Exhaust Size vs HP & TQ
In my experience of putting 3.5" and 4" exhausts on Evo's, Supras, Neons, and Honda's... never once has the car made more power on a smaller exhaust...anywhere.
Was the subaru tuned well? I wonder if the ignition advance was left the same and not advanced for the larger exhaust.
I ran this scenario on the dyno today at work... lost power everywhere. EGT's were higher though
Pre-turbine pressure was also higher.
I'll try it on a stockish evo this winter on Forged's mustang dyno... I've got more than enough pipe to test it.
Was the subaru tuned well? I wonder if the ignition advance was left the same and not advanced for the larger exhaust.
I ran this scenario on the dyno today at work... lost power everywhere. EGT's were higher though
Pre-turbine pressure was also higher. I'll try it on a stockish evo this winter on Forged's mustang dyno... I've got more than enough pipe to test it.
I have no specifics on the subaru, just saw the dyno sheet. It was in '02??? First year the WRX was available.
It very well could have been a tuning issue, difference in mufflers, bends, I don't know. Could just have been difference due to bend radius. Actually, that's been something I wanted to bring up.
Below is a 1-D pipe flow equivalent length chart for bends in pipes. The charted values are for 3" tubing. While a 9" bend radius would be nice, it obviously won't work for a downpipe of O2 housing. But going from the 3" radius that a lot of people tend to use to something like 4.5" or 6" on the more gentile bends with room (around the diff and back by the muffler) makes a very big difference.
For example, let's say you build a turbo back exhaust for a car flowing 60 lbs/min:
(2) 90 degree bends
(3) 45 degree bends
10' of straight
With 3" diameter tube and 3" bend radius
27.5' equivalent length, ~30 PSI pressure drop
With 3" diameter tube and 4.5" bend radius
22' equivalent length, ~24 PSI (20% reduction)
With 3" diameter tube and 6" bend radius
20.6' equivalent length, ~22 PSI (27% reduction)
With 3.5" diameter tube and 3.5" bend radius
27.5' equivalent length, ~13 PSI (57% reduction)
Dang I hate adding anything good here anymore but this is an intersting subject that I've done tons of testing on.
In 2003 we did exhaust testing, hard to even find it now. First we did the axle back, car gained power, then we did the downpipe, car gained no power, then we did an offroad pipe, car gained more power, then we put the stock downpipe back on and lost power, put the new downpipe back on, gained power. The only part left ont the car stock was the stock mid pipe between the cat and the muffler. We had built everything out of 2.5" mandrel bent tubing up to that point. I don't remember how much power the car was making at the time, over 300 on our old Dynojet I'd guess. Then we replaced that one last section of stock turbing with a 2.5" section and the car made some rediculous gains in power, I am thinking over 20 whp. The factory resonators are VERY restrictive.
We then of course built our 3" system, tested it against the 2.5" and stock. The 3" worked the best throughout the entire RPM range. The 2.5" vs 3" downpipe made no difference in power then, I'd say the power levels then were under 400 whp the last testing we did on 2.5" vs 3" downpipes.
Flash forward to about 2 years ago. I had the lighest 3.5" tubing with minimal bends in it bent up for a 3.5" turbo back system to market. I built one for my car, the Bad Bish, put it all together using V-band connections and a Burns super lightweight racing muffler. Loaded the car on the dyno and dyno'd the car back to back VS our 3" turbo back with our bullet muffler. The car was over 600 whp on our Mustang Dyno at the time. The results? 3" had more low/mid and the same or better top end. I could not believe it. I had all those exhausts left at our shop until about 6 months ago when I traded them to another shop looking to build some. I built one for Al's GT42r car, one for Matt's GT42r car and sold the one I built for my car.
To add even more to this. I think it was this spring when I decided that my 700 whp car was just too loud and I was sick of listening to it. The car had run (most of you know this already) 9.0 at 159.6 mph through our standard 3" turbo back with the bullet muffler. I found a small diameter lightweight muffler that I wanted to try to run in the rear. My car runs our Bullet muffler in the center where the cat goes because of the JDM EVO9 rear bumper cover, I don't like to see a big muffler back there. This new muffler I found was very small in diameter, we named it the SD muffler (small diameter). I kept the bullet in the center and added the SD muffler to the rear, it took a bunch of the noise out of the exhaust. Anyway, the car was on the dyno, never unstrapped. I made a couple of pulls with the car that were in the 69x whp ran and then we swapped on the SD, the power went to 700.
Some of what I have seen makes no sense but the dyno doesn't lie and you can gauranfkingtee that if I am testing stuff like this for my own use and my own car there is no BS being pulled, I want my car to be as fast as it can be. To this day I run our complete 3" turbo back, bullet in the center and SD in the rear. The car has made 710 whp like that now on our dyno.
My car also runs a 44 mm Tial external gate that of course dumps some exhaust out the front of the car under boost, definetely beneficial. The odd thing with this is the stock style 02 housings that do the same thing...........I've seen nothing but losses from.
Maybe Myth busters need to buy a dyno. haha
In 2003 we did exhaust testing, hard to even find it now. First we did the axle back, car gained power, then we did the downpipe, car gained no power, then we did an offroad pipe, car gained more power, then we put the stock downpipe back on and lost power, put the new downpipe back on, gained power. The only part left ont the car stock was the stock mid pipe between the cat and the muffler. We had built everything out of 2.5" mandrel bent tubing up to that point. I don't remember how much power the car was making at the time, over 300 on our old Dynojet I'd guess. Then we replaced that one last section of stock turbing with a 2.5" section and the car made some rediculous gains in power, I am thinking over 20 whp. The factory resonators are VERY restrictive.
We then of course built our 3" system, tested it against the 2.5" and stock. The 3" worked the best throughout the entire RPM range. The 2.5" vs 3" downpipe made no difference in power then, I'd say the power levels then were under 400 whp the last testing we did on 2.5" vs 3" downpipes.
Flash forward to about 2 years ago. I had the lighest 3.5" tubing with minimal bends in it bent up for a 3.5" turbo back system to market. I built one for my car, the Bad Bish, put it all together using V-band connections and a Burns super lightweight racing muffler. Loaded the car on the dyno and dyno'd the car back to back VS our 3" turbo back with our bullet muffler. The car was over 600 whp on our Mustang Dyno at the time. The results? 3" had more low/mid and the same or better top end. I could not believe it. I had all those exhausts left at our shop until about 6 months ago when I traded them to another shop looking to build some. I built one for Al's GT42r car, one for Matt's GT42r car and sold the one I built for my car.
To add even more to this. I think it was this spring when I decided that my 700 whp car was just too loud and I was sick of listening to it. The car had run (most of you know this already) 9.0 at 159.6 mph through our standard 3" turbo back with the bullet muffler. I found a small diameter lightweight muffler that I wanted to try to run in the rear. My car runs our Bullet muffler in the center where the cat goes because of the JDM EVO9 rear bumper cover, I don't like to see a big muffler back there. This new muffler I found was very small in diameter, we named it the SD muffler (small diameter). I kept the bullet in the center and added the SD muffler to the rear, it took a bunch of the noise out of the exhaust. Anyway, the car was on the dyno, never unstrapped. I made a couple of pulls with the car that were in the 69x whp ran and then we swapped on the SD, the power went to 700.
Some of what I have seen makes no sense but the dyno doesn't lie and you can gauranfkingtee that if I am testing stuff like this for my own use and my own car there is no BS being pulled, I want my car to be as fast as it can be. To this day I run our complete 3" turbo back, bullet in the center and SD in the rear. The car has made 710 whp like that now on our dyno.
My car also runs a 44 mm Tial external gate that of course dumps some exhaust out the front of the car under boost, definetely beneficial. The odd thing with this is the stock style 02 housings that do the same thing...........I've seen nothing but losses from.
Maybe Myth busters need to buy a dyno. haha
Its about time you chimed in Dave. But the funny thing is everytime you bring in usefull information you usually get some hate.
Its pretty interesting since you mentioned with bullet muffler you made 69X whp then when you swapped in the 2nd muffler (SD) the power went up.
Could there be something more to this all then just exhaust size? Maybe the exhaust tone plays a role in this some how. After all sound is vibrations in air. Maybe a certain frequency in exhaust sound/wave helps flow!? Now thats something to boggle the mind!
Its pretty interesting since you mentioned with bullet muffler you made 69X whp then when you swapped in the 2nd muffler (SD) the power went up.
Could there be something more to this all then just exhaust size? Maybe the exhaust tone plays a role in this some how. After all sound is vibrations in air. Maybe a certain frequency in exhaust sound/wave helps flow!? Now thats something to boggle the mind!
Also to Dave:
Have you considered making a downpipe that starts with the 2.5 inch inlet just like your downpipe flange already has, then slowly increases in size to 3" ? Something more like around the lenght coarse of 6-8" to increase from 2.5" to 3" or what ever results in best transition?
Have you considered making a downpipe that starts with the 2.5 inch inlet just like your downpipe flange already has, then slowly increases in size to 3" ? Something more like around the lenght coarse of 6-8" to increase from 2.5" to 3" or what ever results in best transition?
Dave, have you ever measured the back pressure in the exhaust system?
I look at the numbers that come out of a simple 1-D approximation and the pressure losses seem very high to me. I'm just curious what you've seen.
Also, my assumption of 60 lb/min through the exhaust might not be any where close to realistic on a 700 HP car with the external going to atmosphere.
Regardless, I think your exhaust system proves a point. You've made it light and as straight as possible. You take the time to control the things that really matter and you can make a 3" exhaust work for just about everything.
The evo does have the benefit of a very straight exhaust path. No going over rear cross members, or bending around gas tanks. Just straight down the middle.
I look at the numbers that come out of a simple 1-D approximation and the pressure losses seem very high to me. I'm just curious what you've seen.
Also, my assumption of 60 lb/min through the exhaust might not be any where close to realistic on a 700 HP car with the external going to atmosphere.
Regardless, I think your exhaust system proves a point. You've made it light and as straight as possible. You take the time to control the things that really matter and you can make a 3" exhaust work for just about everything.
The evo does have the benefit of a very straight exhaust path. No going over rear cross members, or bending around gas tanks. Just straight down the middle.
Last edited by 03whitegsr; Dec 20, 2009 at 09:40 PM.
Flash forward to about 2 years ago. I had the lighest 3.5" tubing with minimal bends in it bent up for a 3.5" turbo back system to market. I built one for my car, the Bad Bish, put it all together using V-band connections and a Burns super lightweight racing muffler. Loaded the car on the dyno and dyno'd the car back to back VS our 3" turbo back with our bullet muffler. The car was over 600 whp on our Mustang Dyno at the time. The results? 3" had more low/mid and the same or better top end. I could not believe it. I had all those exhausts left at our shop until about 6 months ago when I traded them to another shop looking to build some. I built one for Al's GT42r car, one for Matt's GT42r car and sold the one I built for my car.
I remember this testing quite clearly. When Ernie called BS to my experience of a 2.5 making more power than a 3 on my particular setup, I was going to ask him to search for your testing of the 3.5 against the 3.0. I was sure you put the 3.0 back on car because the power band was superior. My assessment of what happened was you found the point in exhaust flow at which a 3.0 exhaust has some scavenging effect. the 3.5 exhaust moved the pulses farther apart and the scavenging effect was lost.
at 4-500 hp there likely is no scavenging effect with a 3.0 exhaust. a 3.5 swapped in would likely make more power at that hp level.
My theory of what is going on is there is a certain frequency that allows the pulses to pull on each other. too large a frequency (pulses very close) and there is too much back pressure, too small a frequency ( pulses far apart) and there is no pulling effect.
I remember this testing quite clearly. When Ernie called BS to my experience of a 2.5 making more power than a 3 on my particular setup, I was going to ask him to search for your testing of the 3.5 against the 3.0. I was sure you put the 3.0 back on car because the power band was superior. My assessment of what happened was you found the point in exhaust flow at which a 3.0 exhaust has some scavenging effect. the 3.5 exhaust moved the pulses farther apart and the scavenging effect was lost.
at 4-500 hp there likely is no scavenging effect with a 3.0 exhaust. a 3.5 swapped in would likely make more power at that hp level.
My theory of what is going on is there is a certain frequency that allows the pulses to pull on each other. too large a frequency (pulses very close) and there is too much back pressure, too small a frequency ( pulses far apart) and there is no pulling effect.
Last edited by 94AWDcoupe; Dec 21, 2009 at 12:23 AM.
Maybe someone could do some testing on frequencys to see how much of a role they play. Someone that has done lots of testing before, someone with a dyno and their own shop. Someone with a 710whp evo, i wonder who that could be?
Frequency has already been shown to be a MAJOR contributor to flow in engines.
No testing needed.
Until you want to get to the point of controlling tube diameter to 1/16" increments and having a custom exhaust for each car, it seems like a waste of time?
No testing needed.
Until you want to get to the point of controlling tube diameter to 1/16" increments and having a custom exhaust for each car, it seems like a waste of time?
based on David's findings it almost seems that there is more science behind it than logic...the fact that he added a second muffler and gained power certainly confuses me haha
interesting info all around...good reading
interesting info all around...good reading
Well, no hate so far, so that's nice.
I have not considered making a downpipe that has a slow gradual transition in it. The tooling would be expensive and I have done some generic testing of the sort before. We used one of those Meg's Megafone cone tips, which starts small and goes up to like 4", it's LONG. The transition didn't seem to make any difference when going from small to large tubing diameter.
I have a rule I tell people on tubing size in the exhaust or intake tract. It is fine to go from a small tube into a larger tube but you can NEVER go from large to small and expect good results.
So, IF you have a 2.5" 02 housing and dump it into a 3" downpipe, no problem. You can not put a 2.5" offroad pipe or catback after a 3" downpipe though, it kills the flow. Small to big is fine, big to small is not good.
I have no idea how to test frequency, that is way beyond me.
Back pressure testing, I have no tested back pressure in the exhaust system but have done quite a bit pre-turbine housing.
03white, based on your forumula you posted above, which confuses me, what would estimate the back pressure is in my exhaust at 710 whp? This is not a trick question, I have no ever measured it. I'm curious.
I have a 4.5" radius bends, 80 degree bend in the downpipe, two bends that are about 15 degrees after that, then from the firewall aread to the axle is straight, I'd estimate the other two bends to be under 30 degrees, then the muffler. Just give me a guess. I have something else I want to add to this after hearing your estimate.
I have not considered making a downpipe that has a slow gradual transition in it. The tooling would be expensive and I have done some generic testing of the sort before. We used one of those Meg's Megafone cone tips, which starts small and goes up to like 4", it's LONG. The transition didn't seem to make any difference when going from small to large tubing diameter.
I have a rule I tell people on tubing size in the exhaust or intake tract. It is fine to go from a small tube into a larger tube but you can NEVER go from large to small and expect good results.
So, IF you have a 2.5" 02 housing and dump it into a 3" downpipe, no problem. You can not put a 2.5" offroad pipe or catback after a 3" downpipe though, it kills the flow. Small to big is fine, big to small is not good.
I have no idea how to test frequency, that is way beyond me.
Back pressure testing, I have no tested back pressure in the exhaust system but have done quite a bit pre-turbine housing.
03white, based on your forumula you posted above, which confuses me, what would estimate the back pressure is in my exhaust at 710 whp? This is not a trick question, I have no ever measured it. I'm curious.
I have a 4.5" radius bends, 80 degree bend in the downpipe, two bends that are about 15 degrees after that, then from the firewall aread to the axle is straight, I'd estimate the other two bends to be under 30 degrees, then the muffler. Just give me a guess. I have something else I want to add to this after hearing your estimate.
03white, based on your forumula you posted above, which confuses me, what would estimate the back pressure is in my exhaust at 710 whp? This is not a trick question, I have no ever measured it. I'm curious.
I have a 4.5" radius bends, 80 degree bend in the downpipe, two bends that are about 15 degrees after that, then from the firewall aread to the axle is straight, I'd estimate the other two bends to be under 30 degrees, then the muffler. Just give me a guess. I have something else I want to add to this after hearing your estimate.
I have a 4.5" radius bends, 80 degree bend in the downpipe, two bends that are about 15 degrees after that, then from the firewall aread to the axle is straight, I'd estimate the other two bends to be under 30 degrees, then the muffler. Just give me a guess. I have something else I want to add to this after hearing your estimate.
I was just hoping to be within 20-30% and then use it for relative comparisons between tubing size, material, bend radius, etc.
To be honest, I'm not sure what to estimate for airflow through the exhaust. With the wastegate flow going to atmosphere, even a rough guess from just HP and a rough HP/lbm still gives me nothing on actual exhaust flow as wastegate flow might be any where from 10-50% of the exhaust flow.
Another potential down fall is tube roughness. I just used a generic roughness value for cold-rolled steel.
I'll have to get back to you though, my spread sheet for figuring this stuff out is on my work PC.
Whenever 03white, if I'm not here post it up on our forums and we can continue the conversation there. I got hit with more points last night and should get banned any minute. Interesting topic.
94awdcoupe, that makes no sense to me, I know nothing about all this math and computation though, I put a part on, dyno it, race it and see what happens, that's how I test something. But how can a turbo flow 60lbs, then you add a bunch of fuel to the mix and it then puts out less than half of what it took in out the exhaust? I'm not being a smart ***, I am asking, doesn't make sense to me.
94awdcoupe, that makes no sense to me, I know nothing about all this math and computation though, I put a part on, dyno it, race it and see what happens, that's how I test something. But how can a turbo flow 60lbs, then you add a bunch of fuel to the mix and it then puts out less than half of what it took in out the exhaust? I'm not being a smart ***, I am asking, doesn't make sense to me.