GregBBRD

Keep in mind that none of the aftermarket EFI systems use a knock sensor, which makes the Bosch system "light years" better than anything else out there. Without the knock sensors and the "super soft" Porsche rod bearings, you can plan on a connecting rod "leaving" the engine on a fairly routine basis...unless one stays very, very conservative on ignition timing. Plan on blowing up an engine or two while finding how conservative the timing needs to be.....

gb

Tony

Looking at this what would you deduct is the max duty cycle seen?

The way i interpret an exmaple shown in the ST manual is that when the switch over occurs on the injector firing you divide the max value by 2???

this was a run through 1st 2nd and well into 3rd gear...its not scaled to time but rather RPM which is way the line appears to trace between 4000-6000 rpm

Tony

This is an image of percent MAF on the same run. I interpret this as a max of 105%.

Rick Carter

I'm posting to stay current on this thread.

hans14914

Greg,

While I agree, the ST is one of the singular best developments for the 928, there are plenty of aftermarket solutions with knock detection. Even the open source megasquirt and VEMS kits have knock detection now in packages costing less than $1000 installed. On that note, the way that porsche and ST deal with knock is much more sophisticated than most aftermarket systems.

BC

I do, and no one will be missing an opportunity like this to point at me and mention it I am sure. With all my down time though, I have had "time" to research and talk to alot of people about alot of things. With a complete, intact S4 with everything working, and no long trips planned on the very fragile LH computers, this is the perfect soltution.

BC

The only thing that the Bosch EFI system is light years ahead of is the carborator on my mother's first car after she married - a Ford Torino. Possibly also the computing power of Russia's first satelite - Sputnik.

Many, "newer" EFI aftermarket systems have dynamic knock detection and control, even the much maligned "open source" stuff like MS and VEMS can detect knock and deal with the resulting engine changes needed.

BC

But Tony wanted to stay out of the "this is that and that is this" territory with this thread. We can take it elsewhere to preserve his idea.

To add something useful, I think it was Louie Ott that mentioned Tuning the spark is the biggest gain in power once the fuel is in a safe region.

Vilhuer

IIRR ST only allows one basic value for AF ratio to which it aims for. Of course its possible to adjust each cell into whatever value one wants but current autotune do not understand more than one value IIRR. Knowing what I know about ST user interface I suspect it would be very easy for Niklas to change code to have full map of AF values to which to aim for. Meaning each cell in 16x16 map would have its own personal AF that it tries to reach. These could be changed same way as system allows changes to be amde into map cell. What I'm not sure of is if this is also possible to implement into John's hardware box but I suspect hardware side would not need to be changed at all.

One thing I have tried to have Niklas make is to have autotune which tunes each cell 3, 5, 7 or whatever use sets times in a row and then take average from all tunes or average from n results which were at center of the set for example. We tuned friends SC several times with autotune and it seemed individual cell value could be some distance off from previous result for same cell while both were firmly in green. If same cell were tuned multible times and average taken from results final number would be even more precisely correct than what current autotune gives I think. Difference between tuneups is so small that it probably would not make any real difference and would be just some more fuel wasted in tuning sessions. Still, I think it would be fun feature to have and from user interface programing point of view this is easy to implement. Its obviously possible to achieve same thing already now by doing many consecutive autotunes and then combining results in Excel sheet for example. This is what we did in friends case as we had 3 complete autotune maps from which I calculated average. Resulting map was then verified in fourth tuning session and practically all cells were spot on from the start.

heinrich

Clever thread Tony

kaptnknemo

I will pay to have someone run the ST on my early 86 and burn me some chips. I spoke with all4woody at SITM07 about this and he lives relatively close to me in AL, so he gets to make me the first offer.

KK

AO

Boy, I can't wait to give this stuff a try. We really need to get Z on this thread, he probably has more real world experience with the ST than just about anyone else. I know he mentioned a few tips and trick when we were at SITM that would be beneficial for everyone here... it's just I don't remember them.

As soon as I get my 42# injectors and smaller pulley... I will enlist the help of this thread!

**Officially in my favorites.**

bd0nalds0n

What would be the maximum injector size that the ST can accommodate before the injection system can't maintain a good idle based on stock S4 fuel pressure and the minimum duty cycle of the injector?

Right now I'm running 30# injectors and don't seem to have any leaning-out issues. If I ever planned to use a smaller pully, I don't know whether it would also make sense to bump up to 42lb injectors, or whether 30# injectors will suffice.

I'd prefer not to get into a multiple upgrade cycle with a ST chip for the 30# injectors, then realize I need 42#, then realize I need a new chip, etc.

BrianG

Erkka...... so what A/F target value did you use for the auto-tune mapping?

BrianG

IIRC my initial conversations with Andy, he suggested that 30# injectors would be sufficient to 10 lb boost........... but that was using the BEGI RRFPR.