My 300rwkw Adventures

[Lithium]

My little 2c worth, what I thought long and hard about when I decided what parts I was going to use was actually "What to I want the thing to drive like?" instead of "How much power on the dyno do I want?" etc.

For ME after driving lots of different cars etc, I decided that having a little bit less down low is far from a bad thing with a R33 - having full boost at 2500rpm makes for quite a boring drive, fellow RB25 owners are probably well used to RB20 owners raving about how their cars are more fun to drive because of the more aggressive power delivery. I also found it annoying having positive pressure cruising up the gentlest of inclines, so a bit more lag was a good thing. The 56T still has positive pressure by 2000rpm, and by 3000rpm its giving a decisive push in the back even though it takes until about 4000rpm before its turned into Hyde. I've never been in something RB20 powered that gives as much acceleration at 3000rpm as I get with mine, and the owners of those cars are perfectly happy with what they own. Its not a dig at RB20s, its just using the fact that the "What is acceptable lag" is a relative term.

I guess what I am saying is, there are factory performance cars with much less meat down low to a 56T GT3076R which are running similar redlines - maybe people have based their perception on whats a good turbo for an R33 too much on the fact that they come from factory with a shockingly undersized unit? I really really reckon that the 56T made the car SOOO much nicer to drive, even when not giving it a bootfull and haven't had anyone drive it and think otherwise (or at least tell me haha). The best way of working out what you'll actually enjoy - this is especially worth considering if you have never actually been in a 300kw @ wheels car, is going in someones car with a similar setup to what you'd consider and make sure the way it behaves how you like.

For some, a GT2530 which whacks you HARD from the start and falls over early in the rev range might be their idea of fun - while for others a .82a/r GT3582R which is OKAYYYY up until 4000rpm, where it decides its time for to meet your maker - and won't stop trying to send you there until near 8000rpm. The problem is looking at numbers on paper won't really tell you, you really need to experience it. You may find that looking at dyno sheets, lag looks like you have a compromise - while on the road when your driving it its more of a lull before the storm.

[The Mafia]

Excellent post - I totally agree.

On another note, I drove DaleFZ1's GT2871 powered R33. Went really well off the line, and still had heaps of power to the limiter (6400rpm I think?)

Between that and my R33 running 300rwkw, I couldn't notice all that much lag. It was more of a case of having to squeeze the throttle a bit more on mine to get it to spool a bit more. Power hit a fair bit harder, but thats obviously because I'm running 300rwkw, and Dale's is running 225rwkw.

I honestly don't think twin scrolls are worth it when you look at the cost of $$$ just to make a bit of power in the mid range. In fourth grear, mine is producing 700nm at 3100rpm or there abouts, and holds that much torque until the limiter.

Why would you nee to spend another $2k+ to get a bit more torque, and no more top end? I'd use the money on the 3.0l rebuild. Then you have all the torque in the world, no lag at all on the GT3076 56 trim and 0.82 rear, and yeah, a very nice car to drive.

Just ask Cubes - aren't you making over 10psi in 1st gear? lol. Madman.

I bet you wish that was a AWD..

[Cubes]

Mine just pulled 268rwkw on 15.5psi @ 6000rp; power drops to 252rwkw by 7000rpm. Definitely has plenty left in it.

Z32 afm @ 4.73v; 20degree's ign. timing at peak power. Pushing in another 2degree's (22) only picked up 2.5rwkw (270.5rwkw) so it must be near MTBT?

If we pushed more boost in the power wouldn't really improve all that much considering and the curve began getting all lumpy.

Its been suggested the std R32 RB25DE exhaust cam is too small (232duration 7.3mm lift) and has reached its limit.

We did see 285rwkw but the curve was lumpy.

Fingers crossed its not the onset of valve float.

Feels awesome to drive; never bogs/lags even in first. Just gets up and goes with a strong linear pull that becomes more intense as rev's pick up. With the boost controller connected 3rd gear roll ons from 60km/h = lack of traction with boost peaking at ~3000rpm.

The garrett actuator is quite lazy to bring boost on but it does feel nice to drive and most definitely helps those rears hook up and no doubt doesn't prevents the rb20t gearbox smashing near instantly.

As it is now first 2 gears wheel spin but I can control it with the right foot; third gear over 4500rpm also starts to wheel spin up. Time for some decent tyres.

[wolverine]

far from being an expert but the lift on those cams seems tiny. i am amazed you made as much power as you have. 300rwkw on the cards with more duration and lift?!?!

[Lithium]

Nice results cubes So peak power arrived on 6000rpm? Man, look forward to seeing how mine comes out when I can allow it to rev out cleanly. Sounds like lots of fun to drive haha. Hope the wavey curve isn't anything major, does your boost curve waver to match?

far from being an expert but the lift on those cams seems tiny. i am amazed you made as much power as you have. 300rwkw on the cards with more duration and lift?!?!

Yep those cams ARE tiny. Port sizes on the R32 head aren't great either... car would still be a beast on the road but has plenty left in it.

[Cubes]

Yes peak power arrives at 6000rpm which I am stoked about and there is basically zero boost creep.

Maybe all that port matching and headwork has helped somewhat.

The R32 Rb25DE heads are only different on the inlet; exhaust is the same size.

The inlet ports are ever so slightly smaller and of a different more round cylinder type shape.

The inlet of my head has been ported/port matched and valves deshrouded so no doubt definitely helps flow.

Feels great to drive; at the moment I can't pick that power begins falling over after 6000rpm; feels as if it pulls strong until the 7k rev cut.

The wavey is a sharp jaggered wave; the boost curve wave I think is because of the long feed from the plenum to the actuator.

That will be sorted with an ebc.

The cheapo ball/spring boost controller was next to useless; it would spike up and then tail off to a couple pounds above normal wastegate pressure.

+ If I left it on it would have most definitely smashed the rb20t gearbox to pieces rather quick. :S

The tuner had another car on his dyno with a similiar setup and with a set of 256degree poncams it went from the same 270kw to 290kw on the same boost.

He's got an rb30det same head etc going on in a couple of weeks with a set of reground cams so if that shows good results I'll go with the cheaper $220 regrind option. If results are not that crash hot I'll throw a pair of 260duration+ cams at it with decent lift as the buckets have already been machined.

[Rolls]

I guess the next question is is that amount of power even usable on the street? If its lighting up 3rd what's the point heh.

[Cubes]

At the moment its grabbling at the road in third on most road surfaces, some it just spins up but it is fairly controllable with the right foot; with the boost controller on it it was lighting up the rears. The boost vs accel position is quite linear; so when it begins to spin up you back off a little then feed it back in.

BUT in saying that Im not running flash tyres nor is the diff in that good condition. They are cheapo Federal 535 225/50/16's.

Throw a set of semi's on it and it will hook up much better.

[Adriano]

Curious cubes, where is peak torque?

[Cubes]

Straight off the actuator peak torque is around ~4500rpm. With the boost controller connected peak torque was ~3300rpm.

Interesting we pushed in another 2 degree's advance, no power was picked up anywhere apart from the last 500rpm where it picked up 3rwkw. So pulled that 2 degree's out and left it as is.

With 18-19psi in it made almost 550nm with 15.5 it made ~490nm.

[Dale FZ1]

Sounds like a really worthwhile build Cubes - nice work.

If you have it delivering bulk grunt in a driveable, tractable manner then that to me is what makes a car work. Getting it to the ground is always the key, so it sounds like you're on the edge of "too much"

As Mafia said - you'd nearly be wishing it had 4wd.

My question is - did you run into any knock problems during the tuning process?

[Cubes]

No knock issues. It made the power super easy; so easy we thought bah an easy 300rwkw.

Power started to get all wobbly and spikey once ~275rwkw was reached; from memory with more boost it was making around 275-280rwkw by ~5000rpm so I think its definitely got a 300rwkw in it once its sorted.

If its valve springs or is indeed the std cams I guess I'll find out.

Its interesting to note a mates RB30DET gt35r .82 running 17psi with ~8:1 static comp was making an identical 270rwkw power but with much less ign advance ~12degree's. The exact same thing happened to his; push more boost in it did very little and the curve got all messy.

So more boost bigger turbo but the same power. So my slightly higher static comp + head work + hours spent port matching must have helped.

Shaun is 99% sure its an airflow/cams problem.

Feels strange reving the car past 5500rpm after driving it to 5000-5500rpm with the stock sized turbo for so long.

[discopotato03]

Cubes you may need to somehow probe the exhaust manifold for pressure readings and compare them to your inlet manifold pressure readings .

I seem to remember people saying that the std type exhaust manifold may be hitting the wall at somewhere around 300 Kw .

Also as you'd know not too many people have used a GT3076R on an RB30DET and the choice of turbine housing A/R would depend on what you wanted to achieve with your engine . In a crude sort of way its a bit like the RB25 ceramic BB turbo and its std turbine housing or the larger VG30 one . No question the 1.06 A/R GT30 IW housing would drop your turbine inlet pressure some but whats it going to do to your boost threshold . People with RB30's running GT35R's and the larger housing tell me that the boost threshold is around 32-3400 though this should be lower with a GT30R and similar sized turbine housing .

Anyhow to your engine you have a couple of reasonably easy upgrades in cams springs and maybe a turbine housing . The thing you really need to know is that turbine inlet pressure because if its rising way out of proportion to inlet manifold pressure then thats a major problem . Also once you get some reasonably high boost pressure behind your inlet valves (particularly with NA springs) it wants to crack them open .

Always something , cheers A .

[Cubes]

Hey Disco,

270rwkw is a fair fetch away from the 300-320rwkw brick wall of the std exh. manifold. Its still got a little more left in it yet.

The problem 90% sure lies with the std cams but more importantly its cam timing/centerlines.

I am running GTR valve springs so they 'should' be ok but I wouldn't be surprised.

As I think I mentioned in another post a mates GTS4 making 270rwkw with a GT35r .82 hit the exact same brick wall at 270rwkw with the std cams. Pushed more boost in the curve went all lumpy. The problem was literally IDENTICAL!! No detonation problems just the power curve began getting all shakey. And funny enough Z32 afm voltage sitting at 4.7volts as is mine so identical airflow.

Sky30 ran a similar setup gt35r .82 BUT he ran with the slightly larger RB20DET exhaust cam (240duration 7.8mm lift) + it has a difference centerline; he pushed a touch over 300rwkw; he then dropped in a set of 260degree cams which only improved spool. No real peak difference with regards to rpm and power. So hit the limit of the std exh manifold.

I really don't believe scroll inlet pressure is a problem with my spiky/lumpy curve. The gt30 rear with decent fuel has pushed 360rwkw odd without resulting in a lumpy curve. The problem of excessive scroll pressure shows as not being able to run decent ignition timing and the requirement of better fuel. Of which isn't happening.

At 15.5psi making 268rwkw we pushed in another 2degree's of ignition timing and only picked up 2rwkw; pushed in another 1 degree' picked up no power pushed in another 1 and saw a little bit of detonation. So its fairly happy where it is. Definitely not an indication that there's excessive scroll pressures.

By dropping the larger turbine housing on the GT30 when pushed to its limit I would only see an additional 20hp peak power.

But the trade off being off idle spool, part throttle and low load (1st & 2nd gear) response and general lively feel of the car is most definitely scarified; much more noticable than the 20hp increase at peak.

Tune the boost curve to suit how well the car is hooking up. Makes for a much nicer drive.

There's nothing worse than a lazy feeling motor at part throttle. Of which the larger turbine housings do cause.

I was told if I want the 1.06 sell the turbo and drop a GT35r .82 on it as its spool characteristics will feel the same BUT offer a hell of a lot more top end. I'm not after more than 300rwkw as the cars not 4wd and relatively light so I don't see a need to run a turbo that offers such power especially when it sacrifices mid range.

But yes.. I have this gut feeling its the valve springs. I've seen that wobbly curve many times before.

[@sau]

Nah, I don't think its luck, its all about setup and how careful you are.

Things that can cause failure:

1. Tune

2. Driver

3. Bouncing off the Rev Limiter all the time

4. Cheap parts

5. Identifying Knock

For me,

1. I do my own tune

2. I drive smart, not wreckless towards my engine

3. I don't bounce off the rev limiter.

4. I use Parts that are tested and known.

5. I can see knock BEFORE it happens.

Hey it might last a week more, or a year more, who knows. But I'll take care of her, like I have been for the last 3 years.

I am with you on this one mafia, but i think a car is like a computer, i never turn off my computer thats the only way for me to know how good my internals are same deal with the cars i beat the crap out of it thats the only way how i know which parts are durable and which are not hhhhhhh besides if my engine blows up i kinda think its a chance for you to do something CRAZY to it lol.

[discopotato03]

Hi Cubes , I assume your still using hydraulic buckets and if thats the case they are probably heavier than the RB26's solids so maybe have more innertia for the valve springs to control ? May need higher than 26 rates to achieve the same control .

Just briefly back to those 71mm 56 trim GT35 compressor wheels , the HKS spec GT2835 Pros S is said to hit the wall at ~ 300Kw in 0.87 A/R turbine form on a std ex manifold RB25 . If you check you'll see it uses that GT35 71mm 56T compressor , just like the real GT3071R does .

I went back and had another look at the compressor maps for the above mentioned compressor and a non Garrett map of the 76.2mm GT37 52 trim one . The 71mm 56T hits the 50 lbs/min wall at ~ 3.2 PR or 2.2 bar positive (32 PSIG) . The 76.2mm 52T does about the same at a lower pressure - 2.3 PR or 1.3 bar or 19 PSIG . My map of the 71mm wheel shows the wheel speed as ~ 128000 rpm and the speed line is just about doing a vertical power dive . The 76.2mm map does not show speed line numbers .

Had to say with non genuine maps but if its accurate its easy to see that the larger wheel in a smaller trim is doing the same job at lower shaft/wheel speeds . My thinking is that they both use the same GT30 UHP turbine but the different compressors pump air differently for a given shaft speed . The smaller wheel is demanding lots of rpm and high pressure to move the air while the larger mid trim one is doing it at a less frantic pace . The critical difference overall is how we size the turbine housing to achieve the most usable result . IMO if you want to use the real GT3071R in single scroll form and need it to be responsive then its going have to have a small (for a GT30) turbine housing , then I'd reckon you could forget about 50 lbs air and 30 odd lbs of boost because the turbine inlet pressure would choke it for sure . Its the legacy of having large trim sizes on the compressor and turbine and very much shows how altering the large trim turbines performance with A/R size is not the same as having a smaller trim turbine .

My gut feeling is that the 52T GT3076R/GT3037 would be a better suited to pulling 300 odd Kw from an RB25 if you could get a suitably sized GT30 turbine housing . HKS do GT30 housings in 0.73 A/R but unfortunately they're T28 flanged so no good .

Out of time now more tomorrow , cheers A .

[Cubes]

Hey Disco,

Yes.. std hydraulic head. Regardless others haven't had as many issues as I have. I'm starting to fear the installed heights were fiddled with when it had the work done to the head.

Regarding the HKS2835ProS. Standard Rb25 down this way generally see's them top out around 280rwkw on pump with the .87 rear.

300rwkw can be obtained with cams and exhaust manifold. I haven't seen one crack 300rwkw with a std exhaust manifold/std cams but in saying that here in Adelaide there are some high and some low dyno's varying up to 20rwkw even though they are run in the same DD Dyno shoot out mode.

Here in Adelaide; one has pushed ~313rwkw quite some time ago but that was a full built motor with cams, headwork, exhaust manifold etc. So not just a simple bolt on. Prior to the non std mods so to speak he was only able to obtain 280rwkw. I've since only seen 1 other with a GT2835r .8 which also topped out on pump with the std gear around 280rwkw. They seem to make good power up until around 250-260rwkw on little boost then as you push more boost in they stop making power as easily and eventually start dropping boost. So I'd assume high scroll inlet pressure.

Both have had huge issues with dropping boost. Unsure if its the HKS turbine housing issue that possibly uses a flapper that is too large as a result pushing the flapper open.

[Dale FZ1]

The 71mm 56T compressor actually looks to hit the wall at about 47 pounds of airflow over a PR of 2.25 – 3.1. That it can do so is more an indicator of being able to match the compressor to a wider range of engine capacities/efficiencies rather than the indicator that you’d be attempting to run the RB25 at the higher PR – shaft speeds of 140000rpm being run for what purpose? There is no more efficient mass flow to be had, and you’ll only be running to the PR (ie boost level) required to hit the power target. So it’s pretty safe to ignore those outer sections of the map because they don’t apply to the RB25 application.

It’s unfortunate that there’s no official flow map to suit the 52T 76mm GT37 compressor, but the 56T GT37 shows all the necessary specs to get an idea of what is needed, and where the differences lie between it and the GT35 compressor.

* Effective flow limit ~ 53 pounds/min airflow over the PR range of 2.3 – 3.5.

* Difference of max flow capacity is 6 pounds/minute

* Lower shaft speeds to achieve equivalent mass flows

Those two factors point to the GT37 as being potentially more power productive, obviously dependent on turbine efficiency (amongst other things). So for me the issues to look at are

• Does the compressor unit supply enough mass flow at the required pressure ratio
  
• What efficiency ranges does the compressor pass through with the engine going through the speed ranges
  
• What is the speed trend and absolute values of the compressor shaft rpm
  
• What is the likely required turbine and wastegate performance required to track the compressor speeds
  

As I’ve got some approximate calculations of what it should take an internally stock RB25 to hit 270rwkW, I will use them to assess the value of a GT35 compressor if chasing that sort of power range. Work on a mass flow range of 18 – 47 pounds/min @ PR 2.3, to cover an engine speed range of 3250 – 6500. Then find the compressor maps, plot them and review against the criteria above.

GT35 works over a speed range of 108000 – 130000 with a broad efficiency range in the middle engine speed range. Over 6000rpm, the compressor efficiency is dropping fast + compressor speed is heading skywards for small increase in engine speed. 60% efficiency and rapidly dropping says the compressor has passed its best at this point.

GT37 works over a speed range of 100000 – 113000 with efficiency not as good in the bottom end. The trade off is a high level of efficiency in the middle engine speed range but bugger all drop as the engine speeds get higher. 72% efficiency at peak power points to less heating of the charge and less compromise of ignition mapping or fuelling to cope with knock tendencies.

Comparing the two, at that level the GT35 can’t do much more, and if you ask more it will need a heap more shaft speed, and deliver an overly hot charge. The GT37 can comfortably do more, with much better efficiency.

Just briefly back to those 71mm 56 trim GT35 compressor wheels , the HKS spec GT2835 Pros S is said to hit the wall at ~ 300Kw in 0.87 A/R turbine form on a std ex manifold RB25 . If you check you'll see it uses that GT35 71mm 56T compressor , just like the real GT3071R does .

I went back and had another look at the compressor maps for the above mentioned compressor and a non Garrett map of the 76.2mm GT37 52 trim one . The 71mm 56T hits the 50 lbs/min wall at ~ 3.2 PR or 2.2 bar positive (32 PSIG) . The 76.2mm 52T does about the same at a lower pressure - 2.3 PR or 1.3 bar or 19 PSIG . My map of the 71mm wheel shows the wheel speed as ~ 128000 rpm and the speed line is just about doing a vertical power dive . The 76.2mm map does not show speed line numbers .

My gut feeling is that the 52T GT3076R/GT3037 would be a better suited to pulling 300 odd Kw from an RB25 if you could get a suitably sized GT30 turbine housing . HKS do GT30 housings in 0.73 A/R but unfortunately they're T28 flanged so no good .

For regular use, I see the GT35 working well up to 6000rpm with turbine speeds 8-10% higher. And down lower in the ranges it may well put the lower moment of inertia to good use, with quicker compressor acceleration rates giving improve throttle response. But in the higher speed ranges, the turbine is being asked to do extra work, accelerating to speeds 18% higher but driving a compressor that is (by then) comparatively out of its efficient flow range. The only upside (maybe) is that the rapidly rising compressor speed requirements probably won’t see boost creeping skywards as the turbine speeds rise – it really can’t push much more so the mass flow demands on the turbine + wastegate won’t be overwhelming.

So on-thread, I agree that from a technical angle that the 3071 won’t/can’t perform to the level experienced by Mafia with his 3076 56T. The experiences of a few Adelaidians supports that idea, as Cubes outlined. I also think the idea of a 3076 52T cartridge is worth a look. But lack of product availability and good technical flow data on the compressor means it’s a bit pie in the sky. There must be a good reason why HKS don’t offer a Skyline spec kit with the 52T cartridge.

Regarding the HKS2835ProS. Standard Rb25 down this way generally see's them top out around 280rwkw on pump with the .87 rear.

300rwkw can be obtained with cams and exhaust manifold. I haven't seen one crack 300rwkw with a std exhaust manifold/std cams but in saying that here in Adelaide there are some high and some low dyno's varying up to 20rwkw even though they are run in the same DD Dyno shoot out mode.

They seem to make good power up until around 250-260rwkw on little boost then as you push more boost in they stop making power as easily and eventually start dropping boost. So I'd assume high scroll inlet pressure.

For the money spent - and as a true bolt on – both Mafia and Lithium probably have hit a sweet spot and have the right spec to comfortably hit the 300kW mark. The larger turbine A/R that Lithium chose might well give better control of any likely boost creep and back pressure related detonation problems, but give a different sort of driveability. WI obviously clears the issue of detonation for Mafia. Still worth noting that they are achieving some pretty huge outputs from stock engines so let's hope longevity isn't too badly compromised.

Edited November 6, 2007 by Dale FZ1

[skylinecouple]

So the ultimate turbo working with stock ex. manifold at 6,500rpm max revs, no choking and therefore no boost problems would be..........?

[Dale FZ1]

No simple answer Gary. Every conventional setup has some drawback, whether due to design or budget, or application.

Mafia and Lithium were (as I understand) wanting their cars to do different things. Mafia actually initially ran a 0.82 housing on his turbo but found the trade off between extra power and response not to his liking. Then he went down a size and has something he's happy with. Lithium conversely seems quite happy with his, though he is generating a lesser power level than Mafia.

My views have a couple of facets:

GT3071 will run to around 270rwkW, doable and repeatable - subject to correct cartridge spec and tuning. It's a bit more efficient down lower in the engine speed range, and comparatively runs out of puff up top - at that power level. Technically speaking, it should offer quicker compressor acceleration due to the lower moment of inertia. Don't know that it would actually result in quicker boost response (compared to a 3076) because it needs more compressor speed right throughout the range in order to push equivalent air mass. All of which makes it a good candidate (my view) for a fast road car. Put a 0.63 IW housing on it to make it respond, because the 0.82 would probably dull the very thing that makes the -5023 cartridge needs to make it work - speed and acceleration.

GT3076 will run beyond 300rwkW and still be able to cut it effectively down around 260-270. For road use (my view), run the smaller A/R. Depending on preference, it would liven things up a bit in the response department and the difference in feel won't be huge. If your primary interest was track work then slip the larger A/R turbine on to free up the breathing if you're running more consistently higher rpm and load ranges.

Part of this discussion belongs in that other GT3071 thread, as Mafia's was really focused on a couple of specific issues. But hopefully my view point (which is just my view BTW...) is reasonably clear. I don't believe there is a single best. There are some damn good options that can suit the particular application and driver's preference. And some people like the pissing contest with mine being bigger than yours

Edited November 6, 2007 by Dale FZ1