The 2002 International Congress and Exposition on Noise Control Engineering
Dearborn, MI, USA. August 19-21, 2002
The Pros and Cons of Valves in
Automotive Exhaust Systems
William E. Hill
Tenneco Automotive, Grass Lake Engineering Center
Grass Lake, Michigan 49240
Abstract
In conventional exhaust systems the muffler/resonator volume and restriction control the
average insertion loss or noise reduction of low frequencies. Broadband improvements in low
frequency noise reduction can be achieved with al rger mufflers or smaller pipes or flow tubes.
However larger mufflers increase cost and packaging problems and smaller pipes reduce
engine performance. The use of dual mode mufflers or exhaust systems is increasing slowly.
These systems incorporate valves that reduce low frequency noise at low engine speeds (a
major tuning issue) but have lower restriction at high engine speeds compared to conventional
systems. A benchmark study of production dual mode exhaust systems was conducted to
determine their benefits and limitations. Two groups of valve systems were studied: spring-
loaded valves that are exhaust flow controlled and externally operated valves that are
controlled by the vehicle ECM or the driver. The benchmarked systems were tested on one
test vehicle in order to consistently compare hot backpressure and tailpipe emitted noise
differences. The summarized results show that spring-loaded valves still have tradeoffs with
restriction and noise control while externally-operated valves tradeoff cost and noise control.
1. Introduction
Typical passive muffler development is challenging because the customer requirements can
seem to be conflicting. Most vehicle manufacturers want low exhaust backpressure to
maximize engine power and low levels of exhaust noise. However, the design variables that
reduce backpressure (large smooth pipes, minimum flow direction changes) tend to increase
exhaust noise radiated from the tailpipe. One approach to meet both of these objectives is to
incorporate a valve in the exhaust system that permits two tuning modes – high restriction
under low flow conditions to control low frequency noise and low restriction under high flow
conditions to minimize exhaust pressure drop. The premise is that exhaust noise control is
most demanding at low vehicle speeds when background noise is low. As vehicle speed
increases, higher levels of exhaust noise can be tolerated because the background noise levels
are higher. Historically the aftermarket has had various valves for exhaust systems to provide
driver-selectable horsepower and noise. For original equipment exhaust systems however,
valves have only been sporadically used over the years although they have become more
common in the last decade. A benchmark study of current mufflers with valves was
conducted to determine the acoustic and backpressure benefits and drawbacks of these tuning
elements. Valve durability was beyond the scope of this study.
2. Test Samples and Valve Types
A survey of valve muffler applications in 1999 found 35 different vehicles with some models
having exhaust systems with valves back to 1990. Of these 35 applications, 31 were Japanese
and 4 were European vehicles. Several of the vehicles were sold only in Japan or Europe.
Samples were obtained by purchasing OE service replacement parts from the respective auto
dealers. From the list of 35 applications, 9 exhaust systems were purchased for benchmark
testing.
There were two main types of valves used in these exhaust systems. The first is a spring-
loaded valve (ref. Fig 6) that opens up in response to exhaust pressure or flow. This type can
be completely contained inside the muffler or some components of the valve may be external
to the muffler. The second type is the externally actuated valve (ref. Fig 7) that is opened or
closed based on the engine management program or by the driver. The externally actuated
valves in this study were either operated by vacuum or with a motor and cable arrangement.
Of the 35 valve muffler systems surveyed, 25 had spring-loaded valves and 10 had externally-
actuated valves. Of the 9 systems evaluated in this study, 6 had spring-loaded valves, 3 had
externally-operated valves.
3. Basic Dual Mode Muffler Designs
Spring-Loaded Valves – The basic spring-loaded valve muffler design has a restrictive flow
path that blocks or reflects low frequency noise. The valve is mounted on a tube that provides
a parallel, less restrictive flow path. When the pressure drop across the valve is large enough
to overcome the spring preload, the valve gradually opens. The six benchmarked mufflers
were all three-pass designs with a restrictive path through small diameter tubes or perorations
and a valve mounted on a larger diameter return tube. This valve location may minimize the
valve and spring exposure to exhaust heat. Some designs took extra precautions and located
the spring external to the muffler. Other design differences include mesh valve seats and
weights to reduce buzzing or rattling.
Figure 1: Benchmark mufflers with spring-loaded valves
Externally-Operated Valves – In the three benchmarked designs, all valves were mounted on
pipes external to the mufflers. Unlike the spring-loaded valves, it appears that these valves
are intended to be either open or closed with no intermediate positions. In two of the
mufflers, the less restrictive, valve-controlled flow path is a second outlet pipe (dual spouts
versus a single spout when the valve is closed). The third muffler allows the exhaust gas to
pass straight through the muffler when the valve opens providing minimal pressure drop.
Figure 2: Benchmark mufflers with externally-operated valves
4. Evaluation Methods
The valve exhaust systems were tested to determine restriction and the insertion loss of the
valve. The mufflers with spring-loaded valves were modified so tests could be run with the
valve held completely open, completely closed or valve normal (allowed to open as designed
with exhaust flow). The externally-operated valves were manually operated to be completely
open or completely closed. Evaluating the valve mufflers on the original vehicles was not
practical plus comparing performance data would be difficult. Instead, all tests were run on a
company owned test vehicle, a 1995 3.2L V-6 Acura Legend. This vehicle was viewed as a
good compromise since seven of the nine valve systems were 6 cylinder applications (2.5L to
3.2L displacements). The other two applications were a 4.7L V -8 and a 1.6L L -4.
Figure 3 shows the test set-up with the Acura Legend.
Part-throttle sweeps from idle to 5500 rpm were run on a
chassis dynamometer with a typical road load setting.
Tailpipe exhaust gas temperature was monitored and each
valve muffler & spout
run was started at 250C after a warm-up period. Exhaust
noise was measured with a microphone located .5m and 45
degrees from the outlet. Static backpressure was measured
thermocouple
upstream of the muffler during the sweeps.
& BP taps
Judging acoustic performance by comparing radiated
tailpipe noise would be unfair since these mufflers had a
range of sizes of 13 to 28 liters. Since the benchmark
Figure 3: Valve muffler on
objective was to evaluate the effectiveness of the valve,
test vehicle
valve insertion loss was calculated. The noise level with
the valve open was the noise baseline for each muffler. The tailpipe noise with the valve
closed or in normal operation was subtracted from the valve open noise to show the noise
difference the valve made. Comparisons were made for overall noise and for the dominant
3.0 exhaust noise order.
5. Test Results
By measuring backpressure with the spring-loaded valves in open/closed/normal positions, it
was possible to note the engine speed when the valve began to open (see Fig 4 and Table 1).
SL3 Valve Muffler - Hot BP Results EO1 Valve Muffler - Hot BP Results
Tested on Acura Legend V-6 Tested on Acura Legend V-6
30 30
25 valve closed 25
Static Pressure (kPa)
Static Pressure (kPa)
valve closed
20 20
Valve begins valve normal
15 To open 15
10 10
5 5
valve open valve open
0 0
1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
Engine Speed (rpm) Engine Speed (rpm)
Figure 4: Typical hot backpressure curves for spring-loaded (SL) and externally-
operated (EO) valve mufflers
To measure acoustic performance of the valves, valve insertion loss was calculated by
subtracting the sound pressure levels of the dominant 3.0 order with the valve in normal
(spring-loaded) or closed (externally-operated) position from the 3.0 order sound pressure
level with the valve open. Table 1 lists the average insertion loss of the 3.0 order noise over
the rpm range of 1100-5500 rpm (55-275 Hz).
Figure 5: Valve insertion loss examples (3.0 order).
SL=spring-loaded, EO=externally-operated.
In general, with the spring-loaded valve mufflers, the valve insertion loss was positive at
frequencies below 150 Hz (3000 rpm). As the valve opened, the insertion loss would drop to
zero or sometimes negative levels. There was a similar trend for the externally-operated valve
mufflers except for EO1. This muffler had little attenuation with the valve open and so was
quite loud. Closing the valve reduced the exhaust noise over the entire test range.
The spring-loaded valve mufflers with the highest restrictions with the valve working
normally had the highest insertion losses. The mufflers with the lowest valve normal
restrictions had the lowest insertion losses.
Four of the six spring-loaded mufflers showed evidence of flow noise generation when the
valves were in normal operation. Flow noise was also noted on externally-operated valves
when they were held in intermediate positions during bench flow tests.
Table 1: Summary of valve muffler test results
Avg IL * Valve
No. of valve
Muffler # 3.0 Order Hot Backpressure kPa @ 5000 rpm begins to
components**
55-275 Hz open (rpm)
SL1 7.9 65 17 11 8 2000
SL2 5.2 48 35 16 53 4500
SL3 3.3 25 13 11 24 2700
SL4 1.3 37 24 20 11 2300
SL5 1.2 14 8 6 6 3500
SL6 -7.2 12 9 7 8 4000
EO1 16.3 20 - 1 15
EO2 5.6 25 - 8 22
EO3 3.2 14 - 9 12
* Average insertion loss: valve normal vs. valve open for spring-loaded, valve closed vs.
valve open for externally-operated.
** Valve components only, does not include actuation hardware for externally-operated
valves (motors, switches, wiring, cables, vacuum hoses, accumulators, etc.)
Figure 6: Examples of spring -loaded valves
6. Additional Exhaust Valve Example
Although not included in the original valve muffler benchmark study, one additional
application deserves mentioning. One European manufacturer sells vehicles with dual mode
engines that can deactivate certain cylinders under light load conditions for fuel savings.
Because it is difficult to design a conventional exhaust system that will meet the objectives for
both engine modes, an externally-operated valve was added. In this case, the valve does not
provide an alternate flow path but instead acts as a variable restrictor. In the half-cylinder
mode the valve is closed but it does not block the pipe completely which helps reduce low
frequency noise. Because this valve works as a restrictor, it will perform best when it is
located at points of high acoustic velocity.
Figure 7: Example of an externally-operated valve
7. Conclusions & Recommendations
Use a dual mode muffler when package volume is limited. Non-valve conventional mufflers
rely on volume and/or restriction to control low frequency noise. A properly designed muffler
with a valve can act as a larger volume muffler with good low frequency attenuation at low
engine speeds and reduced restriction at high flow rates for less power loss.
Spring-loaded valve mufflers trade off low frequency attenuation with backpressure. The best
noise control requires a restrictive valve-closed flow path and a high spring preload that
results in higher overall restriction. If the valve-closed path is enlarged to reduce restriction,
then low frequency noise control is sacrificed, same as in conventional mufflers. If the spring
preload is reduced, the valve may open at an engine speed that is too low to be effective.
If the design goal is minimum muffler restriction, use an externally-operated valve system.
The drawback of course is complexity and cost.
If the design goal is to switch the dual mode muffler operation at specific engine speeds or
other points (e.g. cylinder deactivation), then an externally-operated valve is the primary
choice.
Acknowledgements
Special thanks to the co-op students who did all of the dirty work – Matthew Stalker, John
Warmenhoven, and Jonathon Wenk.
