Back pressure?

[BK911]

I remember reading several threads about how MFI likes back pressure. An exhaust system will have a pressure drop from the valve to atmosphere. And there will be a pressure wave bouncing back and forth. But WTF is back pressure? Increasing the pressure drop in the exhaust? Why go with a free flowing exhaust just to add restriction somewhere else?

Why would any engine want a positive pressure at the exhaust valve when it opens? You can tune an exhaust to have a pressure wave bounce off the exhaust valve right before it opens at a certain RPM to create a low pressure area at the exhaust valve. You can also tune the runners to use one cylinders exhaust to create a vacuum at another cylinder's exhaust valve to help purge the cylinder. But again, WTF is back pressure?

Please edjamucate me.

[BK911]

Oh, and happy holidays to all!

[RWebb]

there are numerous threads on this already; but in brief:

you need to understand the physics of waves first, MFI does not "Like" back pressure - but all engines profit when [1] the low pressure pulse of adjacent cylinders are used to 'pull' exhaust out of other cylinders; and [2] those same pulses can if timed properly benefit even a single cylinder engine

a search on 'reversion' will edify

[BK911]

Thanks Randy. I am familar with those two principles, just confused on the term 'back pressure'. I went through about 5 pages of search results and kept reading about how different induction systems need back pressure but nothing on what back pressure is. Somehow I missed this thread:

The value of back pressure in exhaust

[boba]

"like" may be the incorrect description, but MFI is sensitive to back pressure. If you change the exhaust type on an MFI you will need to recalibrate to get correct AFR.

When I switched from stock to sport muffler the mixture went lean and we needed to adjust. Without understanding the need to make the correct balancing adjustments, I guess one could come to the incorrect conclusions that MFI "likes" back pressure.

[RWebb]

Steve may have something on his Rennsport site also.

the term is superposition - might be a wiki or something on it. you don't nee the equations for a conceptual understanding.

[BK911]

Originally Posted by boba

"like" may be the incorrect description, but MFI is sensitive to back pressure. If you change the exhaust type on an MFI you will need to recalibrate to get correct AFR.

When I switched from stock to sport muffler the mixture went lean and we needed to adjust. Without understanding the need to make the correct balancing adjustments, I guess one could come to the incorrect conclusions that MFI "likes" back pressure.

Most (all?) the threads I read discussed MFI needing or liking the back pressure. I just couldn't understand why it mattered. MFI (or any induction) being sensative to it makes much more sense. Well, I still don't like the term back pressure.

[Bill Verburg]

Originally Posted by BK911

I remember reading several threads about how MFI likes back pressure. An exhaust system will have a pressure drop from the valve to atmosphere. And there will be a pressure wave bouncing back and forth. But WTF is back pressure? Increasing the pressure drop in the exhaust? Why go with a free flowing exhaust just to add restriction somewhere else?

Why would any engine want a positive pressure at the exhaust valve when it opens? You can tune an exhaust to have a pressure wave bounce off the exhaust valve right before it opens at a certain RPM to create a low pressure area at the exhaust valve. You can also tune the runners to use one cylinders exhaust to create a vacuum at another cylinder's exhaust valve to help purge the cylinder. But again, WTF is back pressure?

Please edjamucate me.

remember that there are 2 very different fluids flowing through an air cooled engine
1) air
2) gasoline

these have to be mixed in precise ways for best performance

It's not so much that the MFIs like back pressure it's just that the fuel delivery is very inflexible and is tuned for certain parameters of intake & exhaust performance. The further you deviate from expected values the worse the performance and less efective the band aids added on are going to be at optimising the mix.

[RWebb]

Yes, I don't like the phrase MFI likes back pressure either.

I think "back pressure" is a misnomer, or misunderstanding of wave superposition effects....

another search strategy wold to search on:
(muffler OR exhaust) + ("Bill Verburg" OR Steve Weiner")

Bill also posted a wonderful analogy a couple of weeks ago on fluid flow - little soldiers or something like that - it was in exhaust design context however.

some post long ago I gave the cite for an excellent early book on exhaust design - a British guy and I think he discovered the concept - or rediscovered it in IC engines. In the old days they just dumped everything right into a big chamber...

Good Luck on exploring the conceptual landscape...

[jluetjen]

Hi BK. I haven't been on the board for close to 9 months due to work commitments, and pop in today and this is the first thread that I see. I'm not sure if you remember, but I tried modeling this up about 5 years ago(!!!!) and in this thread. To make a long story short, I suspect that "backpressure" is a conventional wisdom term that attempts to describe in a linear fashion a function which is really frequency related. Yes, I'm know that there is a degree of back pressure associated with forcing a gas down a tube, but that is not really the phenomena that is associated with tuning an internal combustion engine unless the exhaust system is significantly undersized. Keep in mind that exhaust gasses will quite happily burn a hole through a gasket or a weak weld -- they really do want to get out of the cylinder pretty badly on their own accord.

Frequency functions are pretty difficult for laymen to understand -- including myself. So rather then deal with the complexity of organ pipe tuning of hot gasses (note that my earlier analysis didn't take into account the change in the speed of sound at elevated gas temperatures since I was looking specifically at intakes), most people say back pressure and assume that less is better. Given that Porsche used essentially the same exhaust system from the first 130 HP model up through the 210 HP 2.7 RS, in most cases I doubt that "back pressure" is the real obstacle to generating HP.

[boba]

Originally Posted by BK911

Well, I still don't like the term back pressure.

I agree, it is a short hand to reference a more or less "open" exhaust but it does not address what is taking place or how this effects the balance of intake and exhaust.

It could be categorized a jargon. Just think of all the business consultant speak only this is auto related.

[RWebb]

welcome back John!

[E Sully]

As I understand it cam timing and valve overlap combine with the exhaust waves to affect the amount of charge in the combustion chamber. At times the intake and exhaust valves are open at the same time during overlap. Wave pressure, or back pressure can prevent the incoming charge from passing straight through into the exhaust, and actually increase cylinder charging.

[BK911]

Originally Posted by E Sully

As I understand it cam timing and valve overlap combine with the exhaust waves to affect the amount of charge in the combustion chamber. At times the intake and exhaust valves are open at the same time during overlap. Wave pressure, or back pressure can prevent the incoming charge from passing straight through into the exhaust, and actually increase cylinder charging.

Umm, I think this is backwards. You want the pressure wave to bounce off of the exhaust valve right before it opens to 'suck' the exhaust out of the cylinder. This sucks the fuel mixture into the cylinder. Once the exhaust valve closes the inertia of the fuel mixture and the piston help fill the combustion chamber.

A pressure wave in the intake plenum can be used to ram air into the combustion chamber right before the intake valve closes. The longer the intake plenum the lower the rpm.

But most importantly, welcome back John!

[86911cab]

I am no porsche expert by no means, But I have played with chevy engines for some 20 odd years. All engines operate off the same principles. I have always believed that alittle back pressure is good. Too little and too much can affect performance. I may be wrong, but I think too little affects low end torque and too much affects performance across most of the power band.

High Back pressure slows down the flow of spent fuel/air mixture(leaving some in the cylinder). The right amout of pressure will aid in increasing the compression ratio inside the cylinder. Lift and duration of the cam also affects this too. Too much B/P will not allow all the burnt fuel to escape causing smaller amount of fresh fuel/air mixture to enter the chamber causing a weaker charge in the cylinder when it fires. In a minute amount High B/P can cause resistance against the piston as it tries to expell exhaust.

Have you ever tried driving a car with just headers to the muffler shop? Not only does it give you a headache, but it feels like you have a big fat wet sponge between your foot and the gas peddle. The engine has a very slow response.

I'd love to hear from the experts from the form about my theory. I welcome the criticisim.

[RWebb]

"The right amount of pressure will aid in increasing the compression ratio inside the cylinder."

- eh?

[86911cab]

By an extreemly small amount. Cams can to the came thing. I spoke to a cam company when I was building my truck engine, and what he told me made sense. Compression ratio may be the wrong term, cylinder pressure may be the better way to put it.

[RWebb]

so, you are saying that pressure acting backwards, up the exhaust port and back into the cylinder, will increase the cylinder pressure during the compression stroke?

if not then, when?

[jluetjen]

The reality is that you've got pressure waves and anti-pressure waves existing in both the intake and the exhaust. The trick is to get them to align (which generally happens across a narrow rev range), while keeping in mind that this potentially happens potentially at multiple places across the entire rev range. The reality is that general length and design (shape, plenum chambers, etc) of the intake will define one set of frequencies that it can be tuned to, and the length and design (mufflers, resonators, bunching configuration) of the exhaust will define a second set of frequencies. Not to mention you really want the intake to be tuned so that there is a pressure pulse over the intake valve when it's open during overlap, while you want the exhaust tuned so that there is a negative pulse over the exhaust when it's open. Ideally you want this to be a dynamic thing so that the intake charge that it pulled by the exhaust negative pulse get pushed (just) back into the cylinder while the piston is approaching peak acceleration just before the exhaust valve closes -- and while there is still a positive pressure on the intake side. Once you've got that figured out -- you have to add in the cam timing and static CR.

The result will be that as the engine rev's across it's rev range that at different times the intake, the cams and the exhausts will come "on tune". If you get them all aligned together, you'll have an engine that will come "on-cam" with a bang, but potentially you'll also have a very narrow rev range. Given the complexity of the problem, I suspect that in many cases the art of intake and exhaust tuning is to design out the worst cases (aka: reversion) so that the engine doesn't have any "holes" in the rev range or reversion. This will result in an engine that pulls from idle well and has a wide torque spread, which will generally result in a usable, strong performing engine.

PS: Running an engine with a incomplete exhaust system just throws all of the design engineer's work out the window, and introduces some fundamental problems such as a system which is too short, thus shifting the desired harmonics to a completely different part of the rev range.

[Bill Verburg]

back pressure builds in any muffled system because the total gas state(PVT) exiting is time delayed from the total gas state entering the system(PVT). It is measurable and depends on the restrictions imposed by the physical properties of the exhaust.

Superimposed on the above are the sonic pressure waves that headers seek to take advantage of. These are variable in frequency, timing and amplitude and consist of a series of compression and rarefactions that refect from the exhaust outlet back through the pipes to the exhaust port. The timing and magnitude of these waves is also determined by the physical dimensions of the exhaust and engine rpm. proper sizing of the pipes is important.

The goal of the superposed sonic waves is to have a rarefaction at the exhaust valve while it is open, a second goal is to have both the intake and exhaust valve open at the same time(aka overlap) that the rarefaction is at the exhaust valve.

Having the rarefaction at the exhaust valve while it is opens causes the extraction of additional gas over the amount that would be extracted at atmospheric pressure or worse some higher than atmospheric pressure.

Having the rarefaction at the exhaust valve while both it and the intake valve are open causes the extraction of additional gas over the amount that would be extracted at atmospheric pressure or worse some higher than atmospheric pressure and at the same time draws additional mixture into the cylinder over and above the amount that would normally be inducted, i.e. improved volumetric efficiency.

Some of this overlap(aka 5th stroke) inducted mixture will in fact go right through the cylinder and allow unburned mixture to go into the exhaust. You can see this on race cars as the unburned mixture is ignited by the temperature of the exhaust pipes, it's most visible on trailing throttle.

Because of the aforementioned raw mixture being allowed into the exhaust pipes modern smog regulated engines have little or no overlap at idle, varicam cars seek to use more and more overlap as revs build because emissions aren't a concern of the EPA at elevated rpm and throttle positions.

In addition all modern engines incorporate feedback from the exhaust(O2 sensors) to change the intake mixture to comply w/ emissions regs at low rpm/throttle w/o adversely affecting optimum power at high rpm/throttle.

Above I mentioned that the superposed sonic waves frequency, timing and magnitude are a function of rpm(among other things), this means that w/ fixed cam timing the rarefaction timing will vary wrt to the exhaust valve opening. There will be 2 rpm intervals of particular interest. In one the rarefaction will be present as desired at the open exhaust port, but this will happen at least partially over a very limited range of rpm. The other is when a compression is at the open exhaust port, this also happens(again partially) only over a very limited rpm range.

The goal of tuned headers is to get the rarefaction to be at the pot at the desired rpm, usually max hp. The side effect is that the second effect ends up in the lower mid-range. On 911s you can hear the sound of both quite clearly, if you listen.

Once again the MFI systems are extremly inflexible, they are designed for a single limited state and cannot adapt themselves well to any changes from design spec