Cracked exhaust manifold, what to do??

[RNS11Z]

Well as the topic says my exhaust manifold has a crack in it. Im looking at upgrading to either a custom manifold or a Jap-version manifold. What prices are they and is there any recommened type for an rb25det with a t88-33d.

Any info would be much appreciated.

Thanks

H.

[Steve]

If at all possible, try and follow the design of the Jap manifolds to suit big singles, ie, 3/2/1.

I am pretty sure they do this to keep the gas velocity high when it hits the turbo, thus spooling it up quicker than longer/tuned lenght manifolds will.

I went tuned length, and ended up with a bit of a lag monster on a turbo that definately shouldnt be laggy, ie turbine housing matched to the engine.

[RNS11Z]

Thats what i have now steve 3-2-1. Still laggy. But i like lagg were as others dont. Im after something thats flows good.

[Steve]

Well what I am trying to say is you can have a tuned length system that flows well, but doesnt generate the exhaust gas velocity that a 3-2-1 does.

Most of the Jap manifolds I have seen (admittedly not alot) have very smooth joins and point the gas flow in the same direction before the join, not as they join - minimise interferance to maximise velocity and flow whilst keeping back pressure to a minimum. From what I have seen, seems to work best for a street tuned car.

But definately, make sure the angle of the join of the runners is as slight as possible, as I am pretty sure this is what is hurting mine (plus the added fact that long runners are being used).

I dont mind a bit of lag, but when its not doing much until 4500-5000rpm, well thats just a little late for me - I would prefer to have 1 bar in by 4000rpm where the RB25 hits the meat of its torque. Either that of I would have to get the bottom end balanced, solid lifters and valve springs to handle a much higher redline, which I dont want to do.

Like I said before, just my 2c worth based on getting the wrong manifold made - expensive mistake, minimum it will cost is for a turbine housing, but ultimately I am pretty sure it is not doing my top end any favours either.

[RNS11Z]

My bad i understand what you mean now. I had a look at a trust manifold for the t88-34d to suit an rb26 and its got short tuned runners, two exits for the external gate which joins to one (tuned length) and its 3-2-1 with the outlets pointing into one direction with a single flange though split into two sections ( when you look from the top you see 2 rectangle holes on the manifold of were the turbo sits). I may try to get one of these. Apparently it costs $800 for one.

H.

[Steve]

IMHO, it would be well worth getting the right manifold (ie, the one the manufacturer sells to suit)

Like I said before, if I hadnt been such a tight ares, I most likely woundnt have the less than spectacular results I have now. I really do wish I had spent that exra 3-400 dollars now, as to get my current setup running properly, it will cost alot more now when you reckon in dump, wastegate pipe, turbo to AFM, turbo to cooler and the cost of a whole new manifold - kicking myself:(

[Steve]

And $800 each, (assuming you need 2?) sounds pretty reasonable. If you could get one second hand perhaps it would work out about the same by the time you had them modified for GTS? (wouldnt take much?)

[Roy]

I know of a T88 manifold for an RB25DET that is for sale. It is a low mount Jap one.

PM me your details if interested and i will forward them onto my friend.

[RNS11Z]

Steve i meant its $800ea for the manifold. Its a 1 piece manifold.

[DoughBoy]

Steve i meant its $800ea for the manifold. Its a 1 piece manifold.

$800 is a pretty good price for a genuine Trust manifold. The last time I checked they were 190,000+ Yen rec. retail.

[RNS11Z]

Its not a genuine, its an exact copy made out of stainless steel. thats the catch

[DoughBoy]

Ah ok, just as a side note the HKS manifolds to suit T04 work as well - same stud pattern. Also the genuine HKS articles are made from 1 piece - no welds in any of the runners to disturb flow or provide a break point.

[Steve]

$800 for a stainless manifold, copy or not, is very bloody cheap - sounds too good to be true.

[RNS11Z]

doughyboy the manifold i looked at had no welds in any of the runners to disturb flow or provide a break point.

[DoughBoy]

Don't hesitate in buying it then. The cost of the stainless and getting it mandrel bent in 1 piece would cost alot more than $800, let alone the flanges + welding it up.

[Guest INASNT]

Don't hesitate in buying it then. The cost of the stainless and getting it mandrel bent in 1 piece would cost alot more than $800, let alone the flanges + welding it up.

Yes that price is way way cheap. Theres gotta be a catch somewhere, especially if it has been cast with no welds

[Sydneykid]

Hi Steve, help me out here....

Let me start off by giving my real world example. In an R31 GTSR the genuine Group A manifold has primaries that are equal length and very, very long, as the turbo is mounted both low and rearward. In fact I would say that they have the longest primary pipes on any turbo manifold I have ever seen. But the response is fantastic from that manifold. When we fitted a shorter high/front mount manifold the engine made less power. Plus it was no better in response, not that I could feel anyway and not that showed up in the boost curve on the dyno.

Now for the theory...........

You posted "Well what I am trying to say is you can have a tuned length system that flows well, but doesnt generate the exhaust gas velocity that a 3-2-1 does. "

Let’s ignore tuned length for a moment, while we examine energy?

Newton said energy can neither be gained nor lost. It can be converted into other forms of energy. The exhaust comes out of the exhaust port with X amount of energy (made up of Y amount of heat and Z amount of kinetic energy). How can less than X amount of energy arrive at the turbo?

Well, we could loose a bit of heat, it might escape from the pipes on the way. But we stop that with insulation, wrap or ceramic coating. We might also convert a little of the kinetic energy into more heat by way of friction on the bends and bumps in the pipes. But that would be minor if the pipes are relatively well made.

So if the turbine inlet is the same diameter, why wouldn't the same amount of energy reach it regardless of the shape, size, diameter or length of the primary pipes? Now if the manifold was so poorly made it had restrictions that caused excessive back pressure, then you could measure it. Which I have done a number of times. But I have yet to see a turbo manifold that has more back pressure at 4,000 rpm than it has at 7,000 rpm.

Thus if the maximum power of the engine is OK, I don't see how you can blame the manifold for slow boost rise.

When we add tuned length into the equation we end up with a different result, but only at the range of RPM that the length is tuned for. The problem is the supposedly “good” Japanese exhaust manifolds don’t have a tuned length that I can see would be of any use. The HKS one for a GTR, for example, has a tuned length at 5,250 rpm, now that is hardly useful. Too high for helping boost build and too low for increasing maximum power.

There are a number of respected turbo book writers that totally eschew the idea of running a tuned length exhaust manifold on a turbo car. They reckon it’s a waste of time, better to keep the back pressure as low as possible and ignore the tuning.

OK, that’s my 20 cents worth on this interesting thread….

[Steve]

I see what you are saying WRT gas velocity. I was originally told (by an exhaust 'expert') - that longer runners will bring boost on gentler than short runners, as the gasses slow the further they move from the engine.

This isnt the total reason for my post, here is another example of what I meant by flowing well and gas velocity - most jap manifilds (I have seen) run a short pipe after the gasses have been collected and before the turbine housing - OK I am a layman, but it appears to me that this pipe will do a similar job as the turbine scroll, in that it channels all the gasses together.

If you have a 6 to 1 collector, without a pipe coming off it, surely the gasses would still be suffering from turbulance to a greater extent than if the gasses were directed toward the turbine for a short way prior to entering the housing.

Am I so far out of left field its not funny, or would this have the effect of accelerating the gasses prior to entering the turbine housing, not as they hit the turbine housing - my guess is this will reduce turbulance, which would increase flow.

Sydneykid, how did the ID of the GTSR extractors compare to the highmount? Also, did all the exhaust gasses meet as they entered the scroll, or before 'jap style':)

[Roy]

Yes that price is way way cheap.  Theres gotta be a catch somewhere, especially if it has been cast with no welds

Be careful that the external welds hadnt simply been polished. Grab a small mirror, or even better a boroscope and look at the internals, the truth will quickly reveal itself

In an ideal world it would be great to have the sections of tube mandrel bent to minimise the amount of welds, but as long as the welds are full pen with no missed edges, no undercut, or weld concavity issues, etc etc, then having a manifold made from welded elbows shouldnt be an issue.

I think its kind of hard to applyenergy equations to an exhaust manifold where you have bends, Air & unburnt fuel-can you treat it as an ideal gas, mitred connections, possible weld dags, exhaust pulses etc, etc.

With experience you can probably ascertain what properties are negligible, but in the real world i suspect (could very well be talking out my a55 :Oops: ) that runner diameters, and blending of sections is most critical when looking at a manifold.