Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Turbo Selection for R32 GTR

Off Topic Discussion

Forum Posts



Tech Articles

If it's not really about tuning or wiring. Then it belongs in here.

= Resolved threads


Hey guys.

So not happy at all with the stock twin turbos on the GTR, not getting to 13psi until after 4000rpm and limited to 13-14psi

Was watching a video on youtube twins vs single where they compare some -7's to a GTX3582r and found the gtx own out in all areas... Now going off the Dyno graph they only show road speed but knowing the ratios can see boost is on about 4000rpm and makes abit over 600rwhp...

So this is about as laggy as the current setup but much better power... As this is 90% a street car with the odd drag meet boost threshold is a high priority. They are using a 1.02 rear housing... How much earlier would you expect the boost curve to be and how much less power if I used to a .6 or .8 rear housing?

I'd be happy with 4-500rwhp especially if I could get boost really early ... Ie before 3000rpm

What would you expect for maybe even a smaller turbo, GTX3076?

From my own experience you simply aren't going to get an RB26 to see full boost by 3000 rpm with anything resembling a worthwhile turbo combination. The GTX3582 is a reasonable choice but to be honest I'd be more inclined to consider a Borg Warner EFR8374.

If you go with the GTX3582 I wouldn't consider the 0.63 housing - It will be too small for for an RB26 and result in too much back pressure. For comparison I've gone from a 0.63-0.82 housing on the GT35 on a few 2.0 litre 4G63's and always gained power and seen zero difference in boost threshold (I know that might seem counterintuitive). Typically the difference between the 0.82 and 1.02 housings is only a few hundred rpm at most and I'd be inclined to think the 0.82 might be a little marginal for a 2.6 litre.

I think one of the key aspects of the RB26 is its ability to rev, with a factory redline of 8000 rpm. If you can reach full boost by 3500 rpm or thereabouts and the rev limit is 8000, this gives you a pretty wide useable power band.

Borg Warner efr 8374 will be a good choice .

I did a rb26 with these turbo a couple weeks a go and make 660 bhp and 545 lb/ft torque at 1.6 bars

I'm putting a package for 4-500WHP for an R33GTR right now and every turbo supplier (UK) are sending me quotes for GTX2860's and GTX2871R's for a twin setup

What is it about RB26's that make them particularly lazy?

Here is a graph comparing my R33 with stock turbo and Nistune with 14psi and my R32 GTR with stock twins and boost turned up to 13psi Nistune.

You can see the 33 makes more power all the way up to 5700rpm, boost is up by 2500 rpm. The GTR is my new daily to replace the 33 but was quite dissapointed really... The 33 feels alot faster and alot more responsive.

Since its 95% a street car low down response is very important... Not after a 600+ horsepower. I would be happy with a turbo maybe not much bigger then the 33 stock turbo but could handle 18-20psi and made power to 7000rpm, even if it only made 350-400rwhp. if it got on boost like the 33... Would Cam gears tuned to down low power be quite beneficial for this also?

Attached Files

What is it about RB26's that make them particularly lazy?

Here is a graph comparing my R33 with stock turbo and Nistune with 14psi and my R32 GTR with stock twins and boost turned up to 13psi Nistune.

OK so a far things need to be considered with that comparison first off:

1 - The GTS is running more boost

2 - The GTR has more transmission losses

3 - The GTS turbo will be showing the MAX it can do there, while the GTR has capacity for even more power - so the flow matching isn't compatible, hence more lag being natural

4 - As well as being smaller/rated at less flow the GTS turbo is a more modern unit, if its an S2 it has a ceramic exhaust side and composite lightweight compressor wheel and ball bearing core. The GTR will be journal bearing old T25ish things.

5 - A key one - The RB25 has VCT which imho more than makes up for the 100cc

So in this comparison the "lazy" thing ultimately comes down to the fact you are comparing dyno plots of two cars focussed on different things. The GTR is also a fair bit heavier car with the same size engine - it's honestly a pretty weird car to buy for commuting tbh, they have a small (for it's weight) old school engine with a setup known to be at its best between 5000-8000rpm.

What fuel are you going to be running the car on? The only sensible options I can think of for what you seem to want are all single turbo - though it all comes down to what you want to spend. Without VCT, lower compression and with a heavier car it will ALWAYS feel lazier than a GTS. I think your best odds of getting something running which will do what you seem to be looking for is getting a decent modern ball bearing turbo with modest power targets, so the likes of a Garrett GTX3071R/Xona Rotor XR5456/Borg Warner EFR7163 and maybe get a smaller-than-normal runner exhaust manifold made to get the exh velocity up a little. They will all do over 500hp @ crank on pump gas and substantially outspool the stock turbos but NONE of them will match the stock R33 for spool from the basement and again you just won't get that... bearing in mind you don't have 200kw @ wheels right now with it.

Don't ever expect the RB26 to push a 1500ish kg 4wd car hard under 3500rpm though, especially on pump gas. You need more displacement and/or some variable cam timing to give the motor itself more ability to push imho. I'd not recommend any smaller turbo than those ones for that motor, if they aren't going to do what you need it's time for more stroke or a different car for daily duties ;)

Hey Lith, it's a Journal bearing turbo on the gts, I believe the GTR turbos are basically the same with ceramic exhaust wheels

Both turbos are limited to about 14psi before they fall apart due to the ceramic wheels... The GTR could go another PSI but the boost threshold would be pretty much the same.

Dyno runs were in RWD mode in the GTR. Still might be slightly more loss I guess tho, but that wouldn't change the boost curve

a 33gts and a 32 GTR are actually pretty close to the same weight of just over 1400kg.. But again won't change the boost curve

In Reguard to VCT, I understand the GTR is more setup for high rpm hence it makes power right up to 8000rpm, which is why I was asking about adjustable cam gears... I'm rarely above 6500rpm and would happily sacrifice top end power for some

More low end...

My main reason for comparison of the two is thinking maybe I would be better off getting a single turbo the same size if not slightly larger then the stock GTS turbo with a split pulse manifold... So yeh maybe a GTX30/71

Or would a compound turbo setup be suitible and give boost at low rpm?

Oh right - so you don't have the stock GTS turbo which is ceramic turbine and ball bearing? What have you got on it now?

14psi isn't a magic fall apart boost level, it's what is generally accepted to be a safe stopping point at least for the GTR turbos - the GTS turbo on 14psi is working WAYYYY harder than the GTR twins and arguably much more likely to fail, we normally stop around 11/12psi with the R33 unit... either way, its running more boost than the RB26 through much of that comparison which will account for some of the difference.

I never said weight would change the boost curve or power delivery, I said it would affect the way the car would drive - the heavier car makes a similarly performing engine feel doughier. Are you going by provided specs to figure both cars weigh the same or have you actually had them on a bridge? Any GTR I've ever weighed have been WAY heavier than a GTS, we did my old R33 GTS25t and an R32 GTR with driver on the same day, same scales and the GTR was around 150kg heavier. No, the GTR driver didn't weigh 200kg haha.

I suppose dialling cams could help you there a bit. If you are thinking about compound turbos then just go RB30, cheaper and will be a better more reliable and easier to implement solution - will fix all your problems.

Oh sorry I thought the stock turbo was journal bearing but your right they are Ball bearing...

RB30/26 is definitely the end goal for this car.

Being mainly a fabrication business I'd actually like to do a compound setup just to show off my work a little and be abit different. just not sure on choosing turbo size ... But a single turbo is definitely much easier to work with some have to weigh that up.

Sorry Viper, my initial suggestion of the EFR8374 was based solely on your mention of the GT3582 - I'd assumed this was the sort of power requirements you had in mind, and in that case the EFR pretty much does everything the GT35 does but better.

Since you're not wanting that much power, the EFR8374 would just result in unnecessary lag. Right, with that out of the way, please continue :)

Just reviving an old thread here... As you know I love low down torque so have decided to build a RB30/76 (when funds allow) so while im just trying to work out a budget and what components Ill need, Turbo being one of them...

Just curious on your comment of the EFR turbo doing everything a GTX35 does but better... what ways is it better?

Ive been using Borg warners Match Bot and from doing the inputs as best I can it looks like the EFR8374 will be the best match and I compared the plotted points to the GTX3582 Compressor map and the EFR is actually a few points more efficent in each spot so I guess thats one reason.

But onto the subject of the Matchbot... these are the figures Ive come up with:


Now on the compressor map you can see points 1 and 2 are off the map, They look as tho they are within the Surge line just the map doesnt extend that far... Just this just mean that they will be in lower efficiency zones? But since the turbo speed will be low and heat input low its not really an issue?

Also by default the VE numbers on the match bot get higher and higher as RPM climbs (and the tutorial video says to leave the numbers as they are unless you have dyno figures) But I was under the impression that VE will peak at peak torque and drop from there so curious why they have it like this and is there a good reason?

Any ideas on how to get better BSFC numbers for a certain motor? is there anything I can log on my current 2.6L to calculate the BSFC? Altho I noticed changing these numbers does not change the compressor or turbine plot points at all only the estimated power outputs.

Also not sure about choosing the target boost pressures at lower RPM's? how do I know what boost pressure the setup is actually possible to achieve?

Lastly I noticed Garret have released their GenII GTX turbo's, how do these compare to the EFS?

Thanks in Advance


If money is no object you can NOT beat EFR for spool and boost response! You can contact Full-Race they will help you, I have an account with them. Also since its kind of a competitor site I wont link to it but there is an interview with Jeff Raicer of Full Race on Youtube with The Tuner School (part 3 interview part 4 is Matchbot) and he goes thru how to use matchbot. BSFC numbers are really just an estimate, unless you have an engine dyno or Mainline with Fuel Flow meter is hard to measure it. I had a r33 gtst with Full Race Twin Scroll Manifold and a gt3582r and ATP twin scroll housing for it, I will try to get the graph for you just gotta start up my dyno computer.

Shot a email off to Full-Race, will let you know what they say.

Get a Reply:

Hello Clint.

I talked to our engineer and here's what he said.

8374 1.05 a/r is excellent on the RB26 and outstanding on the 3.0L / 26 hybrid. This is my recommendation.

8374 1.45 will work on the 3.4L stroker to acheive his targetted powerband, but turbo speed will become an issue if hes looking for more power.. He will need to use a speed sensor and keep engine rpm below 6500 like he said.

9174 1.45 is ideal for the 3.4L stroker and 9180 1.45 if speed sensor will be used.

External wastegate is mandatory on 3.0L and 3.4L - I recommend (2), one wastegate per 3-cylinder bank. EFR internal gate is 0.92 a/r and the engine will choke.

We usually reply within 12hrs (often sooner)

Need Help?

Need help choosing a course?

Experiencing website difficulties?

Or need to contact us for any other reason?