Backpressure

[keithdog72]

Just curious, what are the effects of low backpressure? High? I am assuming of course that exhaust backpressure and quality of performance are directly related. How much bp is too much? how little is too little?

[masraum]

Well, in the most general of circumstances, less is better and more is worse, but like anything else complex on our cars it's just not that simple.

For a race engine or an engine that is in a very high state of tune that is setup to develop peak power at a high rpm and will likely be kept at or around that high rpm most of the time then you want very little to no backpressure, but as racey as our cars are that doesn't quite apply.

We need some backpressure for a scavenging affect in the exhaust, something about reversion and resonance causing the exhaust to suck the gases from the cylinder or something like that. Most of our engines are more like the above mentioned race motors than most other engines on the street, but if you make the backpressure too low you hurt low rpm torque which is as important or more than peak hp for a street motor. It's a very fine balancing act that very few people get well enough to make educated decisions about. I know I couldn't, pretty much all I can do is go on what the experts say, or if I had ton's of money I could spend lots of dyno time looking at power curves and figuring the area under the curve to see what works best for my specific circumstance.

Basically you want low backpressure, but you still do want some.

[Bill Verburg]

We need some backpressure for a scavenging affect in the exhaust, something about reversion and resonance causing the exhaust to suck the gases from the cylinder or something like that

Huh? Scavangeing effect from an exhaust is totally dependant on pipe length, width, rpm, exhaust temp, cam timing events. It has nothing to do w/ backpressure. Ideal in any engine is a vacuum at the end of the pipe(-1bar). In the real world, w/ real air, in real pipes negative pressure can only be achieved intermittantly over a relatively narrow rpm range by having the proper length and width pipes that create a resonance(standing wave in the pipe) at the proper rpm this standing wave will present a trough at the exhaust port(-pressure) which helps to suck the spent charge from the cylinder and draw fresh charge in(if there is enough overlap in the valve events). For every rpm range where there is - pressure at the exhaust port there is a corresponding point where there is + pressure(crest) at the exhaust port(bad)

The most efficient geometry for this is to have all the pipes equal in length and width and merge in such a manner as to speed the exhaust gas out in its expansion and cooling and to allow for the creation of these standing waves in the pipes w/o interference.

interference patterns come from sharp transitions of cross section(as in an abrubt transition in pipe dimeter)reflections from surfaces(as in a muffler) and in general any obstacle to the free managed and yet smooth expansion of the exhaust pulse.

Back pressure is always bad from a performance perspective, yet necessary for noise abatement. There are methods of noise abatements such as the use of Helmholtz resonators, wave canceling internal reflections etc that are somewhat effective at reducing noise w/o commensurate increase in back pressure.

[Randy Webb]

"something about reversion"....

Contrary to what Bill V. posted, scavenging is enhanced at certain rpm by pressure wave superposition. This is described in Phil Smith's intake & exhaust book, and there will be diagrams in an elementary physics text.

In essence, the low pressure on the wave from one cylinder can be used to "pull" the exhaust from an adjacent cylinder. Smith's book gives photos of the old timey exhausts that did not do this, and then shows newer (i.e. 1950's) exhausts where this benefits hp. he may have been the first person to measure these effects - he used a (now) funky manometer setup (similar to the way F-car mechanics tuned the multiple carbs on those cars). Today we would just use a series of peizo pressure transducers connected to a multi-channel data-logger.

This is not a function of nose abatement, necessarily. It is a function of using the rarefaction point of a pressure wave -- as a concommitant of that, there also exist compression points on the wave. so backpressure is not necessarily bad.

The main problem is in thinking in terms of statics or an equilibirum condition -- a steady-state. But the engine is a dynamical system, and you will gain by thinking in terms of transient phenomena.

"most efficient geometry ... is to have all the pipes equal in length" - this true if one is tuning for a peaky engine with max. hp. If one ones more "strettability" -- i.e. a broader torque band, then a 4-2-1 setup is better.

Clear as mud? If this makes too much sense, then start thinking about driven and damped harmonic oscillators....

[masraum]

UUmm, yeah, see... what they said.

Keith, aren't you glad you asked?

[keithdog72]

Didn't realize that I had to be a physicist to understand the relationship between backpressure and perfomance. You guys are very informative, but you are giving me a headache It's amazing to hear the wealth of knowledge that is applied to this board. I appreciate the explaination. Thanks,
Keith

[Randy Webb]

There are some very highly educated people here - quite a few chemists and various engineers. There are probably more people in "electrical" type fields than when these cars were new and fewer in mechanical type fields - but there was no Internet then to connect everyone.

re Physics - yup it pops up everywhere, and our everyday "commonsense" doesn't alwyas lead us to the right or best solution.

Anyways, Smith's photos are worth it for everyone, even those who don't want to read the text. You can ILL the book from most public libraries.

[tsuter]

For a turbo - no exhaust back pressure in the muffler is best. And who cares what it sounds like???

For an NA car I say lower back pressure steals TQ and adds HP so "best" becomes a different standard depending on what you are trying to accomplish.

[slakjaw]

anything ove about 1.5 PSI is too much for a street car.

[Bill Verburg]

Contrary to what Bill V. posted, scavenging is enhanced at certain rpm by pressure wave superposition. This is described in Phil Smith's intake & exhaust book, and there will be diagrams in an elementary physics text.

Once again Randy, I suspect that you do not read my posts or you do not understand the implications of what I wrote, because that is exactly what I said.

Here is the graph from Smith's book showing the effect of backpressure on MPEP hence hp at various atmosphereic pressures

There are 2 phenomena going on at the same time in the exhaust system
1) the hot expended gas is expanding and cooling down the pipe
2) acoustic waves are moving in both directions in the pipes

Both of these can be used to enhace the overlap scavange properties where a low pressure region is present at the exhaust port while it is open, the low pressure at the port sucks additional residual gas from the combustion chamber. If the intake port is open at the same time, a simultaneous enhancement of the filling of the cylinder w/ fresh combustible mixture occurs. This type of event is severly curtailed in modern smog engines because some of the unburned fresh mixture is drawn right through the chamber into the exhaust contributing to air pollution.

Here is an example of the pressure readings caused by both phenomena w/ 4 different pipe lingths at a 2000rpm.



I agree Smith's book is interesting if a bit dated

[Randy Webb]

If it is exactly what you said, then I surely did not grasp your implications.

But thanks for posting his graphs.

[banjomike]

I'll try to simplify (and probably oversimplify) what has been said.
Lets take one side of the 911 engine. There are three cylinder with three pipes coming off the exhaust valves. An efficient design would be three equal lengths that all meet at one point. This is what the early header (and SSIs) have. Now the big question is what length, and diameter? Imagine a cylinder full of hot gasses pumping out of each cylinder in turn. If the length of the pipes is set wrong then the exhaust gas from one cylinder is going to bump into the exhaust gas from the next cylinder and you end up with 'traffic jam'. If you set the length so that the gas from one cylinder it just finishing passing the intersection when the gas from the next cylinder is getting to the intersection then 'traffic' flows a lot better. Now take it a bit farther and set the length so that the gas from one cylinder has passed the intersection and the momentum of that gas is pulling a bit of a vacuum as the exhaust from the next cylinder is approaching the intersection then the momentum of the system is helping pull the exhaust gas out of the cylinders and you have a tuned exhaust!

Of course at different rpms, temperatures, etc. the ideal length will be different. So the designers compromise. If it's a race car they optimize it for high end power. If it is a dodge truck they optimize it for low end pull.

A similar set of circumstances rule the design of intake manifolds which is why Porsche designed the varioram. They vary the length of the intake manifold with the rpm so it can be optimized for multiple rmp ranges. Wo why don't they have a vario-exhaust?

[tsuter]

LOL.....
For you enginerds and theoreticians behind your desks, if....

"Back pressure is always bad from a performance perspective"


then.....

Why do my VERY expensive matched Harley race pipes have adjustable baffles???

[Kahuna]

Why do my VERY expensive matched Harley race pipes have adjustable baffles???

To give you something to play with.

[tsuter]

I thought the response would be " to adjust the noise"!!!
And that is a secondary effect that some may see as the primary effect.

[Bill Verburg]

Why do my VERY expensive matched Harley race pipes have adjustable baffles???

By adjusting the baffles you are changing the effective length of the pipes, which moves the compression and rarifaction up or down the torque curve changing where the max torque is in the rpm range.

Look at chart 2 it shows what is happening w/ different length pipes.

In general longer tubes move the torque down and shorter tubes move it up the rev range.

[tsuter]

And changing the "effective" length of the pipes has nothing to do with back pressure right? Who would want to raise or lower that right??

Because....

"Back pressure is always bad from a performance perspective"

Right, Slick?? LOL!!!!

[Randy Webb]

ts - no reason to attack -- also you don't know how he was defining back pressure.

[tsuter]

Originally posted by keithdog72
Just curious, what are the effects of low backpressure? High? I am assuming of course that exhaust backpressure and quality of performance are directly related. How much bp is too much? how little is too little?

He was asking about EXHAUST BACK PRESSURE and he got a bunch of gobblety gook for an answer.

Too many people want to expound on some theory and either can't read the question or don't care. That's all.

I got another guy on another thread saying that if one of your six spark plug washers is a different type of metal from the other five it will adversely affect the hot cold properties of that plug and adversely affect ignition/detonation!!! I just think he's been riding a desk and thinking waaaayy too long!!!

This is supposed to be a technical forum, so if you want to opine on what is the "best" paint for the fan fine.... but not exhaust back pressure. Statements like. ..."Back pressure is always bad from a performance perspective" are ridiculuous.

Exhaust BP just another variable for an NA car that is carefully managed to achieve TQ and HP objectives determined from how the vehicle is expected to be operated.

Some folks - masraum- even said this but got criticized by linguistic contortionists trying to practice automotive engineering!

[Randy Webb]

Hey - aren't you the guy who wanted every stmt documented? Bill V. may have been unclear and may also made an incorrect stmt., but he has documented more technical items on this forum than perhaps anyone else. I also find Sherwood's stmts. to be very careful and useful. You, however, have not documented anything that I have seen in this, or the spark plug thread. As a matter of fact, I suspect an extra crush washer might well affect heat transfer from the plug (and I have more expertise on heat transfer than most -- do a search).

But what I find most objectionable is the tone of your posts. The forum is here to be a conversation and to exchange information. Lorenfb is able to point out the lack of data with a bit more courtesy. Search for his posts.