Engine and Transmission Tech

Saw this on MGE and I'm confused. Seems to run counter to everything I've always been taught about restrictive exhaust sucking up horsepower. Anybody have an explanation?





Edited 2 time(s). Last edit at 02/08/2016 01:12AM by HealeyRick.

> Seems to run counter to everything I've always been taught about restrictive exhaust sucking up horsepower. Anybody have an explanation?

Vizard reported similar results in one of his early books. The sharpness of the bend at the cylinder head exhaust port exit, lengths and diameters of the primaries and collectors are all still important, as is the muffler flow but you can dent tubes without much effect. The latter doesn't appear to upset the finite amplitude waves which is the dominant header tuning effect. Also, bank separated 4-into-1 headers (on 90 degree V8s with non-flat plane crankshaft) are relatively insensitive to equal length primaries but 180 degree headers are sensitive to equal length headers. One header builder I know tunes his 180 degree headers acoustically, adjusting the length to play the same note.

Dan Jones

It's all cumulative. A single small dent simply creates a venturi and doesn't cause much restriction. Lots of them together add up. The other thing that people don't understand is that ever gram of air passing through the headers had to enter the cylinder through the intake manifold. That path is MUCH more restrictive than the exhaust side usually. This isn't an exhaust pressure issue, it's a mass flow issue.

I'm wondering if the scavenging effect of the tuned exhaust is enough to explain why there was no power loss. I'm assuming that the exhaust is tuned to the RPM where the peak power was made, so there is a low pressure pulse in the area of where those dinged-up spots on the pipe are. Also makes me question what is the best strategy for porting the exhaust. After seeing this, I'm guessing that simply making the exhaust ports smooth, with no sharp edges or sharp bends, is what's important. The Ford 5.0 GT40p heads have small but smoothly curving exhaust ports that out-flow previous production heads despite being smaller ports.

So I'll be making my ports smooth, gradual, and round. I'm also thinking stepped, long tube headers are the ideal. My current project is a Ford 2.8 V6 and the headers need reworking to clear the floor of the Sunbeam.

Gasses being highly compressible, a relatively small orifice effect will not effect flow as much as you might expect. Resonance is also relevant of course, but the gasses still have to move past the obstruction.

I'm curious what you have in mind when you say "stepped".

Jim

Hi Jim.

By 'stepped headers' I mean the design where the primary pipes start smaller, like 1 1/2" and are merged into slightly larger pipes, 1 9/16". The area increases from 1.767 sq. in. to 2.07 sq. in. I think the theory is that the tuning of the length of the step will help cause a low pressure pulse at the exhaust valve. In my case, because I will have to do some modification to the headers to get them to fit, fitting smaller initial primary pipes would also be an expediency to simplify the job.

I can see simplifying the job certainly.

Probably the two things to consider are velocity, and the fact that gasses take less space as they cool. Probably not a consideration that close to the ports though. I do think I've seen stepped headers where the tubes stepped down as they got farther from the ports but I never saw much of anything to suggest there was any great advantage to it. On 2 strokes the pipes get larger and then smaller as the gasses cool and the pulses average out, quite a lot of science and magic goes into those. I don't see where you'd have anything to worry about.

Jim

When you asked about why go with stepped headers and I got to thinking about it, I realized exactly what you had pointed out; as exhaust gasses cool, they contract and that slows the velocity. So a stepped header would actually slow down the exhaust even more, which is not exactly what you want.

I did a back-of-the-envelope calculation to try to get a swag at what the length of the exhaust sound pressure wave would be at 3600 RPM and I came up with something like 15 feet to 30 feet, depending on the temperature of the gas. If I got the calculation right, that would explain why exhaust pipes and tailpipes figure into exhaust tuning. I'd heard reliable reports that for an engine like a 350 Chevy, optimum exhaust pipe size would be something like 2 1/8" and the tailpipe would be smaller, like 1 7/8", all in an effort to keep exhaust velocity up and help propagate the pressure wave back to the engine. For my little V6, everything will be much smaller. That's actually a good thing because part of the exhaust system routes through the frame and large diameter exhaust pipes would make for a tight fit.

Happily, the exhaust pulses on a V6 are evenly spaced on each bank, which makes exhaust tuning a little simpler. According to the quick research I did, the headers for my 2.8 (or 3.4) would have primary pipe lengths in the range of 31" for peak torque at 5500 RPM. I think that the headers I have are bigger diameter and shorter length than what would be optimum. I've never worked on headers before, but I have a 110 volt MIG welder and I might give it a shot.

Thanks for the remarks. You are one of the guys around here who put a lot of thought into how things work and how to make them work better.

I think that we've all been duped a little bit.
That engine should have been good for another 100 or so horses and at least 50 more lbs-ft of torque.
That old tunnel ram with 1500 CFM of carbs was the actual limiting factor to me.
At the 560 power level the headers were probably a bit big which explains the low torque.
Denting the tubing just got the header closer to a size that that combination needed.
It would have been nice to see some of the engine parameters during the pulls.
But I must say that it's still a pretty good showing for a bashed up set of headers.

While we're bashing.... I bought OBX(rice paddy) stainless headers, for my gen III hemi. Other than being really bowed at the head mating surface, they "look" good. How they "work" on a dyno, tbd.