Custom 998 manifolds
Turbo Mini Forums

Actually got into the shed for the first time in aaaaages.....

started on this....




Need more plate so I can do the same for inlet too

Input welcomed!

Edited by wagon_rider on 20th Jan, 2014.

not sure i understand the turbo placement?
or is that plate not staying there?
looks like it could turn out real nice im interested in the intake manifold design

Mmm the plate is kind of tempory to help align the tubes but as also bit of rough idea where the turbo might go.

I dont have a bulkead to hit fortunatly! :)

That short centre runner will hit torque where you most need it.

I made that mistake.

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."

I did consider making it more equal length but I ran out of bend.....

maybe it's worth the effort?

I'm not sure if I exactly know the reasons behind it..... maybe it shoud go and research it

I made a similar manifold for my first 998 turbo in 2007.

Rolling road results showed a very flat torque curve. More recently I built a computer simulation of that engine and that also showed the very flat torque curve. Nothing wrong with a flat torque curve, but what was happening was that the short centre runner was sending positive pulses back to the cylinder at the wrong time, when the exhaust valve was open and that was killing the VE of cyls 2&3 in the 4-6k rev range.

Longer runners send the positive pulse back the valve when it is closed so you don't get the same problem.

Optimum lengths and diameters for a turbo manifold are the same as an LCB.

Edited by Paul S on 20th Jan, 2014.

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."

thats interesting to know as im gonna be making my own 998 manifold in a month or so. you saying they should be equal in length paul?

balls to the blower!!!



1330,stage3 head, gt1549, canems ecu, S/C box and drops

Is this why a side mount gets good results?

Edited by gr4h4m on 20th Jan, 2014.

I run a supercharger and I don't care the TB is on the wrong side.
VEMS + 12 PSI + Liquid Intercooler = Small Bore FUN!

I read an article on exhaust manifold fabrication years ago where the Williams team showed theirs being fabricated.
They went to great lengths to match the runner length, and had different lengths for different circuits to alter the engine characteristics.

Metric is for people who can't do fractions.

im keen to see if mine sinks or swims...

paul do you mean to take the same Y setup for #1+4 then run the long center #2+3 down much father just like the NA lcb is and simply make bends to fit it all in but maintain the same lengths for primary and secondary's?

perhaps you could take the center branch and make a turn straight down then a 180 back up with a short straight/small angle into the turbo?

if you get tight for space perhaps lay the u bend part flat on the block?

The issue is that if you are running 1 bar boost, then the pressure pulses in the exhaust manifold can range from 1 bar to 3 bar over a single engine cycle.

You can either kill the pulses with big pipes like the Mirage manifold or make use of them to improve performance with smaller diameter, longer pipes.

Short small diameter pipes on the centre branch will hurt the performance.

The optimum lengths/diameters are the same as an LCB. For the same reason. With an LCB the pulses come back to the exhaust valve in the negative state at valve overlap and positive when the valve is closed.

I've got some charts somewhere explaining the benefits. I'll dig them out later.

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."

This is a chart I drew up 3 years ago:


The red line is the simulation output for the original engine, calibrated against the NMS RR results.

The purple line shows the output after fitting a big valve head, something like 33/27 valves. The top end improved but something was holding it back lower down. Detailed analysis of the pressure pulses showed that centre cylinder VEs were well down and it was due to the short centre branch on the manifold.

The yellow line is with an LCB dimensioned manifold. A good 10 lbft improvement across the range.

The blue line is the result of optimising the inlet design and a few other tweeks. I've improved it even further since.

It's not just a computer model :)

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."

holy fuck. i guess just knocking one up to go in the elf is going to be a bit more difficult. there goes the sneakyness of not really noticing the turbo.
great

balls to the blower!!!



1330,stage3 head, gt1549, canems ecu, S/C box and drops

hmmm.....

You got me thinking now!

Do you any more pics from teh other side?

Pic of the manifold from the other side:


If I was doing another one, I would do it this way again:

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."

It does make for much better exhaust/ inlet piping routing too....

You'd say the second picture is better? I do like the piping there

Which turbo is that?

Thanks for your help, been very interesting input :)

Has anyone tried playing round with resonator type add on onto the shorter centre branch?
Effectively making it act longer (by changing the pressure pulse velocity), but without having to move the turbo.

As shown: http://somersf1.blogspot.co.uk/2012/07/red...-expansion.html
(Ignore the aerodynamic side of things in this case, the other reason they were used was because the coanda required the exhaust tip to be in a certain position and packaging wouldn't allow the correct lengths.)

Which manifold is which Paul? Is the top one the Mig one that ran briefly and the bottom one the latest design?

Fastest 998 mini in the world? 13.05 1/4 mile 106mph

www.twin-turbo.co.uk

oh my god. i was just going to chuck a mirage on and have done with it! if im only going to be running low boost on a mild motor, would pulsing affect it much?

balls to the blower!!!



1330,stage3 head, gt1549, canems ecu, S/C box and drops

Yes i moved to the darkside

Instagram @jdm_brett

Compared to that, how crap would my turbo phil ported metro manifold be? Just thinking because i could make my own manifold at a later date, right now I'm all for bolt on stuff so i can get used to everything.

Porting will always help but flow is generated by pressure differential across the valve. There is far more to be gained by managing the pressure pulses.

I drew up an estimate of the potential a few years ago:

http://www.turbominis.co.uk/forums/index.php?p=vt&tid=464288

For me, I will always seek the best performance for a given boost. Others will be happy to just turn the boost up.

Tuned inlets have the most benefit on the 5 port turbo, but you need the tuned length exhaust to get them to work at the optimum as can be seen by the blue torque curve above.

Stu, if you have fabrication facilities then a tuned length exhaust would be cheaper than a bought manifold. Mine cost around £100 at the most.

Edited by Paul S on 21st Jan, 2014.

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."