Gt2871 Based Hi-flow Into Rb25

[Dale FZ1]

I acquired a Hi Flow upgrade based on 48T GT2871R, using stock RB25 housings. Selection was based on fairly honest appraisal of usage and needs:

daily driven to work + runabout. 70%

some sporting driving on-road 25%

some motorsport events. circuit sprints, motorkhana, touring road events. anything where risk of run-off damage is low. 5%.

Low rpm boost threshold; big-engine style response between 2000 – 4500rpm, also looking for strong power up to 6000rpm

Decided to fit it myself, sharing the experience with the SAU community.

The process of fitting is fairly straightforward, but time consuming given the relatively tight space constraints about the engine bay of an R33 Skyline. I would recommend the job be undertaken only if you have a reasonable mechanical knowledge, patience and methodical approach to work, access to a range of tools, ramps, stands (or a hoist) and a dry, lockable working area. Do not expect the job to be completed in a single day if you are a home-mechanic and want to double check each stage of your work. Two very full days should be set aside.

The Garrett-style high flow is NOT a straight bolt-up fitment and did take some extra effort to get it finished and running.

The oil feed required a nipple fitted to the turbocharger core, to take an elbow fitting for the pressure line. The more compact dimensions of the Garrett core compared with the Hitachi OE core effectively closed the clearance between the turbine housing and the oil fitting. The nipple was screwed in as firmly as clearances for spanners would allow, leaving me to hope my work was good enough and the tapered thread was going to seal effectively.

Tight clearances about the turbine housing gave a bit of grief when doing up one of the four manifold flange nuts. Not impossible, but best approached by placing the spanner on from between the engine block and manifold. Remember to reinstall the locking tabs and fold them up hard against the nuts.

The oil dump tube was not satisfactory, and required modification. The Garrett core is drilled/tapped for 6mm retaining bolts, but the supplied fitting was drilled undersize – should not have been an issue. The big one was that the angle and length of the tube did not lend itself to satisfactory hook-up with the return-to-sump nipple. The tube was sweated off the flange, a new length of tube with suitable bends and orientation was cut/bent and TIG welded in place. Good if you have access to friends in trade-work, but a problem if not.

The changes in overall dimensions meant the OEM heat shielding over the turbo no longer fitted satisfactorily. Mounting bolt holes did not align properly, and I did not believe the shield design lent itself to simple redrilling. On top of that, it needed a hole cut in it to allow the oil feed line to pass through. Assistance from MR Sheetmetal saw a new one fabricated and fitted.

Advice from Turbo Australia was to fit an oil supply hose with 4mm ID. Nissan engineers heavily restricted the volume of oil supply to the turbo by using a 1mm orifice in the banjo bolt and banjo fitting at the engine block. Failure to integrate a restriction would lead to problems with oil sealing in the turbo at the minimum, and possibly an issue with oil pressure in the engine. Given a parallel (ie. non-tapered) metric thread in the engine block, it was decided to integrate the OEM fitting into the feed line than to source a suitable fitting and restriction. This will be an area to monitor for vibration cracking, even though the length of steel piping used is quite short.

Water lines were a much lesser challenge, though the steel line running behind the block towards the inlet manifold was dealt with by cutting and attaching a fitting that the braided line could be screwed into. It also needed a little bending to get suitable clearance of the braided line from the turbine dump pipe. The other water line attached to the side of the engine block with banjo fittings.

Overall, special care was needed with the “set” of all fluid connections, and also to ensure there was no contact with other metal components that would result in hose failure due to abrasion. This needs an eye for detail, and an attitude of “fix it properly, once”.

The OEM metal gaskets are by far the best quality when compared with what Garrett can supply. Get new ones if you are worried about the integrity of the used ones, or else buy a can of copper spray gasket material. This will fill any small pores in the surface and give a better seal when reusing the originals. It is cost effective too.

Priming the turbo with oil prior to the first start-up is strongly advised. To do this, disconnect the igniter, and disable the fuel pump by removing the fuse (found in the boot compartment beside the battery). This ensures the engine will not fire when cranking, and nor will the exhaust system be loaded with raw fuel vapours that can damage the catalytic converter. I cranked the engine for 15 second intervals, 6 times and deemed that satisfactory. I have heard of another approach whereby the oil drain hose is left open while an assistant cranks the engine. Once oil runs from the drain hose, reinstall and clean up the mess before starting the engine.

Overall, would I do the job again? Yes. It was challenging first time around, but another would be comparatively much less difficult. Hope the pics give a bit of an idea to anyone else wanting to have a go.

I will give comment later about how this particular hi-flow runs on the road.

cheers

Dale

[Robo's]

Looks nice, considerable bigger than the stock one. Let us know on how it performs

[Dale FZ1]

Actually, it is only marginally larger than stock. Have more pics and commentary on selection of this particular unit.

Out of time, more tomorrow.

[wolverine]

go dale!! woohoo

[Dale FZ1]

These were my modification parameters, documented to keep focus on the whole objective:

Power capped through boost and engine rpm limits. Conservative fuel and spark strategies.

Partial throttle response and minimum lag is the means to maximising chassis balance + response, driveability, and vehicle speed over all conditions. Boost response, and overall engine response should be smooth and without a discernable torque “step” or cut-in point. Monster top end power to be rejected in favour of smooth, linear, lag free delivery.

Engine power increase to be achieved primarily through improved compressor and turbine efficiency without introducing undue restriction (inlet pressure) in the turbine.

Standard gearing is being retained, so the speed range of the engine cannot vary significantly from stock.

To remove all doubt about the best fitment to meet my requirements, a post-mortem was conducted of OEM turbos fitted to both RB20 and RB25, specifically to check the measurements and impeller trims used by Nissan. Using a set of vernier calipers, the outcome was somewhat surprising, and confirmed that the GT2871R was going to form the basis of my upgrade. Some basic calculations following Garrett’s literature, and plotting them on published maps also helped give a feel for what would work. This process was useful for estimation only, since the OEM housings were to be used with a Garrett cartridge. Note the wider cross-section on RB25 housing gives a larger A/R.

Comparison of the available cartridges showed the 48T version CHRA 446179-31 likely to best meet my objective. Plots @ 320, 340, 360 hp (@crank) show maximum compressor efficiency is being hit right in the mid-range, only running out of puff (ie hitting choke zone) @ 360hp. Plots run equally either side of the median line, suggesting high overall levels of efficiency and hopefully minimal drive requirements for low-lag.

Conclusion: Highest realistic output for this upgrade probably 350hp @ 15psi. Very comfortable @ 340-350hp. The comparatively high shaft speeds (vs the larger 52-56 trim sizes) for given outputs means this unit would be a little out of depth for a dedicated track car. Should be very nice match on a mild road-focused engine.

Don't know if everyone goes into the detail like this, but it helped me to understand what the upgrade was going to produce, and choose a unit that meets my target.

Bit more to come on initial driving impressions.

cheers

Dale

[Dale FZ1]

I was initially a little disappointed to find that Turbo Australia had fitted an RB20 wastegate actuator, so that I had 10psi of boost straight up. The disappointment was in not knowing just what sort of improvement or change (if any!) would come from changing the CHRA (rotating assembly) within the OE housings. That did not last long, as the difference at stock boost was only going to be academic.

Noise has changed a little, much more obvious induction sucking and a bit more of the F1-111 jet engine whistle under load,. The stock blow-off valve may be having a harder time coping with the extra pressure; at full throttle I get a totally different noise on lifting off, or changing gears. The noise I can identify is best described as a quick, single chirp.

There is little discernable difference at the rpm when it starts to get boost (ie the gauge will start to rise @ 1600rpm in 3rd or 4th gear in roll-on acceleration). The full 10psi arrives by 2500 if the engine is loaded up, and by 2800 in moderate roll-on driving.

There is no obvious power "step" when the turbo starts to work. ie. no firecracker effect. It feels much like a stocker, but just punching quite a bit harder. Running through a couple of my favourite bends it's just smooth response, hard acceleration, and linear power build. Oversteer on exiting tight corners is easy - dial about 2500rpm and squeeze the throttle in 2nd. Running off large roundabouts at the same low-middle rpm in 3rd or 4th gear is satisfying as it just drives smooth and hard – the feeling is like it has gained a litre of engine capacity. Higher in the rpm range it builds linearly and probably still plateaus around 62-6300. I say “probably” because it is now building speed quick enough beyond 5000rpm that you are watching the road more than analysing power delivery. This will change as I get used to the extra amount of stick it has.

On a highway run, the feeling is that the turbo makes the car respond to load on hills by accelerating – but staying under the boost threshold. It is an odd sensation to feather the throttle to avoid picking up 10km/h going up a hill. This is different response from standard, which I attribute to the comparatively aggressive upgrade from stock in turbine spec.

Overall, the match is meeting my expectations. The existing tune is very rich, and logging a few runs is not showing any knock issues. I'm satisfied there is no risk of engine damage running the current maps.

Can't wait for my dyno session next week to get the final tune and boost setting sorted.

[discopotato03]

Hi Dale , the results should be interesting and there is some room to move with obtainable bits if it does not meet with expectation . Say for example the HKS exhaust housings do turn up occasionally and there is a Garrett comp cover in .60 AR ratio that has the three bolt outlet to suit some Nissan apps .

If you want to follow it right through you could measure exhaust manifold pressure and compare it to inlet manifold pressure under load . Also exhaust gas temperature is good to know as well , temp laser gun aimed at dump pipe or housing usually tells the story .

Cheers A .

[Dale FZ1]

Looking at it overall, I could/should have considered the .60 A/R comp cover. Not a lot more stuffing around, but I would have had a bit of pipe work to have fabricated.

HKS turbine housing would be a handy pick-up but I haven't seen any recently. I feel this cartridge shouldn't upset the flow dynamics of the stock housings too much, due to the incremental impeller dimension changes. The HKS/Garrett housings meant for the cartridge should probably add to overall efficiency, but this setup seems to be working well. Would be great to do a back-to-back, but this upgrade is about as far as finances allow.

I run a pyrometer, thermocouple located in the dump pipe. With the stock turbo under high load/rpm I could see temps @ 720 C max. Backing off from a run like that, temps quickly drop back. On cruise it typically ranges between 560 - 640.

I also run an inlet charge temp gauge, just prior to throttle body. Like to monitor what is happening there.

At the moment I haven't seen any significant change in temps but will see what transpires after a tuning session to lean on it a bit.

[Cubes]

Good work....

Very keen to see the results...

[Interloper]

Noise has changed a little, much more obvious induction sucking and a bit more of the F1-111 jet engine whistle under load,. The stock blow-off valve may be having a harder time coping with the extra pressure; at full throttle I get a totally different noise on lifting off, or changing gears. The noise I can identify is best described as a quick, single chirp.

the stock bov will be fine. I get exactly the same noise on sudden liftoff with my stock RB20 bov. It doesn't leak boost so its sweet

OMG cubes, thats some dodgy backyard work! ......now be honset and admit that its your other car

[Dale FZ1]

Ok, first dyno run was done with the stock RB20 spec wastegate running @ 9.5psi. There were some fuel map changes too, leaning it off between 4400-6400 slightly, as there had been a gross over-fuelling (10.5:1 over 6000). So to be fair, there is no direct comparison with the OE unit for full load airflow and power capabilities.

I will have to post up my scanned dyno sheet later today with an overlay of before/after, but I'm still happy with the first effort - 242rwhp, or 180rwkW in metric.

An overlay of the boost curves showed this cartridge had a boost transition point about 150 - 200rpm higher up the scale, but the "knee" as it ramps to full boost was the same shape. This probably points to a slightly heavier / larger diameter impeller assembly having a larger moment of intertia from the initial get-go, but also explains a similar feel in on-road response.

The dyno session was to analyse the effect of hardware changes, and establish a mapping development plan for the higher boost settings. I'm confident that the 300hp / 225kW target will be met when we make the boost/fuelling/ignition changes next time.

Until then, I've got a high flow with more power everywhere above 2500rpm and the same sort of transient boost response. The big teller is the extra torque which means I basically can use 500 less mid-range rpm to get the same sort of acceleration, or use the same rpm and accelerate harder. Very good in my books.

[discopotato03]

Congrats Dale , it sounds like your getting pretty much what you wanted . The fact that the GT cartridges wheels are just slightly larger than the originals means the machine work is reasonably easy .

My first encounter with a GT BB turbo was the DP Turbo and it was interesting that my engine had more torque everywhere (literally from idle) than the RB20 BB turbo it replaced (this was on a later series slightly higher comp FJ20ET) . That DP turbo had the ability to windmill fast enough to feed the engine more air than atmospheric could without inlet manifold pressure going positive at very low revs . Going round roundabouts in 3rd gear was a great novelty !

The turbine it uses (same as yours) is very high flowing for its size mainly because of its open blade nature and wide paddle shaped exducer tips . Its significant that this NS111 series of turbines uses less blades (9 from memory) that the UHP series in GT30/35/37 etc (they use 10 I think) . The Hitachis ceramic turbine was all about being very light and responsive even at the cost of restriction because of its small diametre and the housings small volute . EGT/emissions/catalyst light up was probably involved and may have been a factor Nissan couldn't solve any other way with available technology . At the time they were going into production Garrett was doing heaps of R/D into ball bearing centre sections and wheels mainly high speed open blade designs . As I mentioned somewhere the competition TR30R turbo used recently in WRC/Cart/Audi R8 Le Mans cars is supposed to use a 60mm or approx GT30 sized version of the NS111 turbine (drools) and I have a pic of a (supposedly) HKS/Garrett VATN turbo with what looks like the same turbine .

Off the track again , your power delivery being OEM like is a good sign so driving the thing should be a pleasure . Having realistic goals helps .

Cheers A .

[Dale FZ1]

At this boost level, it is such an easy thing to drive; due to a combination of the factors you've experienced too. That "windmill" effect is really easy to pick at highway speeds when you hit a hill. The gauge stays on vacuum, but the car actually accelerates like the previous cartridge would allow when there was a few psi pumping in.

Also, under the same conditions a slight increase in throttle will see boost coming in progressively whereas the OE cartridge would have pretty much the full wastegated amount, instantly. The way it works is quite deceptive, even in on/off throttle driving through some winding roads. With 1000km of driving now behind the upgrade, it's fair to say that the performance of this cartridge is probably due mainly to the turbine - it does seem much freer flowing, but able to punch out 9.5 psi @ 2600. This is just running off the actuator too - no controllers or bleed valve.

Here's the dyno results as promised, which obviously show what the thing does at full load/airflow, but don't convey well enough how it translates to flexible/easy power delivery.

Boost curve has a knee shaped similar to the OE unit, only it goes up another 2psi.

Mixtures have been kept rich at the very top end while we get a baseline to work with; still plenty to play with right through the mid range too.

Hopefully the completely flat torque between 3400 - 4200rpm can be brought up slightly, although that may be a function of the cams and VCT. It may actually work to make for easier driving and good traction.

Overlay of before/after is a useful tool that helped confirm the seat of the pants opinion. Circled areas shows a slight loss of power/torque.

This particular upgrade won't likely win any bragging rights for dyno numbers, but it certainly feels alright to me.

Bring on more boost and see how it goes!

cheers

[Cubes]

Watch that mid range skyrocket once you do bring on the boost.

[The Mafia]

What that mid range skyrocket once you do bring on the boost.

you read my mind. This is a great upgrade. At least Dale isn't a person "just chasing massive boost and power"

He's putting power where its needed the most.

[MearCat]

Dale,

Did you drill out the banjo bolts to allow more oil flow to th turbo. You said that Nissan heavily restricted the oil flow with a 1mm orifice but did you increase teh size of the orifice - even by 0.5mm??

[Dale FZ1]

No - left the orifice at Nissan-spec. It's a pity I never took photos of it, although my skills are only average and may not have captured the view well.

I figured the oil supply requirements would be no different for the Hitachi core vs the Garrett core, being basically similar designs.

I did actually find using the stock banjo + bolt not 100% to my liking due to vibrations working on the pipe, and have my Enzed mate looking for a suitable Earls type fitting. He has done hose-end mods for guys with Silvias who have gone with Garrett BB upgrades and retained full flow lines - evidently the volume + pressure quickly defeated seals and flooded the turbo with oil. Messy and very smoky, apparently.

Whether the orifice is 1mm or 1.6 or 2.0 I don't think is a big issue, but the point of note was that the OE unit survived quite well with the tiny orifice. I believe the Garrett high flow will do likewise.

cheers

[benl1981]

How much did this turbo cost? How come you didnt go the GCG hiflow - that apparently comes on boost pretty quick also,.,,or a HKS2530..

Just curious. I like the idea of mid range. I have around 170-180rwkw running stock turbo, 10psi, r34 SMIC but it would be nice to have a turbo that would be cranking out 200-220rwkw reliably - with good midrange (I dont believe the stocker is capable of this...maybe on some dynos it does). But 50rwkw above what I have would be PLENTY!

[Dale FZ1]

Garrett produce a large volume of literature showing how to select and match a turbo to an engine. I have an interest in the technical aspects of turbocharger operation and their characteristics, and also in engineering detail generally. Basically I was prepared to do the reading, research, discussion, and mathmatics involved in selecting the cartridge specification for myself, while taking the time to understand and project how it was likely to work, and why.

Owners with Garrett/HKS upgrades to 2530 and GT-RS commented about the abilities of those units, and I figured I wanted to hit a performance capability somewhere in the middle of those two while paying less than full whack for either. The cartridge I selected basically was specified in the middle, although I know using stock Nissan-spec housings will probably reduce the overall abilities of the unit than if HKS-spec housings were used.

Short answer: cost, and confidence in my abilities to select a unit that would do most everything a GCG unit or HKS unit would at a lower cost. I specifically included photos and measurements of the stock Nissan impellers to give everyone an idea of what the factory engineers chose. Yes, this unit is all about response and a flow-limited top end. Even before the tuning and boost level is finalised I reckon it will meet the needs I specified.

Budget less than 1600 for the whole job, but note that there is a degree of stuffing about.

cheers

Edited August 20, 2006 by Dale FZ1

[benl1981]

Nice work Dale.

Sounds like a good package.

If I was to upgrade I would think about this.

Are you sticking with the Rb25 AFM and standard injectors?

Have you got a PFC?

How much did the turbo sey you back if you don't mind me asking?

Thanks

Is this what you got?

http://www.horsepowerinabox.com/HPIAB2/prod414.htm

Edited August 22, 2006 by benl1981