The Garrett Gt30r Thread - Lots Of Info And Q's

[GTScotT]

Dont forget that motor flows more exhaust gas into the cooling jackets then it does out the exhaust. The last word to be put in the same sentence as "astron" or "4g54b" is efficient.

[discopotato03]

Ahh yess , bit of a diesel bore stroke ratio . The balance shaft doubled as a second flywheel reminding the pistons to go back down again .

Agricultural head but at least Magna EFI manifolds were available .

Sigma , don't know what you can say about those ...

A .

[GTScotT]

lol funnily enough they are still in production those motors..... in a DOHC diesel form.

Back on topic, thoughts on response of a 3076 52T .82 vs 56T .63?

and any reason to believe boost control in a .63 IW would not suffer using a 52T? i have my doubts really, but thought we could brainstorm.

Realistically my doubts are based on that of the 56T 3071, that has problems enough with boost control, a smaller trim yet bigger wheel would potentially flow quite similar.. Or perhaps it would just have to be tested before it could be verified as an appropriate theory.

[discopotato03]

Well like I've said in the recent past I can't say how one of these GT3037/GT3076R turbos in 52 compressor trim will perform because I don't know of anyone who has one up and running yet . The theory is that it should be a bit more responsive that the 56T one all else being equal . I can't put boost threshold numbers to it because that could be misleading to others .

What I can say is that in the past people who were concerned with getting boost a bit sooner , and had access to them , chose the 3037 in 52 trim because it suited their needs .

Ditto with 3037's twinned on RB26's , HKS usually shipped 52T ones because any response improvement is gold on an engine like that .

The jury is still out on turbine housing sizes , with garrett the choice is 0.63 or 0.82 though the 0.73 size would have been a better all rounder on a street RB25 IMO . Trouble is no one makes a T3 flanged GT30 IW housing in that size .

I dunno with the 71mm compressor , the best maps I've seen are of the 52T one in the 0.60 A/R T04B comp housing . If you look closely at pics of the RB variant of the HKS GTRS's (GT2871R52T) comp housing it looks like a custom port shrouded T04B 0.60 A/R styled thing . They obviously thought that running the GT28 turbine/housing combination would lead to compressor surge with these or why would they have gone to the effort and expense to produce that PS housing . The SR type GTRS just uses the Nissan/Garrett type T04B comp housing .

Again as I said recently the turbo engineers in the US used to reckon that in a compressor wheel family the middle trim size usually gave the best all round results though there are other minor alterations that can be made to compressor wheels to achieve things sometimes asked for by large manufacturers like tip height .

My personal , and unconfirmed , view is that big trim compressors and turbines are not optimal things and more some sort of compromise for cost or packaging reaons . A port shrouded compressor housing costs more to make than a plain one so this alone would make a GT3037/GT3076R a more expensive turbo to market than the "real" GT3071R .

People probably think I sing a lot of praise for expensive HKS spec Garrett turbos . I do because they went to some effort to develop turbo options to suit specific engines/cars plus they aim for "Power and Response" .

I draw the line at paying elevated prices which is why I'm always exploring options to emulate what they've done with more affordable components . Probably being a cheapscate but if someone elses done the hard yards proving what does and doesn't work why not try to benefit from it .

Really Garrett are the ones not making the full range of PS comp housings or coming forward with as many turbine housing options as HKS have .

Lastly with the full turbine sized GT3071R , I think we be sure that whoever did the turbo type testing would have tried a unit in this form and decided that it didn't suit their requirements . They did make them in croped turbine form and AFAIK this is where that croped GT30 turbine idea came from .

Again I don't think they would have gone to the trouble to do this without a good reason . Admittedly it's no biggie to machine a different profile on an existing turbine and the housing has to be machined anyway so this is what they've obviously done .

Garrett could machine their GT30 IW turbine housings to suit the croped turbine wheel which they produce anyway , I'd say it either hasn't occured to them or they haven't seen a market for it - yet .

Just on this if it ever eventuated it'd give you the option to remachine it for the full sized GT30 turbine later if you decided to . Room to grow without forking out for another expensive turbine housing , I like that idea . Always good to have options .

Lastly the T3 flange footprint is getting on nowdays and I can't think of too many OE things still using it .

I believe turbocharger development is leaning towards smaller engines running in higher states of tune and many of them diesels . Sort of a tug a war between engines running lean enough to meet noise and exhaust emissions , fuel consumption , and as much cooled EGR as they can make fire .

This is an interesting article by JE at Auto Speed about late developments with EGR , most interesting but still a bureaucratic band aid being forced upon the manufacturers by the leaches in this world .

http://autospeed.com/A_110577/cms/article.html

LOL bored yet ?

Cheers A .

Edited February 28, 2010 by discopotato03

[scotty nm35]

Informative article that Disco, It would be interesting to have a map adjustable variable output egr with some sort of cooling (large oil cooler) and run it up on the dyno to check the losses in power and decreased fuel consumption. Seems it may help lower high exhaust temps also and improve efficiency with undersized turbos, especially useful if you are planning a trip to the epa.

[Lithium]

On the GT3071R vs GT3076R 56t vs GT3076R 52t - I'd be interested to see if anyone can bring out a dyno plot which shows evidence of a GT3071R/GT2835R coming on conspicuously faster than a equivalent turbine a/r GT3076R.... before hypothesising that the GT3076R has a major fault in spool versus power production on an RB25 it'd be good to determine how weak it is down low compared to the next smaller.

I am sure its gotta be worth some salt that HKS chose to supply a 56t compressor with their GT3037 units designated for RB25s, they weren't the silliest bunch and they obviously had the option of a 52t version. I am reasonably sure there are people on this forum who have gone from GT2835s to GT3076Rs as well, instead of working on theory - surely the people who have been there and done that would be a good indicator?

For what its worth, me and a mate running a healthy stock turbo R33 (FMIC, S-AFC etc) did a very scientific test the other day - accelerate hard from 70kph in 3rd gear side by side, with my car running straight wastegate boost. He pulled about a foot or so on me before mine made up the boost difference and started pulling him back.

OK it could get more in depth than that, but ALL of the options we are discussing will be laggier than a stock unit - and pulling from lower revs than you'd ever "play" from (unless you completely fail at driving) the stocker hardly has a huge edge over a .82a/r 56T lag monster in a real life comparison, I think there are unnecessarily twisted undies. Just get a 56T GT3076R, if you are too ponsy for that lag then stay stock, if you are unsure if the lag is worth the power then get a GT3582R and know that at least when the lag is over you have a very real risk of dying.

[GTScotT]

Me+Turbos=Bored? NEVER... All these nights spent hawking every active turbo thread I can find on SAU lolll... Not in vain!

Mr Lith, I feel your angst.... I was once an idiot on a forum based around SR20 RWDs... Hunting out CA18 bashers and preaching its good name (lol) and being the only moron on the forum with a FWD, better yet an SR20 lol. But dont worry, I am a firm believer in both FWD and CA, just as im sure you are your turbo

Ohhh mannnn, Im sorry... I wouldnt say it if you didnt atleast half know me hahahaha... you know Im really only kidding.

Your turbo setup is great, yet as discussed in private, I feel it wants to rev a whole lot more than it currently is with the .82 IW. Rev your motor to 9 and run a shorter final drive and I would think it was godlike.

Really I only want to utilise the factory rev limit, its too costly to start playing with it for me, so a .6x based turbine housing on that size of turbo is optimal. Remember boost pressure is not relative to flow. While a .82 housing might see boost come on as early and start making good power, the .63 will have higher shaft speeds. So while the boost is similar the airflow would be exponential, which at the end of the day is pure torque.

Same shit with a stock turbo, it might spool as fast or faster but, flows NOTHING in comparison.

For reference, a HKS GT2835R pro S, the owner of the vid has stated its mods in the comments and its 272rwkw. Nothing about the power or speed seems outlandish to me, so its credible.

Testimony to the power of that turbo. FYI, that video shows a turbo that IMO is very responsive. I am still waiting to see as definitive results from a GTRS based car, perhaps a member can 'find' some and post it up?

[Lithium]

Haha yeah I think we understand each other

Will have a look when I got home...

Something I would be curious to see, is a stock turbo RB25 setup running E85...

[GTScotT]

Not a fan of E85.. Not a fan of lean tuning full stop... Fuel and timing, not less of both.....

Plus, when e85 came out I always thought it was meant to run heaps richer and enable you to run LOTS more aggressive on the timing and charge temps.. But instead they run it at 98 ron lean mixtures... where stoich is like 5:1.

But this isnt an E85 thread

Turbo it up! (wishes he had the brains to do engineering at uni)

[JonnoHR31]

no, they leave the printouts in ulp afr to simplify it for people who are used to ulp mixtures. if you leave your wideband on ulp and tune it to 12:1 for e85 it'll be a lot richer than it would with ulp

[discopotato03]

Informative article that Disco, It would be interesting to have a map adjustable variable output egr with some sort of cooling (large oil cooler) and run it up on the dyno to check the losses in power and decreased fuel consumption. Seems it may help lower high exhaust temps also and improve efficiency with undersized turbos, especially useful if you are planning a trip to the epa.

I'd say the manufacturers are getting a lot out of EGR and ironically it must be a good way to suppress detonation by controlling heat and pressure .

I'm wondering if some enterprising tuner can come up with an engine management system that can toss in varied amounts of exhaust gas before the turbo/s rather that pulling timing or drowning the thing with fuel . The engine would have to run very clean because carboned up intercoolers don't help much .

It appears to be the way to run lean and not get bitten by combustion/exhaust gas temperatures , the only EG cooler I've seen is on a diesel ute at work .

A .

[l0WRB]

Is my turbo dodgy? the cast markings on my turbo pictured on the left are different to every other GT3076R markings

[Dale FZ1]

Looks like this thread is meandering a bit, difficult to fathom direct relevance of Mitsu 4G54B Astrons and KA24 to application of a GT30xx to RB25.

I’ll add the bit that I can, from direct experience with GT3037 Pro S (0.87 turbine), GT3037 (0.63 IW Garrett turbine, water/meth injection) GT2835 Pro S (0.68 turbine, Pon Cams), and a piddly GT2871 high flow onto RB25 in R33 chassis, stock gearing. Max outputs of each were ~ 290rwkW, 300+rwkW, 265rwkW, 225rwkW and not running outrageous boost pressures or tunes to achieve those results. Reference to roll-on response is in 4th gear.

Head to head the 3037 and 2835 DID NOT feel significantly different in the 2000-3500rpm range. There was a slight advantage to the 2835 in roll-on response, made you feel there was perhaps 5-10hp but nothing startling. Above that the 3037 just had bigger lungs and there was a clear advantage that widened beyond 5000rpm.

I’ve said many times that the 3037 never felt laggy. The uninformed/inexperienced might use the term “laggy” when referring to matching a 3037 and RB25, but it’s just not the case. It’s linear, progressive, and with the 0.8x turbine housing gives a stock engine some pretty good high rpm breathing capacity so making and holding max power (in my instance) across 6300-7300rpm. That is holding peak for over 1000rpm, significant in my books. A superb match generally, but it did not quite have the zippy response between 2000-2500rpm associated with stock turbo and/or small capacity high flow turbos. It telegraphed exactly what it was going to deliver, and then delivered on the promise. Advice from a prominent Qld based Garrett distributor was that the 0.8x turbine was the best setup for an RB25 because amongst other things it permitted pressure drop across the head, and hence good scavenging. Also told to expect ~400rwhp @ 18psi. A comment made in response to whether it would deliver strong results under 2500rpm was to be prepared to change down a gear for best results. Dyno results of 390rwhp, some club motorsport, and 10000km on the road showed he was right.

With the 0.63 housing I felt there was a slightly earlier sharpness in torque production, but you could feel things taper around 7000rpm. Not snappy, but it came on a bit earlier and peaked/fell away a bit earlier than the 0.87 turbine setup. Direct advice from a Garrett engineer was that the 0.63 housing would peak around 400hp, and not get the best (highest) output from the compressor but would give stronger transient response. Dyno results showing 400+rwhp and seat of the pants showed he was right.

The 2835 was ”nice” which sort of damns it with faint praise. Cams let it breathe, and it produced power beyond 7500rpm. But (and my opinion only), it felt neither zippy down low, nor I’m-going-to-jail fast up top. That setup did need revs, and didn’t feel overwhelmingly powerful. It was fun and driveable, but for the money I’d want more.

The 2871 high flow was in many ways a surprise packet, because of its strong torque beginning at 2000rpm, and running to 5500rpm. Didn’t look fast or sound fast, but covered ground fast and easily. Relatively low power was easy to handle and not at all intimidating. And it could beat any of the other three setups in a roll-on from 60 to 140.

With respect to compressor trim sizes, I suspect it makes little difference to transient response capability in roll-on circumstance. A couple of posts out there trying to compare the 35 and 37 series compressors by trim size are wide of the mark. Best to get the actual inducer measurement, which tells most about the “gulp” capacity. Bigger mouth = bigger gulp = more top end power. That’s why restrictors work so effectively in rally cars to contain max power outputs. Comparison between the 56T 3071 and 48T 3037 (evidently now out of production btw…) shows their compressor inducers are within 0.15mm of each other, and they should hence be capable of roughly the same max flow capability. Required shaft speeds and moment of inertia will be different, and so they should perform differently – from a theoretical perspective. I had suggested to others looking for ~ 270rwkW that the 48T 3037 teamed with a 0.63IW housing would be a lively unit and better than a 3071 due to the lower required shaft speed to deliver a given air mass (ie power level). That is now impossible to tell unless someone built one special order from parts as a leap of faith. And it would not come cheaply.

The smaller inducer for a 52T GT3037 will offer less max flow than the 56T 30307, and perhaps slightly less inclination to surge – good except that the 56T 3037 doesn’t surge on an RB25 in my experience. I’d concede that at sub 400rwhp the 52T might well offer a technical advantage, with marginally less power (drive) requirements placed on the turbine. There could be some difference in feel to a discerning driver, but until Disco (and a few others apparently) get theirs up and running that is conjecture. Most important is that the chosen compressor is running at peak efficiency across the range of peak engine torque, but not pushed so hard/fast that it’s cavitating and overheating the intake charge at max engine speeds. Not too difficult to grasp.

Rule of thumb offered to me by both engineers mentioned before was that achieving enough compressor flow and efficiency is fairly straightforward but the choice of turbine plays a BIG role in how it responds, and over what engine speed range it runs best. And that is consistent with the turbobygarrett.com tutorials on turbocharger matching. In this case, smaller A/R = lower effective usable engine rpm, bigger A/R pushes things upwards. It really does depend on the use you intend to put the car to as to what is the best choice, not whether it makes 280 or 300rwkW. Power density or area under the curve counts the most.

I know of only two people who used the 0.63 GT30 turbine combination. Mafia ran up around 19psi and did not observe boost spiking or running away. BHDave was looking to keep things down around 16psi and did experience issues. Both of them reported superb low end response – not surprisingly. Running a free floating (EW equipped) 0.63 turbine would offer a satisfactory cure, at a cost for those who demand that sort of torque delivery and usable engine speeds. If I HAD to have a 25DET running stock diff gearing, it is probably the path I would look at – provided I could wear the extra expense of a fabricated manifold and EW. And that also opens up the possibility of the ATP 0.78 A/R split pulse turbine option…

Lithium’s comment about too many twisted undies is accurate. More hand wringing than action, and the 3071 is in my view the orphan that Wolverine referred to it as. Hypergear do offer their bolt-on version of a GT-RS (52T GT2871) for a safe bet of ~250rwkW and quick low rpm response / high power density. Check it at http://www.skylinesaustralia.com/forums/Ch...hp-t309952.html I’d back it to cover ground the quickest and easiest in general road driving, and probably the least expensive $:hp ratio. Skylinecouple has been running the GT3071 for some time, and his preference is to fit a GT3076 simply because it offers more headroom for flow and some power increase where the 3071 really can’t unless it is pushed beyond its efficient speed range.

Final word: I don’t see the point in chasing middle ground between a GT-RS (or equivalent) and the GT3037 (trim size does not much matter). Go one or the other, and if you go for the big stick GT30, choose the turbine housing size according to use the car will be put to. And if funds/legalities permit, take the EW option.

[Lithium]

^ Fully agree with all the above, though I would highlight that the GT-RS as something that you should try for 250kw as an absolute max - some people like to think of them as something you can push up to 20psi but I really wouldn't run one much beyond 1bar, personally. Not with that little turbine flow on a 2.5litre... I'd treat it as a stock turbo setup that the RB25 should have received in the first place.

[chrisR34GTT]

Would any of you know if a HKS 2835 Pro S would bolt up to an aftermarket JJR 3" bellmouth front/dump? Or is the "snail" dump required and then a custom front pipe to suit?

[discopotato03]

Dale I agree with much of what you say but I still think the jury is out on compressor and turbine trim sizes .

When I speak of compressor wheels from a trim point of I view I don't look at the wheels mass or innertia aspects so much , more the mass of the air they are trying to pump . Air has mass and the more a rotary compressor wheel can screw its way through the greater the work load the compressors power source has to shoulder .

The difference between inducer and exducer must have compressor performance variables . Firstly the inducer because thats what the atmosphere has exposed to it , realworld air pumps do NOT suck - they create an area of lower pressure and then atmospheric pressure pushes air in .

I am not a fluid engineer so I can't quote facts on inducer/exducer centrifugal rates but I do know that the larger a wheels diameter is the higher its tip speed is for a given wheel speed .

We don't have a lot of choice with turbine trims because its not very likely anyone has the equipment to change what the turbo manufacturers supply .

I can go back into what I think about Garretts generally large trim GT turbines but I done that here too many times already . Also mentioned that they have woken up to the fact with some non OE performance turbos reducing the trim size from 84 to 78 (GT37R's and GT40088R's) . They could do it with GT30 and GT35 turbines as well but I reckon they won't because if power potential fell for increased response reasons people wouldn't buy them . Mores law and maximums numbers sell .

I also want to see how the 52T GT3037/GT3076R performs but my engine has a few top end mods so it should respond differently to a factory std one . Fingers crossed , for a long time now , that the porting larger exhaust valves Poncams and slightly higher CR will help it breathe better and make a bit more torque on and off boost .

This next bit goes completly off this topic but is still turbo engine related , if it bores you don't read it .

The daily hack is an old 86 Subaru L Series RX Turbo , flat 1800 2 valve SOHC per bank thing with really agricultural porting and factory manifolds (both) .

The Wedge or Vortex had available a better inlet manifold for these EA82T engines so I got one of these fitted .

The heads have siamesed exhaust ports , one ea , and unfortunately flat fours fire the adjacent cylinders in each bank so with siamesed exhaust ports the exhaust pulsing into the exhaust manifold is all screwed up - too much damping . I had a stainless header made up for it because the std thing is a horror - very restrictive and not much room to grow flow or turbo wise . Anyhow the staino header is much les restrictive and I decided to have a Garrett GT2554R fitted except I got GCG to supply that turbo with the larger 76T NS111 GT28 turbine - in an S15 Ni Resist 0.64 A/R turbine housing .

The theory was that removing the restrictions from the hot side of this engine would let it breathe and rev . Also the turbo cams were exchanged for the NA ones because they are slightly longer period and a tad more overlap . Static CR was raised from a pathetic 7.7 to 8.1 , most we could get for the budget .

The engine makes more torque because it does breathe better but boost response is woefull . I got too hungry with turbine flow and the engine can't spin the turbo fast enough to make boost where its needed . I put this down to prehistoric heads and not as much exhaust energy as I'd hoped it to have .

Plan B is to go back to the smallest or GT25 turbine to make the limited exhaust energy transform itself into turbine speed .

The point of this spiel is that you walk a fine line between driving a turbine and not having an unacceptable restriction when the engine is running at full revs and boost . I suspect that this may be part of the problem though to a lesser degree than the full turbine sized GT3071R .

Now an RB25 is a far cry from the lawnmower tech EA82T and will make more power for its capacity everywhere . They don't have infinite exhaust energy and the juggle is getting a turbo on them to boost early enough to make them feel good and not fall over up heigh . Sorry folks you can't have the moon and especially with a single scroll single wastegate turbine housing manifold combination .

I think RB25DETs are good engines and these GT3037/GT3076Rs sound like a real good combination/compromise .

When I get the daily to a state acceptable to me the R33 will get some resources thrown at it .

In time , A .

[GTScotT]

Drop an EJ25 into the leone and get back to the program

I need to learn from your epic win/faux pas

[Lithium]

Drop an EZ30 into the leone and get back to the program

Do I have to set you on the right track with everything?

[discopotato03]

I don't want to go further off this track than is necessary people , ..... but .

No , don't like EJ's . Had I known what I know now it would have been an EJ20 or 22 . The Subaru experience has been ... different ... and I know why some are Rex proof . Less corners to paint yourself into but they're still there . EJ/EZ , more flat engines and trust me after inline fours and sixes you don't want to go there . Don't believe the marketing BS people , two heads a wacky firing order crap manifolding glass gearboxes and on it goes .

The L series is light (1070 Kg) and has an AWD gearbox which was optional on OS RX models , no it isn't the viscous center diff EJ box either . RX ones have a dog clutch diff lock and are a bolt in conversion .

Also trying to work more modern modern turbos into older smaller engines helps keep the feet on the ground with boost and torque needed lower down because they don't make much by themselves .

Just to be 1% on topic I laught loud at those trying to put GT3076R's on EJ20's , really laggy and brave with an aluminium split cased open deck "block" and fragile driveline . EJ is more modern and better breathing than EA but the cork heads join the two adjacent sets of exhaust ports together in their header manifolds and screw it up AGAIN . OS there is a twin scroll header (OE) available and twin scroll/twin gate turbos to go on them , can't for the life of me work out why the Rexers don't all use them , or fab their own headers and hang a pair of externals off them . TS EJ is far easier than TS RB .

Lastly , EJ25 is famous for head gasket issues . The rod is too short for the stroke so you don't get proper use of all 25 beans anyway .

EZ's are not really a performance engine IMO , I doubt you'd get DOHC RB30 performance out of them . Just an excuse for the 3.0L sticker on the porkier ritzier Zoobs .

Anyhow a GTR intercooler is on the cards and will have to look into pipe kits to make that happen in the near future . Next after that should be the turbo and a trip to the dyno to dial it all in .

A .

[Lithium]

FINALLY. Pulled finger, organised some dyno time with my car for tomorrow... tuning it myself. Fingers crossed all goes smooth and I can post a proper dyno chart for this thing! Hasn't been dyno tuned in YEARS!