Backpressure in exhaust (how to add and remove)
Discussion
Jhonno said:
You never want back pressure on a 4 stroke.. It is a complete misconception. Try dynoing or driving an car with a partially blocked exhaust..
What it is is gas speed.. A smaller diameter exhaust will have a higher gas speed at lower rpm, boosting the low end by increasing scavenging at that point. However, at the top end it might then be restrictive due to the fact it can't flow enough volume. This is where back pressure comes from. You don't want it, but it is a side effect of a smaller flowing exhaust which helps the bottom end.
The perfect exhaust would have high gas speeds, and no back pressure, however with many things engine flow related it is a compromise..
How can you mathmatically find the perfect balance?What it is is gas speed.. A smaller diameter exhaust will have a higher gas speed at lower rpm, boosting the low end by increasing scavenging at that point. However, at the top end it might then be restrictive due to the fact it can't flow enough volume. This is where back pressure comes from. You don't want it, but it is a side effect of a smaller flowing exhaust which helps the bottom end.
The perfect exhaust would have high gas speeds, and no back pressure, however with many things engine flow related it is a compromise..
jmanning91 said:
How can you mathmatically find the perfect balance?
Disclaimer : I am not a tuner or exhaust expert, just an enthusiastic amateur who does a lot of reading and works in the dyno industry.I am sure that max_torque has a full CFD simulation at work, and achieves far better results than I ever could, and that there are others with vast experience of exhaust design who can do it better also.
Boldly forging on regardless :
One method is to first choose the target gas speed (240 feet/second is a number I've seen mentioned by a few sources), and choose the revs that you are trying to optimise for.
Then work out the pipe area required based on revs x displacement / gas speed.
I built a spreadsheet based on that method (it also did tuned lengths based on a few simplified assumptions).
When I put in the numbers for my 100 bhp/litre (stock) engine, the predicted optimum revs for the factory 4-2-1 were within 200 rpm of actual peak torque.
I also used it to design an exhaust manifold for another car that was a significant improvement over the (crappy) previous one.
Whether the magical "best" gas speed is calculated or derived from years of "what works best", I don't know. It also makes no allowance for the expansion of the gas, so I assume it's a rule of thumb thing.
AW111 said:
227bhp said:
If you were serious you'd just buy Pipemax for $75.
But the question then is what algorithms they are using.I'm not saying they are wrong, but as a programmer in a technical field, I am interested in how their modelling works.
although he's got a lot of real life data and experience to back it up too.But it does raise the point that these days you don't need to have years and years of practical experience to put together a decent engine spec, just a reasonable amount of knowledge and some accurate data - not much point in putting crap in at one end, as you can be pretty sure more crap will come out the other end too.
Whilst it can lead to better engines being built, it can also leave us questioning why something did what it did. I think 'back in the day' you usually (although many times you didn't) knew why something was doing what it did, it is less so now with computerised simulations, you find yourself going backwards from what the screen is telling you to why it works that way.
AW111 said:
But the question then is what algorithms they are using.
I'm not saying they are wrong, but as a programmer in a technical field, I am interested in how their modelling works.
its based alot on on empirical data to, what the underlying equations and relationships are would be IPI'm not saying they are wrong, but as a programmer in a technical field, I am interested in how their modelling works.
the issue with it is gives multiple options, you need to understand the various options its telling you based on the application you have, if you are building a street engine or maximum effort drag racing engine you'll use different recommendations from the program
stevieturbo said:
It'd be interesting to see the same test with the engine being mapped specifically for each exhaust system.feef said:
It'd be interesting to see the same test with the engine being mapped specifically for each exhaust system.
They also kept the same manifold each time so there would have been a significant step down in diameter between the collector and the 2.5" exhaust. Even so it's not really a surprising result on an engine making that much power, even with the fueling optimised I doubt a 2.5" system would be big enough.Mr2Mike said:
They also kept the same manifold each time so there would have been a significant step down in diameter between the collector and the 2.5" exhaust. Even so it's not really a surprising result on an engine making that much power, even with the fueling optimised I doubt a 2.5" system would be big enough.
There are various other test videos changing manifold tube sizes and system sizes etc.All cool stuff and mostly quite surprising.
But does go a long way to proving people do get worked up a lot over nothing when it comes to exhaust design. Yes there are gains....but they can be smaller than many think. And losses perhaps more difficult to achieve too.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff