Dale FZ1

Click on this link and read to your heart's content. http://www.turbobygarrett.com/turbobygarre...ech_center.html

Congratulations on the good overall result. Similar to a few guys with the HKS 2835 Pro S, but I think they might be tuned/boosted a bit more aggressively. Be interested to see what it would look/go like if you let the tuner play further. I'd thought the fuelling was richer than that. A consistent 12.2 looks pretty tidy to me Depending on how quiet (restrictive) the muffling is, you may find the boost drop related to rising back pressure within the exhaust. My removable baffle 20kW, but is worth it to keep road legal and also avoid tinnitus. The odd thing is that the increased back pressure from it only stops it making more maximum power and does not induce a sharp nose over. Shape of the graph to me looks great - very linear looking and holds maximum across a decent range without falling away. It's good to know that you have extra grunt to access if you wanted to play with ignition. For the odd track day would it be better/desirable to use an octane boost eg. toluene and take advantage of it? That would take care of concerns about the dreaded knock and give a few more herbs. It's pretty evident that the ignition mapping is the only thing holding it back. A 3.5 inch dump/front pipe will fit in, but reasonably tight. 3 inch should be ample at the power level though. With your fabrication skills try doing the dump with a half-pipe attached similar to the HKS Pro S arrangement. I split open a 3 inch bend and shaped to suit. That would allow a bigger venting volume for the wastegate and perhaps reduced turbulence.

Internals are untouched. Around town speeds it responds progressively, delivering predictably and letting you know it's there with the promise of a lot more to come as revs rise further. It doesn't feel insipid, but between 2000-2500 there is noticeably less grunt than available with a 2871 / GTRS installed. However, this is about Gary's 3071 so let's hear from him a bit more.

I''d like to see Gary and Dave meet up to compare results. Boost control and general driveability comparisons are the two things I'd like to know about - but I remain satisfied with the 0.87 A/R HKS 3037/3076 That unit just gets stronger as the rpm rises (especically past 3000), but no massive torque surge to create difficult-to-control traction issues. To me that was one of the best aspects of that unit, and I'd think the 0.82 A/R 3071 might be similar in nature.

Mass-flow capacity through the turbine = easier (generally) power production as rpm, boost and power output rise

Have you searched for answers from this forum: http://forums.bimmerforums.com/forum/index.php Running the same machines, they would surely have sorted through generic type issues, and there are some guys with big output BMW engines. If electronic throttling is a problem, then someone there would have encountered and resolved it.

Looks to me like the range between 3000 - 5000 needs more time spent on it. Your sampling rate and scale will affect how it looks in graphical form, but reading the raw numbers in a table will still tell you it's jumping around too much. Expect to see some saw-tooth effect, but it's the magnitude of change that I'd be targeting, and get him to work at it. 12.0 to 12.2:1 at higher revs looks OK, and it's easy to see the map richened up beyond 6000. IMO you should tune it to run sweetly rather than be concerned about an apparent 20kW loss at this stage. You might find with the fueling sorted he can focus on ignition, which is where a lot of power is forfeited.

If high turbine inlet pressure aka back pressure is an issue, manifold design would be well worth revisiting. Looking at the overall spec and rpm being used, is this thing a dyno queen or drag monster? What sort of power target is there? You'd have to be aiming for fairly big numbers, so pipe diameter / length / merge designs would be critical in keeping TIP down and power production up. How did you decide on camshaft spec? While reduced overlap has helped in some respects, shouldn't cam lift + duration + valve size + port flow capacity be geared to match the turbo size (and vice versa)? As described, it seems that things just aren't working in harmony. Can issues with leaks in the throttle bodies be sorted with new O rings? Using the smoke method would be a good test (if available ) to check for leaks before pulling things apart. Can you remove / blank off the idle valve purely to remove it from the potential source of leaks? Might have to run it at a fast idle while testing, but at least you'd get an answer. Best of luck with things - sounds like a wild ride.

Part # 700382-5012, which will give you the 3076 cartridge fitted with a ported shroud 0.60A/R compressor cover. The ID tag will show a CHRA #700177-5007 Specify the GT30 IW turbine housing in 0.82A/R and ensure it's got the upsized wastegate. Run with a 0.63 if you want something that comes on hard. This housing will require a different dump pipe if changing from a stock type housing. Great unit but you need a good fuel system that can feed the beast. 280+ rwkW should be a doddle, if you need/can use that sort of power. Stock clutch won't last either. Search the forum for opinions and experiences with Walbro. Post up your results when mods are finished please.

Have you altered the boost level? Given they are likely much the same overall in terms of mass airflow capability, unless you've fundamentally changed something (ie the mass-flow), then the ECU mapping should cope easily with a change from high flow to high flow. That is running on the assumption your existing tune is not overly lean or aggressive on the ignition. That said, half an hour of dyno time is a cheap way of optimising for the slight spec change. Might also show you where relative strengths of one setup vs the other is.

The power figure appears in the right ballpark for the boost being run. But that number is an indicator only - how does the thing feel in its delivery? ie does it seem to take a long time to wind up, or does it suit your requirements? I agree that the 1.06 housing is likely a bit big, and rate the 0.82 as a better all round match with the 25 if you want traction friendly delivery and retain top end breathing capability. Yes, please do post up the outcome, and what it feels like to drive. I'd expect to see around 18psi by 3200rpm and a pretty aggressive delivery. Should give you a lot of entertainment.

The 0.68 would be "snappier" as it runs into boost, with an earlier/faster rate of acceleration of the rotating assembly. The 0.87 may result in too slow a rate of spool but certainly give the feel of making lots of power higher up. Bit of a juggling act to get sufficient bottom end response and torque from the smaller cubic capacity engine, but allow sufficient mass flow through the turbine housing to not choke things higher up as revs and power output rise. Dependent on what other mechanical modifications are made (manifolding, porting, cams), the 0.68 might prove to be the better match to give area under the curve. Focus on the torque curve.

It does look like a good start, though I largely share Nismoid's opinion at this stage. (Not so much regarding any comparison with twins though) I've been waiting for this build to come together so that we could see real evidence of what a proper split pulse system is capable of. Let's see the ignition fixed, and tuning continue

The trigger for my question was your comment about using a JJR dump pipe. It seems that there has been likely a 0.7 A/R OP6 Hitachi type turbine housing matched to the GT30 rotating group. A genuine GT30 turbine housing should offer a far superior flow/efficiency capacity and would have been my first recommendation. EW if the funds are available, IW if running tight. I think you'll find the restriction is within the housing itself, due to the internal dimensions and clearances available on the combination as fitted. Turbine inlet pressures likely climb out of proportion to boost, leading to rapid drop in cylinder scavenging and extra retained heat. There will be the source of your knock when trying to push either more boost or more timing into it. In essence, the turbine is having a tough time passing more mass flow than your 235kW without adversely affecting the operation of the engine. That said, I do know of some hybrid type GT3071 combinations making pretty good power. Simon-R32 comes to mind.

Just as a matter of interest, could you post up a picture of the turbo (as installed is ok) and/or the numbers off the ID tag found on the centre cartridge section. There are some Garrett units referred to as a GT3076R that are fundamentally different to the genuine article, and which may employ a variety of turbine housings (eg. reprofiled or high flowed OP6 or RB25 OEM castings). These can be restrictive in the exhaust, leading to lower than expected outputs. I'd be interested to confirm the hardware spec first before delving further into the capability of your tuner(s).

This creeping boost saga is a mind-bender, and I don’t know that there is a clear cut answer. The tendency to rise and fall regardless of whether running straight off the actuator or EBC gives a few clues, and yeah it is clearly a balancing act to get everything working together properly with this 0.63 A/R spec IW housing. Possibly a large diameter wastegate actuator can is part of the solution, as suggested? Given the actuator will have a limited range of travel, is there any evidence of flex/bend in the actuator mounting bracket? That might be a telltale of backpressure trying to force it further open. Observed boost levels are falling back to the targeted level as rpm (and mass-flow of gases) rise. That could be due to skyrocketing turbine inlet pressure; or alternatively massive venting through the wastegate causing turbine speed to fall away. If the backpressure rose out of proportion to boost pressure, then measured boost will drop because backpressure works against the boosted inlet charge. Monitoring turbine inlet pressure would tell virtually all you would need from a turbine efficiency / flow / wastegate capacity perspective. The HKS Pro S housing actually has a threaded hole right below the turbine inlet flange - just the ticket for screwing in a pressure fitting. Tracking turbine speeds is achieveable with the kit now offered by Garrett, but IMO it’s a bit of overkill for a one-off use unless there are the $$ available. Still, it is great that the gear is available to data acquisition nutters at reasonable prices. Getting to a solution without spending big $$ or investing/wasting lots of time, first thing I would do is look for bend/flex in that actuator bracket. Secondly, check for any evidence during dyno running, or your road testing, of knock? (presuming you've got a PFC and Datalogit access). A cluey tuner will have a good idea whether backpressure + scavenging issues are inhibiting the tune and inducing knock. You could go on searching for the cause/s, but the key is whether it is boosting predictably and not just running away when you really want 100% reliability at the track. Apart from the fact that it is overboosting then dropping back, does it run the way you want it to? One thing is clear: sub 4000rpm the 3071 punches hard and really fills out total area under the curve. After all that, if/when you have car down-time, the drilling/tapping operation into manifold or housing is quick and relatively easy. Use an NPT fitting and length of coiled bundy tube to dissipate heat, and a temporary pressure gauge will tell all.

This wastegate upgrade should put the venting capacity somewhere near the HKS Pro S performance level; maybe with a higher level of discharge turbulence due to the casting design. But it will be very interesting to look at your graph with a boost overlay against torque. I'm running a turbotech with 1 bar actuator on the 3037, max boost bumped to 17.5psi with street legal sound levels. The boost curve does not have a sharp knee, in spite of the turbotech. It just climbs and hits near its peak, with a 0.5psi rise/fall between 4-6500rpm. Hopefully this is the sort of creep you're experiencing? Bring on the results of the tuning

www.coolingmist.com There is a different standard of rating octane in the US compared to Australia. MON, RON, and RON+MON/2 all give different numbers for essentially the same level of knock resistance. If they are citing a number of 93, it's probably on par with our 98.

I’ve gone through the exercise of running a 2871 hybrid and then moved up to the 3037 Pro S. If you want early torque, and a reasonable spread thereof, the GT-RS will do it in spades. I wouldn’t think it likely that one would reliably (emphasis there) screw much more than 250kW with the GT-RS from an internally standard RB25, though there is evidence that people are getting more. Simply running more boost and tuning it to suit is not the answer if you want 280+kW. The amount of area under the curve for those running well tuned GT-RS, and the number of satisfied owners tells the story – they are a good unit that makes for a quick road car. Ben should feel happy with what his car is delivering. I’m left scratching my head on a couple of assertions though, given what has been raised over an extended time about what opinions were regarding the optimal unit to actually spend money on. Yes, it means taking the punt that everything adds up, and being confident that you can expect the car to produce its power in a specifically defined manner – and to produce within a reasonably well defined maximum number. For me it’s about the faith in your own research and that you’re not emptying the bank account to try for an unknown result. Why push the GT3071 as offering the holy grail of RB engine performance – strong power from the bottom end without choking high rpm flow, super fast response, and power potential virtually equal that of a 3076, and then step back from that choice? It may well be the case that the 3071 is the duck’s guts, and I’d just like for another punter other than BHDave and skylinecouple to back themselves and demonstrate what this unit is all about. Fuel consumption? I’ve kept meticulous fuel records of my own car over 35000+km. Every litre put in the tank has been recorded, and I have actually slightly improved the rate of consumption when running with a stock turbo @ 170kW, to the 3037 @ 270kW. It’s definitely not being driven any slower, though the fuel costs are definitely a real issue if the weekly commuting trip is anything substantial. My arithmetic average is 12.0l/100km and that is not bad by any measure when the performance on tap is considered. FWIW, it does about 450km/week. The ECR33 is not a race car – fact. That’s what a GTR is all about. Chassis response is not as quick as a lighter weight / shorter wheelbase S13, so not as well suited to many tests of agility. And it doesn’t have the sheer mechanical grip of an Evo, so not a competitor there. But the R33 is a great mid sized Grand Tourer. A damn quick, stable, good handling all-rounder. It will do most things very well and need not cost exorbitant sums to up-spec into something far more potent than the factory provided. It seems that Garrett have now pretty much got the wastegate capacity sorted with the GT30 housing, and even if it looks a tad agricultural compared to the Pro S, that 0.82 housing seems to work very well indeed. It would be great to be shown the way a streetable RB25 should be uprated by the intelligent selection of parts. Come on…. please consider giving it a run

The odd thing is that the only recent (relatively speaking) SAU experience with a GT2835 + 0.87 Pro S was from Grepin (where is he these days?). He made the horsepower, but evidently due to slower acceleration rate of the turbine suffered a loss of low-mid range grunt. Result: go back to the 0.68 housing. But that was with a unit that runs the smaller 56.5mm turbine rotor. Not directly comparable to the 60mm unit. If there is evidence of the “corking” effect of a smaller turbine, it is in Glen’s comment that he can’t run more full load timing than 18 degrees on the map, vs 19 on mine. Remember, I have stock cams, and no WI. The only airflow development comes via external bolt ons, and knock is controlled only via fuel and ignition strategies. Smaller turbine provided extra psi on the gauge, but cams only allowed better cylinder fill and scavenging until the point that the turbine inlet pressure starts to rise. Therein brings the limit of ignition timing. Compressor wise, it’s been bandied about that a well set up (read: housing specs, pipe work, tuning, and perhaps porting + cams) 2835 or 3071 will produce the same sort of max power level as the 3076 but with a more urgent progression into boost and associated bottom end grunt. From the comments by myself and Glen it’s clear that if tested from 2000rpm, trailing throttle and then 100% throttle, there will be a slight advantage to the 2835 that is soon enough knocked on the head as the rpm rises. Real world stuff along flowing switchbacks where there is a lot less than 100% throttle, there was no apparent advantage. In fact for some reason it felt as though the 3076 was producing slightly more useable torque – ignore the boost gauge, just what the chassis was doing and how the speedo needle moved. But refer back to my experiential view that neither is stand out superior @ 270kW. For the generally uneconomic process in acquiring a GT30 Pro S housing, it would be difficult to prove or disprove whether the 3071 cartridge would outperform the 3076 if both used the 0.87 housing and 60mm rotor. However I would be happy to try one out if offered the chance. Meanwhile the best approximation we have is the Garrett IW efforts of BHDave and his 0.63 housing with upsized wastegate, and skylinecouple with his 0.82 housing which also has the larger wastegate. Dave has had issues with getting good boost control, while Gary loved his but wanted to upgrade to the 3076 cartridge. The compressor maps tell the story, and basically the GT35 (71mm) requires about 10 -12000rpm extra to provide the same mass flow as the GT37 (76mm). So for overall performance, it’s all about shaft speed, gas speed, and rising turbine inlet pressures. I have proof in the metal (and mapping) that larger freer breathing turbines allow quite a bit more aggressive ignition timing and result in far better than just “adequate” low – mid rpm performance. My experience suggests that provided boost can be controlled properly, none of those four units (five if you include the HKS 2835 0.68) will be laggy or lacking progressive delivery if tuned properly. But I’ll still back it in that the smaller compressor would be better served by the smaller turbine housing. Do I take it that we will be seeing / hearing results of the Discopotato03 R33 running with a 3071 with a 0.82 Garrett housing, in the absence of an available HKS Pro S unit? Getting that comp cover surge slotted is not difficult or expensive. We have had discussions going on for 18+ months about what you would like to fit, so when do we get the satisfaction of seeing just what can be achieved from this unit? I’m prepared to help out where possible, and have a dump/front pipe assembly that will sort out the exhaust plumbing at a modest cost. Then you’d only have to be concerned with fabricating the inlet trunking as the major impediment to a bolt in fitment. Keep us posted.

No, not necessarily. You set a base timing figure, arbitrarily 15 degrees. The amount you see in the electronic map is added to the base figure, to give TOTAL advance. It is total advance that counts. The reason for checking your base timing is so you don't have to make wholesale changes to the IGN map to compensate for whatever amount it might otherwise vary from the arbitrary 15 degrees.

The 3037 runs 17.3psi max with baffled exhaust, and the 2835 runs somewhere just over 18psi. Full load air/fuel above 4000 is leaner for the 3037, at ~ 12.0 vs ~11.7:1. Ignition mapping was compared and is broadly very similar. Total full load timing for each is around 34 degrees. Different tuners, but with FCEdit we can easily compare all aspects of the tunes and log to compare the results of what happens on road. There is more to be had from either setup, but the approach to tuning is quite similar.

It seems that the 2835 + 0.68A/R and 3037 + 0.87A/R are the IW housing specs most often cited as chosen for fitment to RB25 for various reasons. BTW, the HKS 2835 Pro S uses the 56.5mm rotor, so suggesting the turbine combinations would be evened up by running the same housing A/R is a null argument. The power outputs in the noise-legal specs of these cars is not overly different, and the dyno sheets show likewise. Glen is running Pon Cams and has less noise output. The mechanical specs are not identical, but an example of similar achievements via different routes. We were both of the opinion that it was an even comparison. He has a 0.87 housing spare, but that was not fitted. If it was, and his top end breathing was opened up further then I possibly could remove my tailpipe baffle and release 20kW to keep it an even test. Swings and roundabouts. The point was, there can be similar overall results achieved by both units. But the on-road difference is noticeable. Neither will frighten you with savage wheelspin and a vertical wall of torque gain as they come on boost. Both are extremely driveable, and support progressive delivery of a large amount of power from an RB25.

Tick the box that activates knock warning.

FWIW, I have driven a friend's 2835 0.68 Pro S equipped RB25 (265kW) vs my 3037 0.87 Pro S (270kW) back to back and side by side. There is a discernable difference in delivery, but the mid range torque of the 3037 does not "hit", nor is it "harsh". It is just different, and both our opinions were that it felt better than the 2835. The 2835 was slightly (and very slightly at that) more torque productive @ 2500rpm when loaded in a high gear on a steep uphill pinch, but that was about it. Nice turbo, but felt much tamer by comparison. You could see the boost needle moving faster, earlier, but the car did not seem to respond quite the same. I think it comes back to the concept of "boost being a measure of resistance to airflow".