S13 SR20DET (red top) problems.

Discussions concerning the M7790 cpu

Moderator: Matt

Post Reply
HermaN
 

Posts: 91
Joined: Sat May 22, 2010 10:40 pm
Location: NE Scotland
Contact:

S13 SR20DET (red top) problems.

Post by HermaN »

Got a lad who asked me to map his car for him. It's an SR20DET red top with a 50F00 ECU and NIStune type 3 board. Today I went to his, and we fitted his Z32 MAF & Tomei 555cc (yellow) injectors.

Once done, put the ignition to on, but left the engine off. Resized the injectors from the stock 370cc to the new 555cc ones and changed the injector latency from the stock 580us to 710us for the new injectors, then changed the MAF from stock to the Z32 in NIStune, this changed the k-constant from the stock 21475 to 24682, but when we started the car is just kept hunting then dying.

Only increasing the k-constant up to around 34000 kept it going and prevented stalling. However, at the end of the day, dropping it back to 24682 seemed to work just fine, so have left it as that for now (will fine tune it on Sat). Then when cruising with closed loop enabled in the "Feedback" options the AFR's fluctuate around from as low as 13.0:1 up to 17.0:1 instead of hovering around 14.7:1. It was mainly around the 16.0:1 mark on cruise, but as said, kept jumping.

Then to top it off, it was way too lean on full throttle, spooling at around 14:1 and 12.8:1 on WOT, and it just would not richen up.

After spending 5 hours we called it a day. Gonna look at it again on Sat. Anyone got any ideas?

Cheers!
klattr1
 

Posts: 22
Joined: Tue Sep 22, 2009 3:21 am

Re: S13 SR20DET (red top) problems.

Post by klattr1 »

did you try setting the K value at partial throttle/low load in 3rd or 4th gear to achieve stoich 14.7:1 afrs but without 02 feedback enabled? that should be your starting point for getting the K value right...also, make sure to set TTPmin to 0 and TTPMax to the highest values in all cells temporarily to make sure that's not limiting fueling. I'm also assuming that in WOT it's not using that last column in the load cells where it's going lean correct? lastly, once you get everything else set and if after turning on 02 feedback it seems to be varying too much, then it might be time to replace the 02...maybe try replacing the plugs at the same time too.
tmh983
 

Posts: 22
Joined: Wed Apr 21, 2010 12:29 pm

Re: S13 SR20DET (red top) problems.

Post by tmh983 »

i have the nismo 555 injectors, I found a table that showed latency was .61ms at standard fuel pressure for these injectors. Mine runs perfectly, I only had to tweak the K value the tiniest bit to get perfect 14.7.
I would suggest going back the standard afm temporarily, get the injectors sorted out then put on the z32. I think its better to do one modfication at a time, cos if it all doesn't work properly after doing several mods at once, then you have no idea which of the mods is causing the trouble.
HermaN
 

Posts: 91
Joined: Sat May 22, 2010 10:40 pm
Location: NE Scotland
Contact:

Re: S13 SR20DET (red top) problems.

Post by HermaN »

klattr1 wrote:1. did you try setting the K value at partial throttle/low load in 3rd or 4th gear to achieve stoich 14.7:1 afrs but without 02 feedback enabled? that should be your starting point for getting the K value right...

2. also, make sure to set TTPmin to 0 and TTPMax to the highest values in all cells temporarily to make sure that's not limiting fueling.

3. I'm also assuming that in WOT it's not using that last column in the load cells where it's going lean correct?

4. lastly, once you get everything else set and if after turning on 02 feedback it seems to be varying too much, then it might be time to replace the 02...maybe try replacing the plugs at the same time too.
1. Yes, did try that, but just kept on jumping around, would not sit at a steady AFR reading, so virtually impossible ot adjust.

2. Did not try that, but did adjust all flagged (closed loop cells) to 0 in filtered view while adjusting the k-constant if that's what you mean?

3. Nope, it's using the 3rd & 2nd last columns, occasionally touching the last column

4. I thought the O2 sensor was dodge too, because even with O2 feedback enabled and cruising the AFR's still jump around.

tmh983 wrote:1. i have the nismo 555 injectors, I found a table that showed latency was .61ms at standard fuel pressure for these injectors. Mine runs perfectly, I only had to tweak the K value the tiniest bit to get perfect 14.7.

2. I would suggest going back the standard afm temporarily, get the injectors sorted out then put on the z32. I think its better to do one modfication at a time, cos if it all doesn't work properly after doing several mods at once, then you have no idea which of the mods is causing the trouble.
1. I found a PDF on these forums that tell you the size & latency of all different injectors. The pink Nismo 555 injectors have a latency of 610us (0.61ms), but the yellow Nismo & Tomei injectors (ones fitted to this car) both have a latency of 710us (0.71ms), so that is fine.

He has an adjustable FPR, but I have been told the fuel pressure is set to stock.

2. We are going to try the stock MAF tomorrow and see what happens. If it's still not right I want to put the car back to how it was before and test it out. I do usually do a base run to test these things out before starting, but I totally forgot and realised after fitting the injectors & new MAF.

Car has following mods....
  • Tomei 555 injectors (yellow)
  • T28 Turbo
  • FMIC
  • Walbro fuel pump (now hardwired, will see if this make a difference, but I doubt it seeing is it jumps about on cruise)
  • Adjustable FPR (set for stock fuel pressure from what I'm told)
  • Z32 MAF
  • Tubular Manifold
  • Full straight through exhaust with de-cat
Only new additions are the injectors & MAF, the other items have been there the whole time and used to have a different board (odd/even chips on a single board) with a custom map on it.
HermaN
 

Posts: 91
Joined: Sat May 22, 2010 10:40 pm
Location: NE Scotland
Contact:

Re: S13 SR20DET (red top) problems.

Post by HermaN »

Well, turns out it was the MAF, swapped back to his stock one and the car responded as expected! Got the fueling all sorted today (12.5 on spool, dropping to 11.9/12.0 from 3500 RPM onwards), but laptop battery started to die, so have left the ignition advance for next weekend.

Cheers!
TM_S13
 

Posts: 246
Joined: Mon Dec 29, 2008 11:29 pm

Re: S13 SR20DET (red top) problems.

Post by TM_S13 »

sounds like there are a lot of dodgy mafs circuling arround...
anyway, good thing it's now working good.
and I think I know what car are tou talking about =P
tmh983
 

Posts: 22
Joined: Wed Apr 21, 2010 12:29 pm

Re: S13 SR20DET (red top) problems.

Post by tmh983 »

just getting back to the injector latency for a second, I did a little more googling, I found somewhere that said that all jecs (380, 480, 555 and possibly 740) injectors have the same latency of 0.58us, not sure if this is true or not. It also said that the tomei, blitz, nismo and a couple of others are all identical, they are all made by jecs.
Anyway for an experiment I tried the latency at 710us and at 580us, adjusting the k value to suit both times. The engine runs much better at the 580us setting, the afr's at idle are much more consistant, and the transient throttle is much smoother and the afr's don't sway nearly as much at very slight throttle openings. This is something I noticed with the old ecu I had, was that the closed loop program had heaps of trouble maintaining 14.7 at low throttle openings when the latency wasn't correct. I guess this is because the injection amount vs injector time becomes very non-linear at very low opening times if the latency isn't right.
Anyway, I just thought I might share my thoughts on this, i've left mine set to 580us.
PL
 

Posts: 957
Joined: Fri Feb 03, 2006 3:29 pm
Location: Adelaide, South Oz
Contact:

Re: S13 SR20DET (red top) problems.

Post by PL »

Dodgy Z32 MAFs are a big problem. I know tuners who will not tune a car that has a used MAF on it. Too much trouble.

I've now been bitten many times and I try to get owners to fit a new one if possible. They're not that expensive now. $AU300 for a genuine one.

I also keep a brand new one in my tuning box. At the first sign of trouble I slip the new one on and do a comparison run.

At the very least you should remove the mesh, clean the sensing element properly with a brush and solvent and then replace the mesh. Squirting through the mesh with contact cleaner is close to useless.

The other problem is that vibration fractures the solder joints inside them where the connector pins are soldered onto the internals. My mate refurbs them by cleaning the sensing elements properly and resoldering the joints.

PL
Post Reply