Julian Edgar, 50, has been writing about car modification and automotive technology for nearly 25 years. He has owned cars with two, three, four, five, six and eight cylinders; single turbo, twin turbo, supercharged, diesel and hybrid electric drivelines. He lists his transport interests as turbocharging, aerodynamics, suspension design and human-powered vehicles.
Muffler Design
While car modifiers talk endlessly about exhaust systems, there is surprisingly little hard information about the design of exhausts, especially mufflers.
That’s why I was particularly interested to come across the text that follows, contained in – of all things – a book on corrosion of cars, published exactly 50 years ago. The chapter, by H. Silman of Electro-Chemical Engineering, is a great summary of exhaust muffler design.
The automobile exhaust system consists essentially of three parts:
(1) the exhaust manifold, which collects the discharged gases from the exhaust ports and conveys them by means of a pipe at least one-quarter of the diameter of the cylinder to the silencer,
(2) the silencer, and
(3) the tail pipe, which leads the gases to the rear, the side, or more rarely to the top of the vehicle.
Sometimes twin parallel exhaust systems are used, especially with multi-cylinder engines, whilst it is also becoming increasingly common for two silencers in series to be fitted, the second and shorter unit being located immediately before the final outlet.
The exhaust gases leave the cylinders of an automobile engine at a pressure of around 60-80 lb. per sq. in. and with a velocity of up to 150 ft. per sec. This results in a considerable volume of noise, which must be reduced sufficiently to make the vehicle inoffensive to the driver and passengers, and to the public at large. This legal requirement is achieved by allowing the gases to expand into a silencer chamber, where the intermittent and violent discharges of gas are broken up and emerge from the tail pipe as a continuous and relatively uninterrupted stream.
There are various designs of silencer, but they usually contain a number of baffles carefully designed to avoid excessive back pressure, which might result in a loss of power and overheating of the engine. There is always some loss of power resulting from the use of a silencer, but this need not exceed about 3% with a well-designed unit.
Typical designs of silencer systems are based on a capacity of 12-14 cu. in. per b.h.p. at maximum value. Weight is very important, and a silencer should not exceed 1/10 lb. per b.h.p. (max.) of the engine. As almost all engines are today flexibly mounted, the exhaust pipe and silencer system are also attached to the vehicle by a flexible support such as a woven asbestos [! – JE] strap to prevent restraint of the engine movements and to reduce noise transmission.
The exhaust note of an internal combustion engine is made up mainly of a low- to medium-frequency band of 50-600 cycles per second, and a high-frequency band of 3000-10,000 cycles per second.
The low-frequency vibrations appear to result from resonance between the varying capacity of the cylinder and the exhaust system as the piston moves through its exhaust stroke, while the high-pitch sound originates in the release of the high-pressure gases through the exhaust ports.
The usual type of silencer therefore depends for its action on its capacity, and the damping effect of the large change in section as the gases enter it, the sound waves being absorbed in the turbulent areas where the changes of section occur. Such silencers are most effective in dealing with the lower-frequency noise bands. Higher-frequency sound can be absorbed by a silencer consisting of a perforated tube surrounded by a casing containing a material such as glass or slag wool, but this system is ineffective in dealing with the lower-frequency band and is therefore not greatly used, particularly as it is also relatively costly.
A combination of units of both these types would constitute a good silencing system, but space, cost and weight considerations militate against the widespread adoption of such a scheme.
Various silencers have been designed in which an attempt has been made to reduce the noise volume by dividing the gases into two streams each of which follows a path differing from the other by half a wavelength. Such a silencer will of course be fully effective only at a single wavelength; as there is a considerable wavelength spread in the exhaust note, the theoretical advantages of the method are not realized in practice.
I think that the points made about required volume, and the way in which silencing of different frequencies occurs, are as relevant today as when they were first written.
on March 5th, 2008 at 9:52 am
The combined silencer system that the author mentions is pretty close to the exhaust systems in use (as a performance mod) on large displacement engines, such as the Ford XR6 Turbo/FPV Typhoon and FPV GT. So yeah, it’s still very relevant
on March 7th, 2008 at 2:56 pm
Julian,
David Vizard is considered one of the “gurus” of exhaust design .He did quite a bit of experimental work in the USA several years ago on the effects of exhaust length and silencer type on power output primarily on a 350 Chev, but a lot of his findings have validity for other engines. http://www.geocities.com/motorcity/track/6992/vizard.html
Vizard did work on many types of engines and a Google of “Vizard exhaust” will find many references.
regards Philip A
on March 7th, 2008 at 7:55 pm
According to the article, muffler volume changes effect the ‘low’ frequency sounds, and 14ci is required for every Max hp. ths works out at 23 litres muffler capacity for every 100hp. As such a commodore or falcon would therefore require 60 litres of mufflers capacity for this rule of thumb.(not including the spce required for any glassfibres etc). This seems to me to be a large space to find under a car, and i am sure that most car mufflers do not have that capacity. Just from looking at the mufflers, it seems to me that ‘luxury’ cars like lexus etc that are quiet have really large mufflers whereas ‘sporty’ cars that are noisier like porsche a have smaller muffler capacity. So although the capacity of a lexus muffler def is not 60+ litres it appears that the larger the muffler capacity the greater the noise drop. I wonder if the relation is linear (i.e if i add 1 more muffler to my existing 2 muffler system the noise will be 2/3 of the existing, or if pipe characteristics or pressures play a role. i am sure this could be worked out mathmatically and by experimentation?
thanks . Craig
on March 8th, 2008 at 8:57 am
My Lexus LS400, with two main rear mufflers and others further forward (include the cat converters as they also act as mufflers) would have had a total volume (I’d guess) of ~60 litres. Incidentally, it’s the volume as a whole of the muffler, so just the external measurements minus wall thickneses.
In the past when we’ve tested mufflers, with all other aspects being the same, the larger their volume, the better they work at muffling.
on March 10th, 2008 at 2:50 pm
Hi guys, chiming in as I found this article researching rear diffusers and also exhaust headers. I am planning on embarking on a design journey to design a new muffler. What occured to me is that volume itself provides a method to expand and damp the energy put forth, but what about selective band pass filtering to a particular exhasut note. The flow master is the best that I have seen attempting to actually create and acoustic chamber as well as providing a pressure vessel that is not restrictive. Why are there not more acoustic designs out there? What conventions have lead us to an oval? The ease of manufacturing? I found the viazrd articles a great read and I appreciate the post. I look forward to experiment on my I6 IS300.
Julian, if you ever need some replacemen thood props, give me a ring.
Mike Figaro
on March 11th, 2008 at 8:43 pm
I’ve thought for some time that there is an overlap in basic requirements for exhaust muffling and crash protection.
As we’ve read here, a muffler needs volume, filled with not much at all. A crumple zone needs volume, filled with a bit of crushable structure and not much else.
The bumper bar (or, these days, colour-coded moulded bumper panel) and rear crumple zone of a car are pretty much wasted space. You can’t put the fuel tank there any more, and rear seats (in hatches, third row for wagons and SUVs) have to be set far enough inboard that they won’t get crushed in a rear-ender.
Why not use that space – the under-floor space especially – for an effective large-volume muffler? Why do we waste useful underfloor space further forward with big metal box full of crushable baffles and loosely packed rock-wool? A straight-through exhaust pipe takes much less space than a muffler… the remainder could be used to accommodate deeper foot-wells, under-floor storage compartments, fold-away seats as seen in the (sadly orphaned) Holden Zafira… dare I say, a full-size spare wheel?
Forward-mounted, longitudinally-aligned mufflers are just a silly packaging choice, as far as I can see.
Another crazy idea – has anybody ever played with a Helmholtz resonator muffler system? Helmholtz resonators – effectively an tuned cavity and neck hanging off the side of a pipe or duct – are used extensively on engine intake systems, to eliminate problem frequencies from the induction sound. Exhaust noise is also somewhat tonal… why not hang a set of tuned resonators off the exhaust pipe to attenuate the noise? Again, all it takes is some empty air space, at the back of the car where you need some crumple zones anyway…
on March 18th, 2008 at 11:18 pm
Tim,
I love you idea of using the bumbers as chambers! But the notion of under the floor would be dicey just due to the heat. There is great airflow where the are traditionally mounted, and that airflow is (I’m Guessing) a large part of the reason as to why they are there.
on April 9th, 2008 at 12:35 am
Talking about making your own muffler, I’ve done it. I read some books, look at designs and found nothing like what I made.
The car is my 105 Alfa race car with very tuned 239hp 2L motor. We planed to do 3 tests on the dyno changing from the collector back only.
Test 1
The exhaust consisted of a 2” stainless pipe of about 30 inches into a divergence nozzle expanding to a 3” x 14” straight trough type muffler into a convergence nozzle reducing back down to 2’and then a pipe of about 60 inches with a 30-40deg mandrel bend in the middle. These lengths are not optimized as we just had this stuff lying around This ends behind the front door so the whole system is very short and sounds like a very cranky cat, not very pleasant at all.
Test 2
No exhaust system at all
Test 3
My exhaust system consisted of, well if you could imagine a banana shape with a 2 inch inlet and a 7×3 oval outlet. The internal lining was perforated aluminum shaped to form an oval tapered, curved pipe. The outer covering was made oval and curved from high heat epoxy and fiberglass. The muffling material was just fiber glass wool packed into the void so at the inlet end there was 2 or so inches of wadding and next to nothing at the outlet. It gave a great deep meaty sound which would have been expected due to the density of the materials used in the pipe.
So how did they perform?
Test 1 recorded 239hp
Test 2 added 9 hp but lost torque at 4000 (max torque is at 6000rpm)
Test 3 added 4hp no loss of torque, but me, not being a muffler expert had durability issues. I like every one else thought the epoxy would break down from the heat but the problem was the perforated aluminum expanding and buckling in.
Anyway it was educational.
on November 17th, 2008 at 12:17 pm
The idea of huge rear mufflers being used as bumpers and crumple zones has already been put into mass production in the US market, so obviously you guys are on a winner! It’s only been done on very pedestrian cars though, nothing with any pretence to high performance.
on December 10th, 2008 at 6:32 pm
Hello Julian! I am thinking to build “tailor made” mufflers to serve my customers. Do you recommend any handbook for muffler design? Thank you.
on December 10th, 2008 at 7:08 pm
I don’t know of any such book. Sorry.
on December 22nd, 2008 at 9:20 am
Does anyone happen to have a copy of that David Vizard exhaust tuning article linked above?
I have managed to read all of it except page 5. If anyone can forward the ‘vizard5.jpg’ image I’d really appreciate it.
bonkoh.2007@gmail.com
on August 10th, 2010 at 6:57 pm
I’m a bit confused. Is the silencer and the center muffler the same? Mufflers are the ones which reduce the exhaust noise, right?
Thanks