RomRaider

Looking at the average per injector compensation for each table:

A8DH200X
A -1.06
B 0.82
C -0.71
D 1.03

A2UG000J
A -0.89
B 0.63
C -0.70
D 0.90

A2UI001C
A 1.78
B 0.94
C 1.75
D 1.15

EDIT: A quick check of the 2.5RS/2.5i shows that it uses the older offset tables and they are effectively disabled (all zero). The Tribeca doesn't use it at all.

08 STI per injector comp below.

Anyone got any theories on why individual injectors would need different tweaks at different rpm/load cells? That just seems like overkill to me, but I'm guessing they wouldn't have gone through all that trouble if there wasn't a good reason...

(All academic to me since I have an 05, but I'm curious.)

The only reason that makes sense to me would be the flow patterns in the intake manifold and fuel lines.

What are the corresponding cylinder locations for A,B,C, and D?

We don't know for sure. Maybe A-D is #1-#4.

I Have an 07 STI limited plus i drive 120 miles a day. Send me the ECU defs and what you want logged and ill play around with it all week.

Unstable AFRs?!

I thought that I might contribute something to this. I've been running into an issue with my low pulse width AFRs where at around 1k RPM, s*** hits the fan per say. In no way am I able to obtain stable AFRs and by extension stable IDLE at this point. I've seen this strange phenomena occur while trying to tune both the SD based hacked ROM and while tuning the MAF.

Things to consider are that my injectors (JECS/NISMO) 720s run substantially lower latencies than stock. What I surmised is that the compensation tables discussed here have a much greater impact on the AFRs as the compensation values make up a greater percentage of the total pulse width.

This is bad.

Here are my tables, note that around 1k RPM, the rapid change of additional pulse width. Ignore table values, I adjusted the calculations to play nicer with my coloring.


What did I do? I zeroed out ALL 4 compensation tables. I suppose a less drastic change would have been to reduce all values by the percentage change in latency.

Results, a more stable IDLE at this point. NOT perfect, but I have yet to retune the MAF scaling at this point. Something I'll attempt tomorrow night. My exhaust note was also different, but that might have been more because of very moist air.

I'll post further results as I get em.

Cool - I've been waiting for someone to play these. Thanks for being the guinea pig!

Mine are zeroed as well with better results in overall fueling on DW850's.

I just tried to work around the wall I hit at 1k RPM Though not as bad with these tables zeroed out its still there.

Nothing I do can fix this stumble I see at this point. Watching latency and pulse width at this transition point and it goes nuts. AFRs lean out randomly.

I checked out timing was sane at this point. Fueling points were on target. I don't think I can do anything more with the MAF scaling.

Merch/Tea, you hiding any other compensation tables at strange points like .24g/rev, 950-1.1k RPM or 1.42V MAF ?!

Is it possible for injectors to be bad at a single point in their operation?! I know this is off topic a tad, but insight would really be appreciated. The previous owner of these injectors and my flow reports reported nothing strange.

More playing with things. I have a perfect idle now. Just some MAF adjustments. I still have some weirdness at at 1k and other places. I think its just side effects of using injectors intended for larger engines. FWIW though, these injectors have done quite well all things considered for $300

Meh. I'm going to max these out this summer, so I might as well pick up some 850s and not expect as much driveability as stock injectors.

Back OT, for those hitting this thread, zeroing out this table can help with stable fueling for non stock injectors

There are many undetermined compensations to fueling.

FWIW, my Legacy GT drove exactly like stock after I got my Perrin 816s dialed in. The very first upgrade I did was those and a fuel pump (I happened upon a good deal on both, used), so I was driving for a while with no other changes to the motor, and no problems at all.

For that matter, they drove pretty well even before I got the scaling and latency set right. I have to wonder if the alleged driveability issues with large injectors have really been due to crappy injectors or bad tunes.

Found the issue.

Stupid UTEC
http://www.iwsti.com/forums/2-5-liter-l ... ution.html

Along the injector lines, in my search for the cause I installed new DW 740s They're the only larger than stock injectors that maintain a V spray pattern which matches the ports of our heads. Cut stockers have a conical pattern thats too wide and of course lacks the V pattern to make the fuel easily enter the cylinder. DW has an 850 side feed with new pintel welded on to narrow the conical spray pattern FWIW, better than cut stockers.

OT, thought I'd share what I learned. I talked on the phone with the Deatschwerks and Witchhunter owners. I also learned Gordon from Witchhunter claims to have previously done the injector work for DW.

FWIW, I've been driving around with these tables zeroed out too. I don't have definitive evidence as to why I like this change, however, I do. It just feels - smoother - especially on the highway when getting on the gas just a bit. FWIW, I too am running the JECS/Nismo 720s (mine came in at 730 from witchhunter). Also running a ported intake manifold, TGV deletes, APS fuel rails, Walbro, and an SX AFPR.

t