Exhaust Manifold Design Considerations
Turbo Mini Forums

Myself and Paul will hopefully be starting work on my exhaust manifold this week. I am still having some thoughts about the design and hope someone can shed some light.

Since I won't be able to get the centre branch as long as the two outer branches this will obviously effect the exhaust pulses as the come together at the turbo flange.

I was trying to think of a way around it. If I keep the volume of the centre branch equal to the volume of the two outer branches (by using bigger diameter pipe) then would this keep pulses apart? Or would the reduction in velocity of the pulse in the centre branch negate any gain in keeping volumes equal?

I haven't seen an equal length 5 port manifold and the only considerations on pulses seem to try and get them pointing away from eachother (ala Paul S GT1752 manifold). Then again, the Mirage points the outer branches together.

Should I just shut up, stop being so fussy and build whatever will work?

Edited by GaryOS on 7th Sep, 2008.

On 12th Nov, 2009 Paul S said:

I think Gary OS has taken over my role as the forum smart arse

i like the idea of equal lenght branchs on the exhaust manifold, but is it worth the effort on a 5 or 7 port head?

my first point is the standard metro turbo manifold has a shorter centre branch and is proven to work, i beleive mini williams uses the metro manifold and so do i. next is the mirage manifold that also has a shorter centre branch, this has been proven by Nic to work well ( that covers 3 out of the fasted 4 or 5 at avon park this year)

next, as the centre branch is siamesed (sorry spelling) does that meen the centre branch should be twice the lenght?

on the other hand if no one ever tried something out we would never develope faster cars

Thanks Turbo Shed. I am aware of the other two manifolds but as you say, why not try something new?

Given that the two centre ports are 180 degrees apart, my understanding is that if all branches were equal length then the inner branch pulses would fit in between the pulses from the outer branches, all being equal distances apart once they reach the turbo

I'm probably completely wrong though

Edited by GaryOS on 7th Sep, 2008.

On 12th Nov, 2009 Paul S said:

I think Gary OS has taken over my role as the forum smart arse

Hi,
Even with equal length runners on any engine the pulses will not all arrive at the turbo at the same time, so what difference will it make?
Cheers,
Gavin :)

VEMs Authorised Installer / Re-seller. K head kits now available!

WB/EGT gauges. Click here for customers write-up

Visit www.doyouneedabrain.co.uk

My Mini build diary

In theory there would be some hp gain by keeping the ports equal length.

With the Metro or Mirage manifolds, the pulse from the centre cylinders will get to the turbo earlier.

However, it's not like tuned length LCB legs and using pulse reflection to help the cylinder VE. Under boost, with signficant pressure between the exhaust valve and the turbine, I'm not convinced that the pulse has any real power to it.

The other problem is the practicality of getting the additional length of pipe in without incurring significant flow restriction or compromising the inlet.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Bat, is that due to the fact that the pulses will be pointing at different angles relative to the impellor, therefore making the distance of each branch to the impellor different? Or is it due to the turbulence among pulses once they join at the turbo flange ?

On 12th Nov, 2009 Paul S said:

I think Gary OS has taken over my role as the forum smart arse

Hi,
Firing order, one pulse comes after another, no two cylinders are on exhaust stroke at the same time (on a 4 cylinder anyway)
Cheers,
Gavin :)

VEMs Authorised Installer / Re-seller. K head kits now available!

WB/EGT gauges. Click here for customers write-up

Visit www.doyouneedabrain.co.uk

My Mini build diary

I would have thought that by far the most important aspect is to separate the inner port from the outer ones on a 5 (or 7) port head thus preventing interference during blowdown.
This is standard on a turbo manifold.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Thanks Paul, I suppose I have been equating this manifold design to a tuned length LCB

Bat, obvioulsy no two cylinders are on exhaust stroke at the same time. Isn't the sole purpose of an equal length manifold to maintain the spacing between pulses, which is created by them firing at different intervals? If cylinders fired at the same time then an equal lenth manifold would be the worst thing possible as the pulses would arrive at the turbo at exactly the same time

matnrach, phisically separating pulses does seem to be a bigger concern of siamese heads but I am just wondering if I should consider length aswell so that pulses are as far away from eachother as possible

On 12th Nov, 2009 Paul S said:

I think Gary OS has taken over my role as the forum smart arse

I agree that it would probably be better to have the pipes the same length but it would not be as significant as separating them. (hence the design of the standard manifold)
With the relatively low overlap of these engines, I would doubt there is an enormous amount to be had with dynamic exhaust tuning anyway.

The point of seperating exhaust pulses is to reduce lag because the individual 'puffs' from each cylinder are more effective than a constant 'puff' at acelorating the turbine wheel from standstill. That is almost a word for word quote from ForcedInductionPerformaceTuning Book..
PaulS having the manifold underpressure might make each pulse more effective because the denser gas will transmit it along the manifold?
I think the mirage would benifit greatly from having all headers pointing at the turbine flange rather than making the gass bash into the exhaust from the other cylinders and making an abrupt rigth angle turn into the turbo.. If bends in intercooler plumbing have a pressure loss across them then the exhaust manifolding will be no different!

haven't built one myself. The ones I have seen custom-made weren't on Minis, but on those, the fabricator made an effort to bring the pipes into a collector so that the pulses arrived in the same order as would contribute to turbo rotation. I don't see how to do that on a Mini 5 port head.

Here is one of zillions of articles which might help...
http://autospeed.com/cms/A_2605/article.html?popularArticle

Mine happens to have equal length runners on my 5 port... no idea if it helps or not, because I have not tried an alternative manifold. To make this one was a nightmare though...

Its a completely different design though, with the turbo mounted over the clutch housing.






Pete

Edited by pristic on 10th Sep, 2008.

1132cc Turbo ~120 bananas ATW - Fun Fun Fun

That's a beautiful manifold.

I'm not sure about what actual pressure you get in the ex manifold, but even a bit of a rise will cause the resonant frequency to rise and bugger up any equivalent of shockwave tuning. Besides, a spinning fan thing is hardly a good reflector.

I'd say it's for 2 reasons. Sticking the gas into the lowest possible area of pressure for flow and also an inertial effect. The whole lot will be pushed along better if each pulse is pushing/sucking the previous/last.

Actually getting the spikes of pressure is a different thing and very dependent on the exhaust lobe/valve/port sizes.

The tube area is kept the same for average flow reasons, most certainly not shockwaves. If you had the tubes the same size, the velocity in the central tube would be higher and perhaps interfere with the other "tubes"

An interesting point is that old downton LCB's used to have the Y piece even further back.....IIRC... Like 18" or so.

I'm sure it's much of a muchness for a turbo engine anyway. Bearing in mind fluid flow is about the most complicated area of maths and an exhaust manifold is going to really be very complex explaining every last wiggle of the graph.

I'd just whack it on and go for it.

Bugger off, I'm getting there.

Edited by GaryOS on 10th Sep, 2008.

On 12th Nov, 2009 Paul S said:

I think Gary OS has taken over my role as the forum smart arse

that shit needs to be heat wrapped pristic big time!

Hey Hario, no heat probs at all with it mate.
I dont actually use Heat wrap - dont like the stuff.
Ceramic coating - fine, heat wrap - I dont like it.

Pete

1132cc Turbo ~120 bananas ATW - Fun Fun Fun

I work at a thermal spray company and we recently coated some rollers for a textile company with something ceramic, it was for wear resistance not heat retention but still the ceramic powder that we coat with was £45 per kg. and we used about 2kg. per meter per roller"!
looks nice though lol