dual 3" exhaust no muffler vs. dual 4" with mufflers

[ptuomov]

Here's a quick question about the flow and pressure drop. The question is how much is the pressure drop for nine feet of exhaust pipe?

This is a turboback exhaust question from twin gt3071r's feeding a 5.0L V8. B so we can ignore the wave tuning. The engine will put out somewhere between 500hp and 900hp in various possible configurations, I don't exactly know yet. Right now, what's being put in is dual 3" pipes, very high flow cats (flow more than Random Tech flowbench can accurately measure), and a Borla muffler on each side.

How much am I losing because of using 3" pipes instead of say 4" pipes? How big is the loss in "muffler units?" That is, is there a way to say that 3" and no muffler will produce the same peak power as 4" pipe and a specific muffler?

[ptuomov]

Another very related question. I just looked up some of the math on wikipedia and I am having hard time believing it.

http://en.wikipedia.org/wiki/Hagen%E2%8 ... ble_fluids

What this says that the flow is proportional to the fourth power of the diameter. So dual 3 inch exhaust pipes flow only about as much as single 3.57 inch exhaust pipe. Is this true?

If this is true, why does anyone ever use dual exhausts? The weight is going to be much lower and the single pipe with only 19% larger diameter is not going to materially influence the ground clearance etc.

[96Mustang460cid]

Here's a quick question about the flow and pressure drop. The question is how much is the pressure drop for nine feet of exhaust pipe?

This is a turboback exhaust question from twin gt3071r's feeding a 5.0L V8. B so we can ignore the wave tuning. The engine will put out somewhere between 500hp and 900hp in various possible configurations, I don't exactly know yet.

The airflow difference between 500hp and 900hp is significant and greatly affects the answer to your question. Also, as the exhaust cools, it will compress and its velocity will decrease. That also greatly affects the answer to your question.

With turbulent flow (Re > ~2,400):
pressure drop = [(pipe friction coef) * (pipe length) * (density) * (flow velocity)^2] / [2 * (pipe diameter)]

Your density and flow velocity varies throughout that length of exhaust tubing.

Another very related question. I just looked up some of the math on wikipedia and I am having hard time believing it.

http://en.wikipedia.org/wiki/Hagen%E2%8 ... ble_fluids

What this says that the flow is proportional to the fourth power of the diameter. So dual 3 inch exhaust pipes flow only about as much as single 3.57 inch exhaust pipe. Is this true?

If this is true, why does anyone ever use dual exhausts? The weight is going to be much lower and the single pipe with only 19% larger diameter is not going to materially influence the ground clearance etc.

I looked at the link you referenced and don't see where it says that. Are you looking at the Poiseuille's equation for compressible fluids? If so, you can't just pick that single term (r^4) out of it and draw that broad of a conclusion. Yes, it's a dominating term, but there is more to it than that...

Comparing CSA's: dual 3.5" = 19.23 in^2 while dual 3.0" = 14.13 in^2, a 36% difference in CSA. I'm not accounting for wall thickness.

Have a good day!
Michael

[vic_dahn]

"very high flow cats (flow more than Random Tech flowbench can accurately measure)"

serious? most flow benches will go 400+ cfm or you can calculate it up to 28in if you can’t get the pressure. why would you want to use cats, are you going to emission test this thing?

[vic_dahn]

[quote="vic_dahn"]"very high flow cats (flow more than Random Tech flowbench can accurately measure)"

serious? most flow benches will go 400+ cfm or you can calculate it up to 28in if you can’t get the pressure. why would you want to use cats, are you going to emission test this thing?

I'd go with the 4in no mufflers no cats approach, that will give you the best flow!

[ptuomov]

With turbulent flow (Re > ~2,400): pressure drop = [(pipe friction coef) * (pipe length) * (density) * (flow velocity)^2] / [2 * (pipe diameter)]

Are you saying that the pressure drop is inversely proportional to the pipe diameter? The wikipedia formulas on the linked page say something quite different.

For incompressible fluids, they say


For compressible fluids, they say


In both formulas, the increase in diameter has a much larger (more nonlinear?) impact on the pressure drop than inversely proportional.

What's the source of your formula?

I looked at the link you referenced and don't see where it says that. Are you looking at the Poiseuille's equation for compressible fluids? If so, you can't just pick that single term (r^4) out of it and draw that broad of a conclusion. Yes, it's a dominating term, but there is more to it than that...

Yes, that's the equation I was looking at.

What's wrong with looking at the single term, leaving everything else constant, and drawing conclusions from that? If you have a dual 3" pipes, how large single pipe would flow the same amount at the same pressure drop? The answer is 3.57" diameter single pipe. It was surprisingly small increase in the diameter.

most flow benches will go 400+ cfm or you can calculate it up to 28in if you can’t get the pressure.

I agree, they should be able to convert them. The second hand info is that the test pressure differential dropped so far under 20 inches and they didn't like converting from there for some reason. Probably because it was close to the lunch time.

why would you want to use cats, are you going to emission test this thing?

Because I can, which is also the reason for the whole car project! Well, at least I think I can... The idea is to make a lot more power in nicely drivable street car while keeping the emissions down below stock.

I'd go with the 4in no mufflers no cats approach, that will give you the best flow!

We have dual 3" pipes with a cat and a muffler in each. It's set up in a way that the cats and the mufflers can be replaced with straight test pipes very quickly, so I'll be able to test the loss (if any) due to the cats or the mufflers.

[96Mustang460cid]

With turbulent flow (Re > ~2,400): pressure drop = [(pipe friction coef) * (pipe length) * (density) * (flow velocity)^2] / [2 * (pipe diameter)]

Are you saying that the pressure drop is inversely proportional to the pipe diameter? The wikipedia formulas on the linked page say something quite different.

For incompressible fluids, they say


For compressible fluids, they say


In both formulas, the increase in diameter has a much larger (more nonlinear?) impact on the pressure drop than inversely proportional.

What's the source of your formula?

The first formula you listed is for incompressible fluid and is irrelevant to this conversation. Compressible vs. incompressible is two completely different animals. The second formula you listed is not a formula for pressure drop. The formula I listed came from a fluid dynamics book I have in my office. To quote a source, though, pick a link:

http://www.google.com/search?q=pressure ... =firefox-a

I looked at the link you referenced and don't see where it says that. Are you looking at the Poiseuille's equation for compressible fluids? If so, you can't just pick that single term (r^4) out of it and draw that broad of a conclusion. Yes, it's a dominating term, but there is more to it than that...

Yes, that's the equation I was looking at.

What's wrong with looking at the single term, leaving everything else constant, and drawing conclusions from that? If you have a dual 3" pipes, how large single pipe would flow the same amount at the same pressure drop? The answer is 3.57" diameter single pipe. It was surprisingly small increase in the diameter.

What's wrong with picking a single term out of a complex formula and drawing a broad conclusion??? I won't answer that directly. Instead, look at the answer you came up with. Does that seem like a realistic result?

Have a good day!
Michael

[ptuomov]

With turbulent flow (Re > ~2,400): pressure drop = [(pipe friction coef) * (pipe length) * (density) * (flow velocity)^2] / [2 * (pipe diameter)]

Are you saying that the pressure drop is inversely proportional to the pipe diameter? The wikipedia formulas on the linked page say something quite different.

For compressible fluids, they say


The increase in diameter has a much larger (more nonlinear?) impact on the pressure drop than inversely proportional.

The second formula you listed is not a formula for pressure drop. The formula I listed came from a fluid dynamics book I have in my office.

The wikipedia formula seems to say that the for a given flow rate, the square of the required inlet pressure is inversely related to the fourth power of the pipe diameter. That would be roughly consistent with your formula if the denumerator would be diameter squared instead of two times the diameter.

[96Mustang460cid]

The wikipedia formula seems to say that the for a given flow rate, the square of the required inlet pressure is inversely related to the fourth power of the pipe diameter. That would be roughly consistent with your formula if the denumerator would be diameter squared instead of two times the diameter.

Quoted from the Wikipedia link you listed above:

For a compressible fluid in a tube the volumetric flow rate and the linear velocity is not constant along the tube. The flow is usually expressed at outlet pressure. As fluid is compressed or expands, work is done and the fluid is heated and cooled. This means that the flow rate depends on the heat transfer to and from the fluid. For an ideal gas in the isothermal case, where the temperature of the fluid is permitted to equilibrate with its surroundings, and when the pressure difference between ends of the pipe is small, the volumetric flow rate at the pipe outlet is given by:

IMHO, that is not an accurate formula to use in this instance.

Have a good day!
Michael

[vic_dahn]

do you guys have any assumptions on these factors:
what is your pressure or volume that you are using?
what is your evaluation of the reynolds number?
what temperature are you using? or just the specific gravity at that temp will be fine...
are we using a standard steel, seamless, Sch 10 pipe?

[ptuomov]

With turbulent flow (Re > ~2,400): pressure drop = [(pipe friction coef) * (pipe length) * (density) * (flow velocity)^2] / [2 * (pipe diameter)]

Looking at the formula more carefully, isn't the flow velocity is also a function of pipe diameter. This formula may in fact be very accurate, it's just that when thinking about the impact of pipe diameter on pressure drop, one also has to consider the indirect effect of pipe diameter on flow velocity. (I think the pipe friction coefficient may also depend on the pipe diameter, so the indirect effect from that channel would have to be taken into account as well).

Ok, back to the books...

[cjperformance]

A formula I use that works well for n/a purposes is,

0.024sq" (of cross sectional surface area) per Hp of engine out put.
ie, 0.024 x 'HP' = cross cectional area required.

-This relates to the pipe diameter needed for max Hp.

-It does not account for balance pipes, X's etc. These will increase flow potential if done right.

-Turbo's will also increase flow potential by removing pulses.

-Dont forget to factor in wall thickness.

-A specific muffler will need to be flowed to calculate its effect on flow. Some are BAD!

-Its not the be all and end all but gets you very close.

-Super charged apps will very too.

Will go on if any one wants but must go right now.
Cheers,

[ptuomov]

do you guys have any assumptions on these factors:
what is your pressure or volume that you are using?
what is your evaluation of the reynolds number?
what temperature are you using? or just the specific gravity at that temp will be fine...
are we using a standard steel, seamless, Sch 10 pipe?

The outlet pressure is going to be normal atmospheric pressure, think New England weather. The current setup is two about 10-foot or so exhaust pipes with 3" diameter. The material is 304 SS. There's a cat and a muffler, but the cat can be quickly unclamped, swapped out and replaced with a straight pipe. Muffler is Borla XR-1, shortest version with 3" inlet and outlet, both in the muffler center. I am estimating that the turbine outlet temperature will be maybe 1100F, but your guess is as good is mine I could be off by hundreds of degrees F. The target for this iteration is to take out 600 dynojet rwhp and for the next iteration (less static compression, a bit more cam) about 800 dynojet rwhp. It would be nice to have some guess of what the turbine outlet pressure will be with this exhaust and at these two power levels.

[vic_dahn]

Thank you my assumptions are close Im using Moody’s and some software to get the # sorted out. It's also nice I live with 5 other engineer's

[ptuomov]

A formula I use that works well for n/a purposes is,

0.024sq" (of cross sectional surface area) per Hp of engine out put.
ie, 0.024 x 'HP' = cross cectional area required.

-This relates to the pipe diameter needed for max Hp.
-It does not account for balance pipes, X's etc. These will increase flow potential if done right.
-Turbo's will also increase flow potential by removing pulses.
-Dont forget to factor in wall thickness.
-A specific muffler will need to be flowed to calculate its effect on flow. Some are BAD!
-Its not the be all and end all but gets you very close.
-Super charged apps will very too.

Fro 18 gauge pipe, I got an ID of 1.451 inches, area of 6.61 inches^2, the hp capacity using your formula of 275hp per pipe or 550hp for the pair. I am shooting for more, so this means that I'll be losing something because of the restrictive exhaust. The question is how much?