
598.8
563.8
559.3
565.5
5.8
5.7
4.5
5.4
3rd Berg
Approach
Average 3rd Berg
Approach. 5.8 38.6 571.9 5.4
Final 12-2101 1/17/72 8.9 112.5 504.8 4.1
Baseline 12-2104 1/18/72 8.7 118.8 563.7 4.7
12-2108 1/19/72 7.0 97.8 521.4 4.1
12-2114 1/20/72 8.3 105.3 487.4 4.7
Average Final
Baseline 8.2 108.6 519.3 4.4
------- |