Idle and A/F fluctuating CA18DET

[janko]

I having an issue with Idle and A/F fluctuation when the car warms up.
Setup is:
Z35 MAF in ID=74mm housing
DEKA FI114961 630cc @ 3bar (resistor removed)
INNOVATE LC1 (as A/F monitor and ECU feedback)
S14 recirc vale
running on RON 100 pump gas.

K value is set to 270, Inj Latency to 640us and volt latency to 120.

The spec for the DEKA says 409us@ 14V.

Under load the A/F is OK.
When the car idles its starts to go between rich lean and rpm starts to fluctuate. If I disable O2 feedback to the ECU car idles fine and A/F near 14.7:1.

Already checked for air leaks but cant find anything.
I'm running the CA without the crank driven FAN. Instead I use 2 electric FAN's. When they turn on the idle and A/F gets stable. So with just a little load it is fine.

Injection time reported by NISTUNE is above 2ms and my DEKA 4 should work fine down to 1,5ms.

[Matt]

Need a full log and tune file so I can check it out further. Screenshot doesnt give me enough info

[zbeffreyCA18]

Check your timing and your IACV. I just had a recent idle "hunting" situation and in my experience, those two things are usually to blame. Also, if it helps any, with my Z32 MAF and Deatschwerk 550cc injectors my K Constant is 383 and my Latency is 850 with 3bar (44psi) static fuel pressure.

Hope this helps a little!

[janko]

Thanks for the support!
Timing is OK. IACV should also be OK. The car was idling/running fine with a Z32 MAF stock injectors and K value of 580. Also passed emission testing(EU).
Now the car was standing for 7 months. First start with the DEKA's and Z32 the idle fluctuated very badly. Changing the MAF to Z35 helped a little but still fluctuates. The reason for changing the MAF was a reversion problem with the Z32 MAF when my recirc opened.

OK in the 7 months I've also changed the FPR, Turbocharger (brand new GTX3067R), exhaust header (4-1 self made) and the down pipe. All V Band. The LSU is now much closer to the turbo but should still be OK. But from just little to full load everything runs fine. Also found no exhaust leaks. Just the idle does not work.
Other thing I was thinking of was minimum on time of the Injectors. The DEKA's should run down to 1,5ms NISTUNE reports >2,2ms but will check it tomorrow with the scope.

Air/vacuum leak is still my favorite because I need to add fuel (increase latency) at idle to get 14,7:1. Latency should be set to 409us. But with 409 the car idles very lean with O2 feedback disabled.
The car idles OK some seconds and then it starts to lean out.I do not see any change in injection time, voltage, MAF, Fuel trim it just leans out. The ECU then tries to compensate for by changing FTS and then the fluctuation starts.

Does Min Load (TP) limit the minimum injection time. If yes how is the conversion?

Will make log of just Idling tomorrow and post it if i'm not able to sort it out.

[janko]

hi,

found a small leak on the recirc actuator. But still idles like crap. Log and map attached. Also measured the duty of the injectors 1 and 3. injection time is OK. But what are the short turn on times? Anyone seen something like this? Voltage is measured directly on the ECU pin.

[janko]

and with a different time base. the short on's appear erratic on all injectors. Dont belive that <200us can trigger the injector but looks strange to me. It always appears just before another injector opens.

Has anyone measured the overshoot voltage at turn off with standard injectors?

[janko]

reduced TP Min. Idles fine now.

[Matt]

Latency should be set to 409us. But with 409 the car idles very lean with O2 feedback disabled.

Compared against most latency figures 409uS is quite low. Adjusting TPmin can change injector time just at idle so does work as an effective workaround