rob82

Well in that application it is modelled as 2D linear algorithm lookup table but it does become 3D once cam phasing is used. And as for MAP based systems - they are getting very complicated these days and manufactures are now using artificial neural networks to model cylinder fill. And at then end of all that most systems(Bosch) still use a MAF sensor to determine cylinder fill and only fall back to the ANN when the MAF fails. In my mind there is nothing that beats a MAF in terms of metering of air into and engine. The more complicated engine's get (variable cams, variable length manifolds, direct injection etc etc) the better off you are going to be in using a MAF sensor.

I struggled on an EXOX with the 2000cc getting back to closed loop idle using std ECU even on E85. Ended up pulling them out and going back to the 1000cc injectors.

Tuning resolution has very little to do with the accuracy of a tune. Dont ever let tuning resolution sway your judgement on what ecu to use. The BA-FG V8 engine is speed density based and only uses 2 1x14 lookup tables to map the entire engine operation. And it still gets through euro3/4 emmissions compliance.

Does the dyno plot look normal - like the cams are relatively in the right spot? If so that compression pressure is pretty much spot on. My RB30/26 with CP 9:1 pistons was around 160psi +/-5psi - when I advanced the inlet 7degrees there was very little compression presure difference (maybe 5psi at most). This would suggest to me that your engine has the correct static compressure ratio of 9:1. So many things to check before pulling the engine apart. Start by checking the CAS like I said.

So where do you get the genuine bosch 2200cc injectors? And is there an injector in between that isn't the 1600cc? I personally havn't tryed the 1600cc but to look at the lowend unlinearity on the injector dynamics website they will be almost identicle to the 2000cc in terms of low pulsewidth drivability. Is there something around th 1400cc mark that has similar characteristics to the 1000cc or 2000cc?

Once visteon finish their direct acting solenoid valves the world will be a better place - ie just punch in the valve opening and closing time and how much you want it to open. Then the petrol engines may catch up to the diesel in terms of torque and fuel economy with the help of HCCI and direct injection.

How did you test the CAS? Lock the timing at 20degrees and take it to 6000rpm and check how much timing variation you get any more than +/- 2degrees and its time for a new CAS or its time for a crank trigger setup. BTW - nice setup. Looks like a well thought out combo.

I dont see why people dont test their fuel systems before tuning - its the simplest thing to check. Boost gauge and air compressor is all that is needed. Put whatever boost pressure your going to run onto the fuel reg - watch you boost gauge to see the pressure. Put the return line into a measurable container (arround 5L), then start and run car for 1 minute. If there is 6x600cc/min = 3.6L/min then you have enough fuel system. We need a sticky on this - as I know it pisses me off as a tuner when you get setup on the dyno only to find no fuel flow on the first dyno run.

The problem with percentage changes is that most MAF have a non linear transfer. ie - its expotential. Have you setup your fuel table to resemble a commanded AFR map?

Not exactly following what you are trying to say but do you realise that the majority of cars dont have a flat torque; hence if you biased your commanded equivalence ratio to cylinder pressure you would end up having a rich midrange with a leaner topend. Dont confuse cylinder pressure with boost pressure - in most applications they are not a simple function of one another. And adding addition timing past MBT will only add to higher chamber temps and EGT's.

By that logic you run more addtional fuelling at peak torque rather than at peak revs? Thats dumb - you want to increase the cooling with increaseing heat cycles (RPM) not cylinder pressure. Cylinder pressure has no bearing on any commanded equivalence ratio or Catalytic/exhuast temp model I've seen. And I've tuned all sorts from direct injection BMW's to port injected vovlo's that will run stioch throughtout the majority of peak rpm no matter the load. This topic is quite trivial as there are so many factors about the engine design that influence what additional fuelling is required at high load/rpm.

Ahh - the difference between theory and practise. 1. Your using quite a large injector how much power are you after? 2. Those injectors are quite poor in terms of spray pattern and low end drivability. 3. The test benches that injector manufactures use don't always match the how the factory ecu's drive injectors so the provided latencies aren't always that good of a guide. The problem you are experiencing is non-linearity of the injector from driving a too smaller pulse width. You can correct it by increasing your latency time or to increase your ttp min. I would put in your calculated k value and try increasing your latency first, failing that increase your ttp min.

Was the AFR the same with either turbo?

Was the AFR the same with either turbo?

Its a good comparison in that most cars are probably only going to run around this level of boost but this result is exactly what I would expect - apart from earlier spool. If you had of reved it to 7500rpm you may have actually seen some results. I just dont know why you guys are so dissapointed - you only need to compare the compressor maps to see that there are f**k all gains unless your running around 2bar of boost with the gtx. PS what type of engine management?

Why dont you just organise a dyno day and log MAF volts(of a common MAF let say a VH45), rpm and boost to the dyno computer. That will tell you a whole lot more in terms of mass flow rates rather than just boost vs RPM. You can also see how different setups convert the mass flow into power.

Why dont you just get a Garrett EVO9 replacement twin scroll GT3071 or 3076 with internal gate. You will find that the 0.76AR TS GT3071 will spool about the same rate as the 0.63 GT3071 but the power will be more - ie closer to the 0.82. Twin scrolls are about broadening the torque curve. My pick for 280rwkw with the rb25 would be the TS GTX3071 with the 0.73A/R. I would expect that power at 15psi with full boost around the 3000-3200rpm range.

ACtually in reading the compressor map the -7's have no more to give over 20psi of boost. But it will also depend a little on setup.

The issue is still the reversion and its overfuelling becasue the MAF signal oscillates on the transition to closed throttle. You can reduce the fuel at this rpm/load point but then you will have a hole in the map at normal operation. They really need what the RB20det ECU's have which is a closed throttle fuel map vs rpm.

Max flow is around 22psi for both turbo's @ around 31 lh/min the difference is the -9's are slightly more efficient.

GT60

Yes - but I never put the cam covers back on.

I spaced the whole subframe down 12mm and cut two slots in the bonnent and it fit fine.

I beleive he was talk about back EMF off the solenoid creating voltage spikes on the driver. They will have most likely account for a fair amount of flyback votlage - I wouldn't be too concerned. Like I said just compare the impeadances of both solenoids.