Info Exhaust and muffler design - interesting video

I've tried to watch this guy's YT vids, but (in my opinion) it's 90% filler and his voice/delivery is coma-inducing. If he just got to the effing point, then it might be worth it. if someone wants to summarize his video, then I'd be up for a debate about exhausts. I think too many people confuse a 2-stroke engine & exhaust with a 4-stroke engine and exhaust. I've had people tell me stuff that makes no sense whatsoever. All this is my opinion as a non-mechanic.

 

I cherrypicked my way through. Pretty much what he said at the talk he gave here some years back.
I have the handout booklet given away after the talk. Exhaust section is just a condensed version
of the talk.
I've been building exhausts for long enough now to have my own ideas. Racebikes don't have the
noise problems he's trying to get around.
The important points for fours with 4:1 pipes are 1/ The primary length sets the peak torque RPM
and 2/ The overall length (to air) sets the peak HP RPM

 

Hey Greg, do you have pictures or schematics of those 2 points? I asked a tuner about more power (from the exhaust) for the ZXR and all he said was he could raise the point in the rev range that it made peak power by shortening the pipes (headers) before they get to the 4into1 collector. He said it's a lot of dicking around for little reward, so I passed. I also spoke to a guy who ceramic coats the exhaust (inside and out); he said it helps get the spent charge out more efficiently. I wasn't too sure he wasn't speaking shite and was keen to move money from my pocket to his, so I passed on that too. Your thoughts?

 

No pics sorry. Initially my system lengths were calculated and I still have the formulas tucked away here. But once I realised
that I could tailor the torque curve to suit the engine's application everything devolved to fairly similar lengths. Some of my best pipes have been off the cuff advice on phone calls,lol. I'd built the engine for a local WSB entry - the first of the watercooled
Suzuki 750's - and the owner rang up and said Suzuki had a pipe for him but it was $3000. I reeled off lengths and diameters
and they built it. Later Suzuki gave him the kit pipe and it was identical to what I'd prescribed - but heavier. They said no thanks.

Primary lengths will be in the range of 24inches to 28inches. 24 will give a torque peak around 8000rpm 28 is around 6500rpm
Overall length is usually in the 54 to 58inch range. 54 will want to give peak hp around 11000rpm 58 about 9500-10000.

Couple of cases. Jet boat engine - big Chev. Guy who'd worked out the pipe said it was peaking slightly too high rpm for the
jet unit. They're pretty much constant speed engines running at the torque peak. I told him to lengthen the headers by 2inches.
It worked.
The GSXR1100 on Methanol I've mentioned previously in it's original form with a pipe built by a local specialist had a huge torque peak around 5000rpm. Hell to get off the line but probably a good road pipe. I deliberatley sacrificed some torque by shortening the headers 4 inches.
Making them 26inches. I was frankly surprised by the result. Getting away was easier - but hit 7000rpm and it broke
into fierce wheelspin. We ran it in this form at a local flying 1/4mile. 1.3km run-in on each side of the 1/4 - and it ran
190plus MPH while laying rubber right into the traps.

Speedway TQ's -like your compact speedcars. GSXR750's on Methanol. 28in X 1.5inOD headers 58in overall length incl
a straight-through muffler on a 2in OD tailpipe. Nice broad power spread. My engines and pipes won our International
TQ series against the Americans 3 years running.

Never got into 4: 2 : 1 systems as I suspect you'd need to computer develop those.

 

Experimental and Computational Analysis of a High Performance Four-Stroke Motorcycle Engine Equipped with a Variable Geometry Exhaust System

 

@gregt I played with this http://wallaceracing.com/header_length.php with peak power set at 9500 rpm and using my 400 specs. I assume it is for a 2 valve so I doubled the exhaust valve area. Comes in at about 22". I shall carry on with my great exhaust plan with that number.

What do you think of the paper I just uploaded? Interesting about the layouts, but I can't see how that exhaust valve would connect to the rest of the system where one is supposed to be a 4-1 merge collector and the other is 4-2. As I see it, there would be 4 pipes in and 6 pipes out. When on 4-1 the four primaries would continue into the 4-1 merge collector and when switched the 4 primaries would switch into the 2 secondaries.

 

@GreyImport Hey Chief, could you add a Research Papers section to the resources?

 

I haven't looked at the paper yet. My original formulas came from Phillip Smith's book. For single pipes I worked out
my own formula based on known published pipe specs for known motors. One at least of Phil Irvings books gives a
table of pipe dimensions which I found useful.
4:2:1 pipes in my experience don't give more peak HP than a 4:1. But they can give more off peak.
Everyone seems to want their race engines to have a straight line power curve. Not a fan -get an electric bike.
The 360 system should be better than the other pairing. I've built one 180/540 system but never dyno'd it.
If I was doing the same motor again I'd do a 4:1

Did one pipe for an FZR400. Memory says 24in X 1 3/8in headers. 13/4in OD tailpipe 54in total length incl muffler.
Worked well. Kit cams and ignition, Keihin FCR 33's NZ F3 roadracer.

Most of my later 4:1 pipes I used a "Cosworth" merge. This consists of an extended merge cone tapering down to around
85% of the tailpipe diameter followed by a short expanding taper out to the tailpipe diameter. The expanding taper is usually
one pipe diameter long. This is said to amplify the return wave. It works to extend the powerband in myexperience.
definitely worth doing.

 

That was my understanding also. The paper is based on simulation software - no actual results.

https://www.speedyauto.cc/howto/sportcar/aftermarketparts/201603/241237.html
Dyno Data: Stock vs. Tri-Y vs 4-Into-1 (Horsepower)

 

OK,I've looked at the paper. There's too much info missing to draw the conclusion that a variable geometry pipe will be better.
I'd like to see the cam timings as lobe center positions make a huge difference to where the motor makes power
We have no idea if the cam timing has been optimised.

It would also seem that the factories have decided VVT is better/cheaper/easier to do than Variable pipe geometry.
And I tend to agree with them. In practise I'd think the flow characteristics of the cross pipes and valves would
sap power.

Two strokes, yes a variable tuned length pipe is proved to work But they're very different.

 

Those lobe center figures are odd. Modern 4valve heads prefer the exhaust LC to be a larger number than the inlet.
IE slightly advanced from exactly symmetrical timing. This example is running the whole timing diagram retarded.
If that's optimised it's very unusual.

 

That Honda car engine dyno comparison that I posted uses very short primaries and long secondaries.

In response to this article https://low-offset.com/workshop/4-2-1-headers-vs-4-1-headers/ a reader corrected the author and said:
"A Tr-Y header is not exactly the same as a 4-2-1 header. In a tri-y header (which is what your article actually describes) the secondary pipe is a tuned length (second harmonic usually). A true 4-2-1 header has a very short, if any, secondary pipe. The primary pipes are long like a 4-1 header then they go into a pair of 2-1 merges which in turn immediately go into another larger 2-1 merge. The 4-2-1 header behaves very similar to a 4-1 header with a little advantage in the midrange if the pipes are paired right.

This Yoshimura system is an example of that 4-2-1.

I am modelling my exhaust on something like this simply because trying to make a 4-1 merge collector in titanium, nah never going to happen.

 

If you can roll a cone in sheet Ti, you can make a collector.
But why TI ? The Suzuki pipe I quoted above was made from 18 and 20 gauge mild steel - and was lighter than the kit Ti pipe.
Lasted two seasons as well of hard use and I believe still survives.