RB20DET lean on cold start

[MidnightR32]

I'm running a Haltech Platinum Pro with the stock map for the RB20DET + tweaks for my build. When I started it this morning, the IAT was at 32F/0C (Outside temp was ~42F) and the AFRs went off the scale for my AEM gauge. I let it run for ~30 seconds then turned it off. I then did a smoke test, by hooking into a vacuum line directly behind the throttle body, and confirmed there was no leaks ( i let the smoke tester run for ~5 minutes / confirmed it was generating smoke before hand ). I then started the car again, after fixing a power steering leak.... and it seemed to be happy for ~30 seconds (was at ~11/12), but then seemingly shot back up to ~18 again. I then let the car run for a few minutes, thinking this was just cold start related, but the AFRs never came back down. I didn't let the car fully warm up, so maybe this makes sense, but I'm not sure what to do?

The AFRs, when I last ran it, were normal, but because of personal reasons I haven't been able to break in the car. It was rebuilt last year.

 Engine / Mod list:

    Supertec pistons

    Spool rods

    Kelford 270 cams and Springs

    Stock crank

    Deatschwerks 900cc injectors

    Walbro 255 (Wired directly to the battery relocated in the trunk)

    NKG Audi R8 Coil packs + Conversion Harnesses from Wiring Specialties

    Holset HY35 turbo

    Haltech Platinum Pro ECU

    Haltech boost controller (wired in for high boost

    RB25 Neo transmission + needed bits for the speedo to work

 

[MidnightR32]

I've uploaded logs / the current map to Gdrive if anyone has any ideas. https://drive.google.com/drive/folders/1FPnr22Z-EilG7tYdHSi0Qb07P2jfFLKV?usp=sharing

[GTSBoy]

Put a pressure gauge on the fuel rail.

[MidnightR32]

I have one and it was at ~43.5PSI.

[GTSBoy]

I shall assume that was primed, not running.

Now it is time to suspect the wideband sensor.

[Duncan]

I agree, most likely a sensor issue. You didn't mention how it was running but above about 20:1 it won't run anyway

[MidnightR32]

Unless my sensor just died, was only a couple months old, I strongly believe it didn't. How can I test it? I never mentioned it being 20:1+?

[MidnightR32]

Oh, my misunderstand.

When the car was running, it sounded ok, but if I gave it any gas it wanted to die but caught itself afterwards. It's very different from how it was a couple months ago when it was warmer outside. The logs show that the AFRs are better during, what I assume, is warmup enrichment. Because it's cold, and air is more dense, should I work on the enrichment bit?

[GTSBoy]

1 hour ago, MidnightR32 said:

Because it's cold, and air is more dense, should I work on the enrichment bit?

You shouldn't need to massively fatten up the mixtures for cold conditions. For one thing - 0°C is not that cold. For another, the Haltech will be using the IAT sensor to tell it how dense the air air, and calculate the correct amount of fuelling. Then the cold start enrichment is added as a % on top of that, so it should scale with the main fuelling.

You might also doubt the IAT sensor at this time. You're not using one from an RB26 are you? Using a nice Bosch sensor or similar?

2 hours ago, MidnightR32 said:

Unless my sensor just died, was only a couple months old,

Happens. Some wideband units take great pleasure in killing their sensors.

2 hours ago, MidnightR32 said:

How can I test it?

Put another wideband in the tailpipe and compare. Or just swap the sensor to a brand new one and see.

[MidnightR32]

Ah, fair enough. For the IAT, I'm using a legit GM sensor that was used on the car prior to my current build.

I'll get another wideband and IAT ordered and follow up when they show up. Thanks for the help.

[MidnightR32]

Update on the car... I took video of about 9 minutes of it running, shortly after start up, and the car sounds OK? I replaced the ATS and confirmed that the new/old one are showing warming than freezing teamptures. I will replace the wideband tomorrow, but maybe this video can give some insight.

Logs of warmup - https://drive.google.com/file/d/1bpvfmsLfZEjg_rBZw8IseAuM3iWq3BGN/view?usp=sharing

[MidnightR32]

It was the wideband. 🤦‍♂️

[joshuaho96]

23 minutes ago, MidnightR32 said:

It was the wideband. 🤦‍♂️

Yeah not a mystery here. Those AEM wideband controllers kill sensors incredibly quickly. They go full heat the moment they get power. If you don't put them on a switch to only turn on a few seconds after engine start the condensation cracks the sensor and you get results like what you described.

[MidnightR32]

On 12/8/2024 at 11:25 PM, joshuaho96 said:

Yeah not a mystery here. Those AEM wideband controllers kill sensors incredibly quickly. They go full heat the moment they get power. If you don't put them on a switch to only turn on a few seconds after engine start the condensation cracks the sensor and you get results like what you described.

how would I do that?

[joshuaho96]

22 minutes ago, MidnightR32 said:

how would I do that?

I don't know, honestly. Easier to just use a CAN wideband with the Haltech so you can get the ECU to control the wideband in the manner described.

[GTSBoy]

57 minutes ago, MidnightR32 said:

how would I do that?

Small analogue delay circuit with a relay. Just need a trigger, say, fuel pump running for more than 5s or so?

Or, you could go full mental and write a sketch for an Arduino.

[Dose Pipe Sutututu]

Just get a Haltech CAN wideband, no need to wire any delay circuits, or relay, etc.

Since I've moved to a Haltech CANbus wideband, I have not replaced a single sensor for over 5 years (if not longer, can't remember when moved to Haltach).

[Murray_Calavera]

Agreed, another vote for the Haltech CAN wideband. 

No voltage offset errors is a beautiful thing.