Looks Like Garrett Slipped Another Gtx30 Variant In .

[discopotato03]

Your call but I think GT28 stuff is a little small for an RB25 unless it's in a very mild state of tune , I definitely wouldn't swap your turbo for a GT28 based one in a GT28 housing like the HKS GTRS one for this reason . I got mine (GTRS kit) for a reasonable 2nd hand price off Wolverine because it was an easy install and has more potential than the std RB25 ceramic dryer . If 270 at the wheels = 300 at the crank then that's what I got with a few top end mods on E70 . Turbo on its knees .

GT30 turbines are about 6mm larger in diameter than the GT28 and in 84 trim size vs the 28s 76 trim . So GT30 turbine blades are going to act as longer torque arms than the GT28s ones which is better where possible to drive compressors into boost .

Again if you look at Garretts questionable turbine maps the gas flow through the GT30 0.63 AR is very similar to the GT28 in its larger 0.86 AR version - which is only available in T2 flange .

Hot side sizing is always a juggle of turbine speed vs exhaust flow restriction and the "restriction" options are basically housing AR size .

Since you can't easily get a 0.86 T28 housing in T3 flange the logical solution is to go GT30 in 0.63AR T3 flange .

Uninformed miss guided - mmm . No offense but I can't see any less restriction or potential boost control issues with the "big" 28 hot side vs the "small" 30 hot side so going 28 for the sameish flow doesn't make sense to me .

I think everyone agrees to get boost earlier the turbine has to spin faster sooner so the usual method is to go down in turbine size .

I feel sorry for the lack of love for that 0.63 housing and its hard to imagine Garrett making them to start with if they are so useless .

This much I can say , the 0.63AR will spin your turbine faster sooner and is the simplest option available . I can also say that people get 300 plus wheel KWs with the standard exhaust manifold so it probably isn't as useless as some would have you believe .

I see your point about an increased flow restriction at part throttle and off boost but engines lightly loaded generally have low exhaust flow , this is still way way better than Nissan had it standard and significantly better than a GTRS GT28 0.64 AR hot side .

Headers , not cheap or easy or as stealthy and I think a lot of though would go into a manifold that's going to excite a GT30 turbine in a 0.82 AR single scroll housing at low med revs . I think the best shot there would be with a twin scroll 0.82 housing but that's more money external gate/s and still an unknown quantity .

My instincts are telling me you want more shove (torque) further down the rev range so solving that and not being worried by dyno numbers is probably the path .

Mr Mafia did it , it was reliable , it didn't kill the engine and it was very likely driven and dyno'd hard . I don't remember him mentioning boost issues and while he did use WMI it all worked out - on a GT30 turbo with a larger compressor and compressor housing . AND with that poor unloved GT30 IW 0.63 AR turbine housing .

Numbers like 320 RWKW and something like 700 Nm of torque .

I think if you got to 270 RWKW with 450-500Nm of torque your engine would hardly choke to death . Mafias was an unopened R33 RB25 meaning standard pistons and they didn't turn into melted pizza cheese .

Gotta run but look into dramas with 0.63 housings because I reckon you'll find other factors involved .

[M@&k]

Don't get me wrong, I have no intention to change to GT28 turbine. I have every intention of getting the most out of the GTX3067. But rereading my post sounded otherwise. I only raised as a possible topic of discussion is all.

I was more thinking of someone reading this thread in the future and contemplating the GTX3067 vs the GTX2867, why they would go for the larger turbine. I understand conceptually that the GT30 hotside will let the engine breath better and should be better suited to the RB25. But if that is the case, surely this must show up in results somewhere?

[Dale FZ1]

Best bet would be to hunt around and find typical results for GTRS equipped RB25 running on 98. It's not a reasonable comparison for anything running on ethanol or using water injection.

Consult with Wolverine for input, but I reckon you'll find the GT28 turbine would limit you to somewhere around 235-245rwkW for a good one. You'd probably see the torque curve up a little earlier, and fall a little earlier, and the on-road drive experience quite a bit different to the GT30.

Best description of the GT30 setup is that it is progressive, and throttle position plays a bigger role in how the grunt is delivered. GT28 will do a lot more on part throttle, but doesn't seem to do much more on 100% throttle than say 65%.

[discopotato03]

I think it is possible to compare turbo systems with/without WMI before heat and pressure gang up and detonation sets in . It'd be a pretty piss poor turbo engine that wanted to detonate just as the turbo started creating boost and or at low boost pressures . That's why I suggested lower power and torque numbers (less heat/pressure) than Mr Mafia had and I think its reasonably obvious that a 3067 is going to be spinning faster for just about any air flow rate than a GT3076R 56T . A faster spinning turbine presents less flow resistance than a slower one for a given housing size .

Comparing a GT28 based turbo like a GTRS isn't going to teach anyone much about GT30 turbos even in smaller housings . The 28 turbine is ~ 6mm smaller with narrower tips and the nozzle and volute is nothing like a GT30 housing . NS111 and GT30 UHP turbines are nothing alike , chalk and cheese and everyone at Garrett that ever spoke of them said the 60mm NS111 is a more efficient animal than the GT30 UHP . If GT30 was the titz competition turbos would use them but they don't because they don't work very well in serious competition applications . They won't give us the real 30 series turbine so we soldier on with the high temp material GT30 diesel turbine in 10 blade form . I did once see a pic of a Garrett competition turbine that was even bigger than 60mm and I suspect there's a GT35 sized version of the NS111 or similar too .

Anyway IMO a GT30 turbine in a 0.63 AR turbine housing is going to work better than a GT28 in even the "big" GT28 0.86 turbine housing , not that there even is a T3 flanged version of that - from Garrett anyway . I agree that Garrett should have made a 0.73 AR GT30IW turbine housing and I'd have one if they did . I'm still not convinced the 0.63 is a waste of time if top end power isn't important to the user . In fact just today I changed from E42 to E85 so my next tune has the best chance of blitzing 300 plus on my car . If I'm still not convinced I'll consider trying a 0.63IW housing and if anyone wants to sell me a good used one PM me please .

I'd love to see one extrude honed to give it every chance of being a good road car system .

I don't imagine there is a lot of difference performance wise between a GT30 0.63 and a 2835 0.68 hot side . Not too many people bitched about surge or boost control issues with the 2835 Pro S and that's not exactly a GT30 0.82 is it . HKS sorted their turbo so it didn't f*ck up and by that I mean surge or have boost control ills . You'd think if the GTX67 hot side was going to turn to shit it would do so worse in GTX2867 form than GTX3067 form . Literally the same animal with a smaller turbine and turbine housing . I'm pretty sure that if you threw a GTX2867R in the HKS 0.64 AR IW T3 flanged housing at an RB25 it would have similar limitations top a GTRS (GT2871R 52T) or the KAI version of the 2835 mentioned in the other thread . If they were going to surge they'd be doing it on 4 and 6 cylinder engines and I don't hear anyone with SRs etc having issues .

Now your calls but I think the X67 cold side was designed to go to the limit of the GT28 hot side , then Garrett decided to turn it around and fit a turbine that would take that X67 cold side to its limit with less exhaust restriction . The GT30 0.63 hot side flows as well as the biggest GT28 hot side options get and its hard to guess how big a GT28 turbine housing would be to get anywhere near a GT30 0.82AR flow - if that were even remotely possible .

From past examples we know GT2871R 56T turbos were problematic , laggy and surged because Garrett didn't have port shrouded compressor housings for them . HKS fixed it with a 52T compressor and a port shrouded housing . The last issue was a T3 flanged turbine housing and they had these too .

Anyhow to each there own , I know what I'd do .

A .

[AngryRB]

interesting how you say the GT2871 56T turbo's were laggy and surged because I had one in highlflow form inside stock housings and I can tell you I had no surge issues, it pulled hard from 2500-3500 and then sizzled out thru to the top end with only

214kw.. so didn't have much to offer my engine. the GT3067 has pissed on that from my view point, although the tuner didn't

really push it to its potential in my case, so it should of gone 230kw's I think.

My thinking of this whole situation is that an RB25 is a naturally revving motor, it really likes revs and is

really suited to a turbo that goes hard after 3500rpm, anything before is really working against its nature. The

Mafia's setup should be the minimum setup in my opinion as it gave the motor some breathing right at its point of

wanting it ( around 3200rpm and over) , its not an RB30 so why try to make it like one with a small turbo combo

that runs out of puff up top..

Best to work with the engines natural characteristics which seems to be a free reving no restrictions 3500-7000rpm, more boost the more angry it gets and this is the sweet spot, GT30 is like a perfect marriage and with E85 its orgasmic, just

look at matts results. The GTX3067 is an attempt to force the RB25 into being a torgue monster which it isn't

really, but might come close with a 0.63 rear but the top end will suffer which the motor really doesn't suit.

Edited July 12, 2014 by AngryRB

[M@&k]

I hear what your saying about this potentially working against the RB25's nature, but my aim is to make the power more accessible in street driving conditions. I rarely get the opportunity to rev out to 7k without drawing unwanted attention.

Speaking of which, I did get that rare opportunity today. Found some quiet space with a friend and took some vids of pulls in 2nd and 3rd and 0-100 runs. Did a best of 5.75s as measured by GKTech thingy, but certainly room for improvement with a decent driver who can get off the line better.

The runs had us both giggling like school girls all afternoon.

Will post up vids when I get a chance.

[discopotato03]

I'd have to re read Mafias 300 Kw adventures thread to get his exact opinions but I do remember him saying it pulled like a V8 from I think around 3-7000 . If it wasn't a volumetrically efficient system it wouldn't have worked as well as it did . I have no doubt that up top things would have been toasty without the water mist cooling but he never cooked or melted anything and his leak down tests were always good .

Now , I think the perception of the 0.63 housing being a choker chain is in relation to larger compressor ends because that's all we had until recently .

The idea of an RB25 being a rev engine isn't something Nissan worried about considering the miniscule turbine side they get standard . I find when I have my tune right that I get surprisingly good part throttle torque and flexibility and that's off boost .

You can argue it any way you like but the smallest GT30 hot side is a big increase over the standard turbo and still higher flowing than the few GT28 alternatives .

Now I don't disagree that easy exhaust outflow is a good thing but not getting a power boost from positive air pressure where you want it is hardly a good thing . IMO it isn't good enough to tell people that's how it is so lump it . They are the ones who know what they want and they are the ones they have to please . If they can't run full bore up the road at 6-7000 revs the bit they sacrificed down low is all for nothing .

One day we will hopefully get a comparison of a 2835 Pro S and a GTX3067R 0.63 AR . If the result is similar what does that say about its hot side - or peoples acceptance of the HKS turbo ?

[AngryRB]

That's awesome, its the school girl giggles that make everything worth doing, great to hear your enjoying it

Was she getting loose? love it when they break traction in anger

[M@&k]

Conditions were good yesterday so it was holding traction just. Will certainly kick out sideways in 2nd if road is not ideal. But certainly doesn't bake the tyres, which makes it easier to drive really. Just foot to the floor and concentrate on gear shifts. If I lived further out of the city, I don't think I would be as worried about the response as I am. More 80+km/hr speed limits and less traffic. But in the city where I am I'd like to be able access the power sooner...

Anyway, here's vids as promised. Taken with my phone so not brilliant quality. Tried to get both tacho and boost reading in frame.

Edit: doesn't seem to want to embed third one

Edited July 13, 2014 by M@&k

[M@&k]

3rd gear run:

[discopotato03]

I went back into Mafias 300 RWKW Adventures thread to see why he chose the 0.63 AR housing on his GT3076R . He said he'd had butchered VG30 and RB25 turbine housings then a 0.82 and finally the 0.63 one .

What he was looking to avoid was turbine lag and detonation . Pre WMI his wasn't knocking up to around 270 RWKW , someone asked him what the difference was and he said 10 degrees of timing and 30-40 RWKW . He mentions 20 degrees of full load timing with WMI and 700 Nm from 3100 to the redline .

This was ~ 7 years ago and you'd have to think that within its limits a GTX3067R is hardly going to be less efficient than a GT3076R .

My parting comment is that water methanol injection cannot stop an engine choking on its own exhaust even if you drown detonation . If it really had been super restricted it wouldn't have been happy to rev and been very reluctant to hold 18 pounds of boost much past the upper mid range .

The X67 wheel has to be spinning faster to pump the same weight of air at any point than a GT76 one so the turbine turning faster also creates less resistance to exhaust flow .

If Mafia claims he was only just reaching the limits of his turbine housing its hard to see how the lil X67 is going to create mayhem .

A .

[The Mafia]

I eventually pushed my setup to 22psi and I would guess it was making around 320rwkw. It went hard and early. Always on full at about 3400rpm in 2nd, 3200rpm in third. Never had issues with detonation.

One point I should make is that I could not run boost lower than 18psi even though I had a 12psi actuator. The rear housing wasn't the extra ported internal WG, only the standard one.

Felt the same on the RB25/30 with a 0.82 rear housing. Just more down low. Same rwkw, but torque went from 720nm to 850nm+

[GTScotT]

Before I say this I will say I am happy to receive a PM asking for more detail, but I would really say this needs to be done on ones own skills. I would not pay a mechanic to set this up for me, or recommend anyone else for the same.

- Hang a Tial 44 off the factory manifold. Contact Stao to get him to mod a manifold and post it to you.

- Buy a Kando 8cm housing (move up to 10cm if it chokes, they are CHEAP and personally I would be happy to trial and error).

- Make a 3.5" downpipe and reduce to 3" at the cat.

Prepare for serious levels of carnage and ball tearing torque.

The money you can get for your .82 iwg will mostly pay for the above.

[badhairdave]

He must have also run one on an SR......it was a 3071 .63 with external gate and Unigroup cams.

Posts 83 & 85 (where Dave chimes in).

http://www.nissansilvia.com/forums/index.php?showtopic=413229&st=60

I really should have changed my nic to sandyvag based on those responses.

The same turbo was on the sr and the RB, post rebuild as the compressor kissed the housing when I crashed the 32. Turbine housing was changed as I didn't want boost control issues on the sr that I had on the rb25.

I can't really provide informed comment on the x67 as I haven't spent enough time looking at results or comp maps.

I will say that the iw .63 housing is to be avoided on an rb25 when running a small compressor with a gt30 turbine.

Scott's suggestion above is probably the most cost effective way of getting a lower a/r housing with enough wastegate flow onto an rb25 and is similar to what I have recommended to friends recently.

If you have to spend more to fit a gtx3067 there are more cost effective solutions out there with similar results.

Edited July 13, 2014 by badhairdave

[badhairdave]

Oops, Tapatalk double...

Edited July 13, 2014 by badhairdave

[discopotato03]

Well I'd like to hear what Mr Maffia thinks because his results largely fly in the face of what many here think . Maybe the laws of physics are different in Mackay .

I don't see how anyone can reason that a big trim turbine in a reasonably big housing is gonna spin a modestly sized compressor fast enough to boost properly in a car that doesn't get many opportunities to stretch its legs . 4000 rev boost threshold on a 67mm compressor - piss off . How f**ked would an Evo be if it did that with the 67mm compressors they have - factory ...

[Dale FZ1]

By a turn of coincidence I also drove Mafia's 0.63 setup, at the same time my R33 was running a 0.87 Pro S setup.

There was definitely more down lower in the rpm range ie 2500-3500 but beyond that point things largely evened out a bit and the 0.63 had done its best by 6500. Not surprisingly the less restrictive 0.87 was freer breathing and it pulled to 7500. Mine was running on 18psi and capped at 290rwkW so when you think of the area under the curve available from 3000-6000 Mafia's car with ~310 was clearly the quicker car. It was all about where it delivered its torque. And it was a good thing for the street.

But that also indicates how it was choking up higher. And 20 degrees of full load timing admittedly is probably a touch more than what I'd consider "typical" for a 0.8x A/R setup. I'd be thinking more likely 16-19.

What goes unsaid is how much work was required to get Jono's setup to hold boost.

Lastly, Mitsubishi has its sights set on a World Rally Championship with the Evo turbo. Production runs were big and they had to get it right. Be interesting to know how many of the 0.63 IW housings were produced by Honeywell.

[M@&k]

If you have to spend more to fit a gtx3067 there are more cost effective solutions out there with similar results.

Agreed. While external gate would provide better performance (when does it not?), it's not what I set out to achieve with this setup. I was looking for something like the Pro S performance; early boost, ~260rwkW and stockish look. It sounds like this is possible with a change to the 0.63 A/R housing if decent boost control can be achieved. I was previously considering an external gate off the housing, but decided it was overkill for what I was chasing.

Looking at the comp map for the GTX67 it's in peak efficiency up around 20psi. The boost control issues seen by others may not be an issue if running at this pressure anyway is my thinking.

By a turn of coincidence I also drove Mafia's 0.63 setup, ...

There was definitely more down lower in the rpm range ie 2500-3500 but beyond that point things largely evened out a bit and the 0.63 had done its best by 6500.

...

I'd be happy with this. I do most of my driving below 5000rpm realistically. To be honest I prefer not to have the temptation to wind it out to redline all the time

Edit: Going back and looking at the videos, doesn't seem anywhere near as impressive as in real life, haha.

Edited July 14, 2014 by M@&k

[discopotato03]

It generally never does .

And there we have it , the user wants GT2835 Pro S like performance but with something available off the shelf and as compact/stealthy as is reasonably possible .

We know that you can hide a GT3076R 0.82 reasonably easily and the only non std bits needed is an inlet tube like Scottys and a spacer plate . This assumes you keep all the standard airbox and snorkel - and the airflow metre . If you get creative with a turbine housing and dump shield most people won't see too much out of the ordinary . Hiding a GTX3067R should be easier because from memory you don't need the manifold spacer and the compressor housing is smaller . The inlet boss is smaller , I think 3"/76mm vs 4"/100mm) . The AFM from memory is 80mm so not much bigger than the turbos inlet .

Dale you must have been one of the few who had the larger Pro S turbine housing on your 2835 Pro because I believe they came standard with the 0.68 housing - in kit form .

Now of all the GT3071Rs , 2835 was the HKS handle for cropped GT3071Rs , their Pro Ss gave the best all round results and I put that down to their turbine housing (both sizes) being GT30 format and their comp housings being port shrouded .

I suspect the people with full sized GT3071Rs and Garrett housings (both ends) tried and to a degree failed to get good all round results because they were trying to fix issues bcaused by lack of port shrouded compressor housings . This can really easily cause surge and boost control problems and changing turbine housing AR size (generally larger) often isn't a complete fix . It bleeding obvious , to me anyway , that HKS found the problems and fixed them before they became grumbling customers dissatisfaction .

HKS didn't always use big or even medium sized turbine housings but most of what the marketed worked - remember the tag states Power and Response . Some early efforts like GT2540s and GT2510 were superseded by GTRS and GTSS but that's another story .

To keep this close to home ask Wolverine what he thinks of his GT2835 Pro S in 0.68 AR form on an RB25DET . I'll suggest nobody had boost control or surge issues with these turbos using the smaller 0.68 turbine housing , very few people changed to the larger 0.87 turbine housing . I have a fleeting memory of someone doing this 10 or more years ago here and I think it typically moved the power range up evenly but not by a huge amount . It got them closer to their target number which was the point of the exercise .

I think the GTX2867 and 3067 are aimed squarely at being GT2871 56T and GT3071 replacements , one for one . Most GTXs of similar wheel size to the earlier GT versions are intended to make more rather than the same power more efficiently , X67 wheel is (base diameter anyway) 4mm smaller than the GT71mm wheel it's replacing . I think the comp maps even show similar flow rates GT71 to X67 compressors .

With Evolution Lancers yes Ralliart and MHI were trying hard to make a two litre engine perform like a high torque rev limited diesel . All the manufacturers were hard pushed to turn in some cases like Ford with their RS500 Sierra engine a rev donk into a torque one . The restrictors forced they to use everything they could think of to make them boost from nowhere and haul within the restrictors flow ceiling .

Dragged away for domestic bore now , back soon cheers A .

[M@&k]

Does anyone have any solid info on this mythical update 0.63 IWG housing from Garrett with larger wastegate? I've only read that they had a -M part number suffix or something, and that they all should have this from now on.

Any idea when the change came in? Any idea on how to identify? Anyone actually know what changed?