• Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

How do HP, Torque, and backpressure get along?

M

Mopar500

NAXJA Forum User
Location
Colorado
I am trying to get this all straight in my head.
Stock exhaust = restrictive, low horsepower and low torque
High flow exhaust = less restrictive, high horsepower and low torque
Somewhere inbetween exhaust = medium horsepower and high torque

Now enter into this equation the noise factor.
I don't care what anybody else says, these 4.0s sound terrible when loud. Very ricey IMO. I guess there's no getting away from it. I started out with a new 2.25" exhaust system. (Evidently if you figure in the bends on the stock exhaust it's only 2".) The new system is 2.25" even in the bends. I have a high flow cat (Magnaflow) and started out with the straight-through Maganaflow muffler. I have two in the 2.5" dual exhaust my big block Coronet and it's almost quieter than it was when stock.
I lived for about 24 hours with the straight-through Maganaflow on my XJ before I coundn't stand it. The local shop replaced it with a baffled version that was much quieter and seemed to have a noticable increase in torque. The shop guy (who seems honest, good guy) explained that the increase in backpressure from the more restrictive muffler was responsible.
So now, even the more restrictive muffler is getting annoying. The shop guy told me that to make it any quieter would greatly reduce performance. Originally I responded NO. I will keep every bit of performance I can get. But now I am having second thoughts. I am wondering how much I would loose. I am thinking I could live with it, if I didn't sound like a ricer everywhere I went. I am way too old for that!

What are your thoughts?
 
I also hate loud Jeeps. The quietest performance muffler I have ever run was a 20" Dynomax muffler. Unfortunately, it doesn't fit with the suspension crossmembers I now run. A close second was this $14 14" muffler from Summit. Nice tone, but very quiet. My muffler's don't last long enough in the rocks to rot out, so I have no idea how long it will actually last.........but the price is right. Link for 2.25" version.

http://store.summitracing.com/partd...122&N=400304+1002+4294922710+115&autoview=sku

I use a 2.5" exhaust from the header to the tip, on a 4.6 liter stroker.

Mopar500 said:
I am trying to get this all straight in my head.
Stock exhaust = restrictive, low horsepower and low torque
High flow exhaust = less restrictive, high horsepower and low torque
Somewhere inbetween exhaust = medium horsepower and high torque

Now enter into this equation the noise factor.
I don't care what anybody else says, these 4.0s sound terrible when loud. Very ricey IMO. I guess there's no getting away from it. I started out with a new 2.25" exhaust system. (Evidently if you figure in the bends on the stock exhaust it's only 2".) The new system is 2.25" even in the bends. I have a high flow cat (Magnaflow) and started out with the straight-through Maganaflow muffler. I have two in the 2.5" dual exhaust my big block Coronet and it's almost quieter than it was when stock.
I lived for about 24 hours with the straight-through Maganaflow on my XJ before I coundn't stand it. The local shop replaced it with a baffled version that was much quieter and seemed to have a noticable increase in torque. The shop guy (who seems honest, good guy) explained that the increase in backpressure from the more restrictive muffler was responsible.
So now, even the more restrictive muffler is getting annoying. The shop guy told me that to make it any quieter would greatly reduce performance. Originally I responded NO. I will keep every bit of performance I can get. But now I am having second thoughts. I am wondering how much I would loose. I am thinking I could live with it, if I didn't sound like a ricer everywhere I went. I am way too old for that!

What are your thoughts?
Click to expand...
 
Here's a write-up on this subject that I did some time back. I wrote it mainly to debunk the myth that back pressure "creates" torque, but it includes a general description of the relationships...

Back pressure does not create torque and back pressure is not a good thing for your exhaust system. It adds nothing! It reduces performance and gas mileage whenever and wherever it occurs. It is, however, practically impossible to avoid and therein lies the confusion.

The key to any exhaust system is the velocity of the exhaust pulses through the system. A good header and exhaust system work together to create smoothly and quickly flowing pulses of exhaust. As the pulses move through a well designed system they can create a vacuum behind them that, at certain RPMs, will actually suck exhaust gasses out of the cylinders as the exhaust ports open.

When you hit that RPM range where the exhaust pulses are flowing as quickly as they can and the exhaust is actually being sucked out of the cylinders you are said to be "on the pipe." How the exhaust system is designed (size and length of pipes, size of collector, etc.) will determine at exactly what RPM this happens. At RPMs below this range the pulses will not be coming out of the cylinders fast enough to work together and create the vacuum. At RPMs above this range the exhaust pulses will be coming out of the cylinders faster than the system can flow them and back pressure is created.

So, obviously, the exhaust system needs to be designed as a part of the total engine design, and you have to take into account the RPM range where you want to get your maximum horsepower and torque. Smaller pipes will facilitate fast-flowing exhaust pulses at lower RPM ranges, but result in back pressure at higher RPMs. Larger pipes will flow more exhaust at higher RPMs, but result in slow-moving exhaust at lower RPMs.

This is where the myth that back pressure creates torque comes from. For maximum low-end torque, good manners in traffic, towing, and off-road driving you generally want smaller pipes that give good exhaust velocity at low RPMs. Pipes designed this way will have back pressure at higher RPMs. It's not that back pressure "creates" torque, it's that an exhaust system designed for low-end torque will practically always result in back pressure at higher RPM ranges.

Run large, straight pipes and you reduce the exhaust pulse velocity at lower RPMs so that it feels like you've lost torque. But you haven't actually "lost" it, you've just moved the RPM range up so high that you have little torque or HP down at lower RPMs.
 
The quietest exhaust I've ever owned was in an 81 Cadillac. It had an ordinary passenger car muffler, but also had a long resonator about 18 inches before the exhaust tip. I suppose the engineers placed the resonator there after testing different distances from the muffler for maximum noise damping.
The resonator literally detunes the exhaust pulses. It cancels and scatters the sound waves and absorbs them into the glass packing.

You could make a silent Cherokee exhaust, if you could find the right spot to add a resonator, but it would be trial and error. It does add some back pressure, but you would be riding luxury car quiet.
 
Hmmmm.
dmillion, that explains things very well. Thanks.
So if I understand correctly, tuning the exhaust can move the torque curve.
I know that in '96 the torque curve was moved 1,000 rpm lower on the 4.0 liter. I am told that was due to mainly a cam change. Would it be possible, or realistic, to attempt doing this with exhaust tuning? I assume not. But with the way you've explained it, what creates the limitation in that respect?
Thanks again, I can feel my brain growing!
 
Mopar500 said:
Would it be possible, or realistic, to attempt doing this with exhaust tuning?
Click to expand...
To some extent, yes, but only to a very limited extent. That's because everything has to work together. If you put in a cam that's designed for high RPMs with an exhaust that's designed for low RPMs you're going to create problems.

And don't forget the intake. There, too, we have to balance the ability to flow large amounts of air with the velocity that the air moves through the system at different RPMs. Much like the exhaust system, you want an intake that works well at the RPM range you're designing for--not just the highest possible flow rate.

This is why people say things like, "I'm annoyed, because I put on a header that's supposed to give me 10 hp, but I can't feel any improvement." Well, just a header by itself will do very little. You have to always look at the engine as an entire system. A change to one thing rarely does much and may actually make things worse.

Everything works together.
 
If you want to get deeper into the technical side, the exhaust flow speed that peak torque occurs is between 240 and 280 feet per second (~.25 Mach). This speed is similar for both exhaust and intake gas flows. This is the gas flow speed that the friction restriction of the airflow does not exceed the benefits of momentum. This is the speed where the pull from the mass of gas traveling down the pipe develops a vacuum upstream in the pipe network to help suck exhaust out of the back of the exhaust valve.

The gas momentum is the force of the mass of gas flowing down the pipe: one of the laws of physics, Force = 1/2 mass * Velocity squared. This momentum effect is part of Newtons second law, "things in motion tend to stay in motion" (until resistance is encountered). We need to balance momentum with resistance.

Below 240 feet per second the momentum exists, but the full benefit has not been realized. When you get gas flows above 280 feet per second the friction of the gas on the pipe or port walls grows, and the friction loss begins to exceed the momentum the benefit.

In real life the differences are small. On a 4.0L XJ engine the difference in peak torque rpm by changing from the stock 2.25" pipe to a 2.5" exhaust is about 300 rpm (IIRC, ~900 & 1200 rpm). Not much change when you have a fairly large pump (engine) and small pipe (the 4.0L pump can push against considerable more resistance and rpm without significantly impacting the power output). The effects from 2" to 3" pipe are more obvious with smaller displacement engines (like on half of a 5L V8).

One missing issue is that the peak torque is defined by the smallest pipe in the system. Installing a 2.5" can-back exhaust leaves the stock 2.25" downpipe. This leaves the 2.25" downtube as the smallest pipe in the system, and as the defining gas flow velocity (where the friction loss will exceed the momentum benefit first). When adding a cat-back system the peak torque rpm will not change, but the overall system restriction will be reduced (with less pipe length in the more restrictive 2.25" size after the cat). The result is the same peak torque rpm with a slower increase in the friction loss restriction as rpm increases (the Velocity squared in the 2/3 of pipe behind the cat is reduced, reducing the friction).

One other piece of the puzzle of relating seatofthepants feel to reality is why a pipe change can appear to reduce the peak torque at the peal torque rpm (it just does not feel the same)? This is due to the other half of the physics, the mass in uniform motion. When we make a pipe diameter change we alter the volume and total mass of gas traveling at the uniform velocity. The exhaust gas after the cat slows down and is "decoupled" from the mass that defines the peak torque rpm. With less length of 2.25 inch pipe we lose some of the conditioned mass that helps the F-mA momentum exploited to pull exhaust from behind the exhaust valve. The true reduction in the peak torque is small, but it appears to be larger than reality because what it is gaged against is smaller, the drop off in torque at higher rpm is less (we do not feel our pants lift out of the seat as quickly as the rpm climbs).

The two elements to exploit when designing an exhaust system are shaping a large mass/volume of gas at a constant velocity (how long the pipe is at the constant defining diameter) and the sizing the pipe to match the rpm where the sweet gasflow velocity exists (size the diameter of pipe needed to achieve 240-280 feet per second). We see these considerations in header and exhaust designs all the time. Long-tube headers (when room exists) perform better than shorties (the peak torque rpm is better defined). Right sized header and exhaust pipes perform better at low rpm than excessively large pipes (the joeracer 2" head pipes may look way-cool but do not help much below joeracers targeted 7500 rpm track speed).

The 96 and later 4.0L changed more than the cam. The exhaust ports are smaller, increasing the friction loss and lowering the peak torque rpm. The exhaust was also changed to a cast header. These changes were made to improve the emissions (reduce NOx) by trapping more exhaust in the combustion chamber, leaving less oxygen to burn, and lowering the combustion temperature. Since the Engineers had to make changes to lower the emissions, changes that lowered the rpm where the gas momentum benefit occurred, they redesigned the intake and exhaust, and engine management system, to make the best power with the changes (resulting in the same peak torque value 1000 rpm lower than the year before). The result was a better intake design and engine management system to ovecome the forced poor exhaust design. The 96 and up head may not be the best hotrodder choice for high rpm performance but the total package worked well.

HTH?
 
Ed A. Stevens said:
If you want to get deeper into the technical side, the exhaust flow speed that peak torque occurs is between 240 and 280 feet per second (~.25 Mach). This speed is similar for both exhaust and intake gas flows. This is the gas flow speed that the friction restriction of the airflow does not exceed the benefits of momentum. This is the speed where the pull from the mass of gas traveling down the pipe develops a vacuum upstream in the pipe network to help suck exhaust out of the back of the exhaust valve.

The gas momentum is the force of the mass of gas flowing down the pipe: one of the laws of physics, Force = 1/2 mass * Velocity squared. This momentum effect is part of Newtons second law, "things in motion tend to stay in motion" (until resistance is encountered). We need to balance momentum with resistance.

Below 240 feet per second the momentum exists, but the full benefit has not been realized. When you get gas flows above 280 feet per second the friction of the gas on the pipe or port walls grows, and the friction loss begins to exceed the momentum the benefit.

In real life the differences are small. On a 4.0L XJ engine the difference in peak torque rpm by changing from the stock 2.25" pipe to a 2.5" exhaust is about 300 rpm (IIRC, ~900 & 1200 rpm). Not much change when you have a fairly large pump (engine) and small pipe (the 4.0L pump can push against considerable more resistance and rpm without significantly impacting the power output). The effects from 2" to 3" pipe are more obvious with smaller displacement engines (like on half of a 5L V8).

One missing issue is that the peak torque is defined by the smallest pipe in the system. Installing a 2.5" can-back exhaust leaves the stock 2.25" downpipe. This leaves the 2.25" downtube as the smallest pipe in the system, and as the defining gas flow velocity (where the friction loss will exceed the momentum benefit first). When adding a cat-back system the peak torque rpm will not change, but the overall system restriction will be reduced (with less pipe length in the more restrictive 2.25" size after the cat). The result is the same peak torque rpm with a slower increase in the friction loss restriction as rpm increases (the Velocity squared in the 2/3 of pipe behind the cat is reduced, reducing the friction).

One other piece of the puzzle of relating seatofthepants feel to reality is why a pipe change can appear to reduce the peak torque at the peal torque rpm (it just does not feel the same)? This is due to the other half of the physics, the mass in uniform motion. When we make a pipe diameter change we alter the volume and total mass of gas traveling at the uniform velocity. The exhaust gas after the cat slows down and is "decoupled" from the mass that defines the peak torque rpm. With less length of 2.25 inch pipe we lose some of the conditioned mass that helps the F-mA momentum exploited to pull exhaust from behind the exhaust valve. The true reduction in the peak torque is small, but it appears to be larger than reality because what it is gaged against is smaller, the drop off in torque at higher rpm is less (we do not feel our pants lift out of the seat as quickly as the rpm climbs).

The two elements to exploit when designing an exhaust system are shaping a large mass/volume of gas at a constant velocity (how long the pipe is at the constant defining diameter) and the sizing the pipe to match the rpm where the sweet gasflow velocity exists (size the diameter of pipe needed to achieve 240-280 feet per second). We see these considerations in header and exhaust designs all the time. Long-tube headers (when room exists) perform better than shorties (the peak torque rpm is better defined). Right sized header and exhaust pipes perform better at low rpm than excessively large pipes (the joeracer 2" head pipes may look way-cool but do not help much below joeracers targeted 7500 rpm track speed).

The 96 and later 4.0L changed more than the cam. The exhaust ports are smaller, increasing the friction loss and lowering the peak torque rpm. The exhaust was also changed to a cast header. These changes were made to improve the emissions (reduce NOx) by trapping more exhaust in the combustion chamber, leaving less oxygen to burn, and lowering the combustion temperature. Since the Engineers had to make changes to lower the emissions, changes that lowered the rpm where the gas momentum benefit occurred, they redesigned the intake and exhaust, and engine management system, to make the best power with the changes (resulting in the same peak torque value 1000 rpm lower than the year before). The result was a better intake design and engine management system to ovecome the forced poor exhaust design. The 96 and up head may not be the best hotrodder choice for high rpm performance but the total package worked well.

HTH?
Click to expand...

Amen
 
Ed -
I know we'd talked about this before, but that was several system recoveries, a couple updates, and at least one brain trauma ago...

Didn't you give me a simple mathematical formula for an optimax exhaust pipe diameter at a give RPM and displacement? If you did, could I get you to give that to me again? I seem to have lost it, if I'd ever had it (I just don't remember anymore - but I don't seem to see it in any of my notebooks...)

As always, a wealth of information. Glad to see you're still around!

5-90
 
5-90 said:
Ed -
I know we'd talked about this before, but that was several system recoveries, a couple updates, and at least one brain trauma ago...

Didn't you give me a simple mathematical formula for an optimax exhaust pipe diameter at a give RPM and displacement? If you did, could I get you to give that to me again? I seem to have lost it, if I'd ever had it (I just don't remember anymore - but I don't seem to see it in any of my notebooks...)

As always, a wealth of information. Glad to see you're still around!

5-90
Click to expand...


I have a spreadsheet somewhere with the formulas. I need to find it again but with my work load of late it's not on the priority list.

I should correct that the sweet velocity for momentum (ram tuning) effect on primary pipes is the 240-280 feet per second, found at the second harmonic. The sweet velocity on a collected gas flowstream is at the first harmonic (1/2 the primary pipe velocity, or ~120-140 feet per second).

One thing to remember is that a 4-stroke is a positive displacement pump and it can overcome considerable gas flow restriction. The ram tuning effect is slight but it can enhance the low rpm torque signature (flatten the torque curve on the low rpm range of the overall peak). This is called improving the torque bandwidth and should be considered with compatible compression & cam timing to assist the extraction of exhaust from the chamber. The ideal is to minimize the unburned exhaust traveling back into the chamber at low rpm with the ram/expansion chamber effect, and still have quality port & valve flow for high rpm power (the ideal is opposite from the designed in EGR emissions control effect built into the factory 96 & up cam overlap). The short story is significant torque gains can be made across the rpm range with a wider lobe center cam change and a little porting and compression on the 96 and up 4.0L engine (any of the 4.0L engines).
 
Read David Vizard's brilliant article on exhaust science:

http://superchevy.com/technical/engines_drivetrain/exhaust/0505phr_exh/

It's the most comprehensive and informative article that I've come across on the subject. Add it to your favorites.
Since a minimum of 2.25cfm/HP exhaust flow is needed to keep restriction to less than 1%, and exhaust flow in a straight pipe is approximately 115cfm/sq. in., I've derived a simple formula for sizing the pipes for single and dual exhaust systems based on HP. Using these formulae, I made an easy to read spreadsheet so all you have to do is reference the target flywheel HP output for your engine and select the appropriate size pipe for your exhaust system:

http://www.angelfire.com/my/fan/Exhaust_Pipe_Size.htm

For a single system, the pipe diameter should be the square root of flywheel HP output / 40. For a dual system, it's the square root of HP output / 80 for each pipe. It's very accurate.
e.g. If you want to get 225-250hp from your 4.0, you need a 2-1/2" system. If you want 275-300hp from a stroker, you need a 2-3/4" single system or a 2" dual.
Looks like I'll have to upsize my exhaust pipes to 2-3/4" since I'm sitting at about 270hp with my 4.6 stroker and I have a 2-1/2" system that appears to be slightly restrictive.
When it comes to selecting a muffler (two for a dual system), total muffler flow needs to be at least 2.25 times the target flywheel HP output in cfm. Therefore a 300hp engine needs either a single 675cfm rated muffler or dual 340cfm units to achieve near zero restriction.

0505phr_exh_12_z.jpg
 
Last edited:
Don't forget that every bend in the tubing adds equivalent lenght to the actual tubing lenght.
 
What about turbos? From what I understand, after the turbo starts spinning, and the engine is running at a constant RPM, the pressure is equal on each side, only the exhaust cools off a bit (because of the recuperation of that otherwise "wasted" energy).

I was wondering about this earlier today, as I plan on building a stroker which will eventually be turbo'd, and if I have to modify the exhaust significantly (ie, from the turbo/header back), I may as well just build the motor turbo'd from the get-go. I would imagine, Dyno, that your formula still applies to some extent, but would it remain as accurate as you claim (and I don't doubt) it to be for NA applications?
 
Turbo'd cars are totaly differant. They don't rely on exhuast gas pulses for scavanging or anything like that. Bigger pipe is better for them (after the turbo). That's where the stupid ricer kids got the dumb idea the bigger the fart can, the better---on thier n/a honda civic with 3 1/2 hp.
 
I thought they did that just to be noticed? I didn't think any of them were actually dumb enough to think they gain HP from those things....well that's off-topic, and I guess I'll try searching for threads about good turbo exhausts. As you stated, and as I thought but wasn't sure, there are no gas pulses after the turbo, just a steady stream of exhaust....thanks!
 
krakhedd said:
I may as well just build the motor turbo'd from the get-go. I would imagine, Dyno, that your formula still applies to some extent, but would it remain as accurate as you claim (and I don't doubt) it to be for NA applications?
Click to expand...

I don't see why not. Horsepower is horsepower whether it's turbocharged, supercharged, or naturally-aspirated.
If you're going to build a turbo stroker from the outset, you might wanna combine a Clifford Performance dual outlet header with dual downpipes, an X-pipe, dual cats, and a true dual system.
 
You must log in or register to reply here.

Similar threads

X
2001 xj high-flow cat recomendations?
Replies
2
Views
3K
maxbraketorque
M
X
2.5 exhaust Questions
Replies
5
Views
5K
75SV1
7
S
96' Jeep XJ 4.0 Modified Exhaust Experience.
Replies
0
Views
1K
streetxj
S
J
2000 jeep cherokee rebuilt 4.0 with cam, Low Vacuum at idle
Replies
15
Views
2K
RCP Phx
R
ShoeterMcgav
The end all Exhaust/ Header thread!
4 5 6
Replies
109
Views
41K
tonytech
T
Back
Top