Jump to content
SAU Community

Recommended Posts

Hi guys, After blowing a turbo gasket, I’m in the process of fitting 2 stock RB20 turbos to my Nissan Syline R32 GTSt. All the work is been done at a engineering firm in Bunbury WA where my dad works.

We are making 2 separate manifolds one for each turbo. Front 3 exhaust outlets to one and the back 3 to the other turbo. There is going to be a balance pipe that goes between both manifolds to equalise pressure into both turbos. We see there is no need to have 2 wastegates so one will be blocked off. All water and oil lines are been ‘T’ . There will be 3” pipes coming off the back of the turbos converging into a 4” exhaust system. Air intake for the turbos will consist on the 1 pod filter, splitting into 2 pipes after the afm then to the turbos. The outlets for the turbos will converge into one tube entering the front mount intercooler then engine.

This project is well underway, and if all goes well will be completed in a month. Before starting this project I had 200bhp running a stock turbo with 13psi, with the intercooler and a 3” exhaust. I have had it down the drag strip with a time of 13.7 @ 167km/h.

I was interested if any one had any comments on how laggy it will be, the power output I might achieve, or any concerns about what im doing. All comments will be appreciated. thanks, Ewan.

Link to comment
https://www.sau.com.au/forums/topic/19560-rb20dett-twinturbo2litre/
Share on other sites

  • Replies 141
  • Created
  • Last Reply

Top Posters In This Topic

Did some reading on twin turbo setups. Where twins will help, if run in parrallel, not talking about sequential, is if you are aiming for xxx amount of flow (power), rather than fitting one large turbo, two smaller ones with a combined flowequal to the flow of the larger one will spool up quicker - that is if the turbine wheels are half the size etc. Its not exact in the real world, because of so many other variables, anyway starting to digress....

This has to do with the energy required to move the wheel varying exponentially, rather than linearly, as the radius increases - so two smaller wheels (turbine) require less energy to spin at the same speed as one larger wheel.

So, where the advantage will lie, is if you were wanting to fit, for exampe, one 500 hp turbo, two 250hp turbos of half the size exactly, will come on boost quicker than the one bigger turbo.

Perhaps it would be worth seeing if you cannot find a couple of smaller turbos to fit?, second hand GTR ones or even smaller if possible? This would give you boost alot quicker. Remeber that while the rb20 turbo isnt huge, it is designed for a 2L engine, so if you put 2 of them on, you probably have a car that would be good for the drags, not so much of a streeter.

For wastegate, perhaps it would be possible to run one wastegate on the crossover pipe between the manifolds? but probably alot better to just use the stadard wastegate on the turbos - these could be controlled by one boost controller, with the line t-ed no probs.

As rev suggested, exhaust cam, also a good manifold design, decent dump pipes, good exhaust, 'high' flow cat etc will all make boost earlier.

I saw the other thread on the same topic first, but rev210 is exactly right. Each turbo will have to spin up to the original shaft RPM to generate the original boost, but with only half the exhaust volume to do it. As a result boost threshold RPM will double, and rate of boost rise will be just about half.

It will be pretty awsome at the very top end though. Some smaller a/r exhaust housings would be what I would be looking for, or a couple of stock GTR turbos at a friendly price.

Further to Steve's comments on large single versus twins has to do with the relative needs of airflow and boost pressure for various engines.

If you look at any flow map it leans to the right, with the most efficient operating islands in the middle somewhere. Now this means a given sized turbo will give you more available flow as the pressure increases, and you operate higher on the flow-map.

It also means that at low boost levels, there will be a far lower efficient flow available.

Now suppose you have a fairly large capacity engine and you only want to run low boost. You could either fit a huge single, or two smaller turbos, each supplying only half the flow.

Ford on the XR6 have gone the huge single way with a monstrous GT40 turbo, which everyone thinks is oversized, but almost certainly is not. Or Nissan with the GTR have two very small turbos to do much the same thing. In each case there is pretty high airflow requirement at fairly low stock type boost levels.

On the other hand if you have a small capacity engine and you want to run very high boost, a large single is going to be better because it will fit better onto the flow-map. High flow and high boost go well together with a large single.

So as Steve says, if you want 500 BHP it really depends on the boost level you need to make that power. If it is going on an eight litre monster that only needs 4psi boost, go multiple turbos.

If its a one litre engine running 50 psi boost or something, a single will get it done best.

AFM will be dependant on how much air flow, ie, how much power you are making - no point forking out dollars if you havent had the car on a dyno. Lots of other things to consider, like fuel system, clutch, ECU, gearbox (perhaps) etc - more important than doing an AFM straight up.

What other mods do you have already?

R32 GTSTT, what sort of power are you looking for? RB20/25 afm's are good for ~370bhp (thats at the engine). If you want any more, you'll need a Z32 or better afm. Whats your power goal? Chasing power without setting a power goal is always going to end in disaster.

A twin turbo RB20 is a different way to go about things and I like it. Twin RB20 turbos is the wrong choice though. Look for twin RB26DETT turbs (T28 flange) or twin VG30DETT turbs (T25 flange), they'll be a better match.

Originally posted by broken

i would get a aftermarket computer and ditch the airflow meter completely for a map sensor.

Why? A 80mm pipe with filament running down the centre isn't going to be a physical restriction, and afm's are much better at measuring engine load than map sensors. Work with the ecu you've got R32 GTSTT. Its more than capable.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Latest Posts

    • No, you're wrong, and you've always been wrong about this. The Nismo has 2 sets of openings. One is a real 2-way, and the other is a 1-way. There is no 1.5-way possible with the ramps that they offer. A real 1.5-way does exist. That Cusco stuff I posted is a prime example. If the forward drive ramps are, say 55°, and the overrun ramps are, say, 30°, then you will get about half as much LSD effect on overrun than you do on drive. It is real, it realy works. OK, you're slightly right. The Nismo has 55° and 45° ramps on the 2-way, so it does offer less LSD effect on overrun. But, I think that just means that they've (probably) sensibly established that you do not want actually equal LSD effect on overrun. You just want "quite a lot, but not quite as much as the drive LSD effect".
    • Just wanted to unearth this and post my baby with the new front ❤️😝 Took her to my wedding rehearsal today. Next up is getting wide skirts (after wedding)
    • Yea, that is what I was getting at in my ramblings too. The nismo one actually is a 1.5 way and a 1 way. They don't do a *2* way because a true *2* way would have equal ramp angles. Or is that a true 1.5 way? Realistically I think a "1.5 way" does not actually exist. A diff can either lock in two directions or one. It also doesn't help that a LOT of people in Australia speak about 1.5 way diffs are referring to their 1 way diff.
    • Well, the trouble with that ^^ is: The configuration shown is absolutely a 1-way, not a 1.5-way. There is no way that a 1.5-way can be said to offer LSD action only on acceleration. If Nismo cannot get that right, then it is impossible to believe their documentation. That ^ is not a 1.5 way setup. That is a 1-way.   And so now I have allowed all doubts to flourish and have gone back to look at the MotoIQ video. I originally made the mistake of believing him when he said "this is a 1.5-way" at the ~6:10 mark. Because what he did was take the gear assembly out of the 2-way opening and just rotate it one place to the left to drop it into the 1-way opening. When he dropped it in there, the cam was "backwards" compared to the correct orientation shown in all other photos of that config. The flat shold have been facing the 1° ramp side of the opening, not the 55° ramp side. And I thought, "gee that's cute", but I was concerned at the time, when he put the other ring back on, that the gap between the rings looked like it was wider then in the 2-way config. And then I said a lot of things in my long post on Tuesday that could only make sense if the guy from MotoIQ was correct about what he'd done. BUT... I have now done my homework. I grabbed a frame of the video with the 2-way config, and then grabbed another with the "1.5-way" config, snipped out the cam and opening of that frame and just pasted it direct on top of the 2-way config. I scaled it so that the triangular opening was almost exactly the same height in both. AND.... the gap between the plates is wider with the cam installed in the triangualr opening backwards. That is.... it cannot go together that way. There would be massive force on the plates all the time, if you could even reassemble it.  So, My statement on the matter? The Nismo diff is actually only a 2-way and 1-way. There is no 1.5-way option in it, regardless of what they say. Here's a photo of a real 1.5-way ramp opening from Cusco (along with the 1 way option). And the full set of 1 through 2 way options from their racing diff, which is not same-same as what we'd typically be using, but...the cams work the same. A little blurry, but it comes from this Cusco doc, which is quite helpful. AND.... Cusco do in fact do what I suggested would be sensible, which is to have rings that do 1 and 1.5, and 1.5 and 2. Separately.  
×
×
  • Create New...