Suddenly running very lean

K-Series Programmable ECU installation questions / support issues
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Suddenly running very lean

Post by sinistrs2k »

Hi all,

I just got my kpro installed 2 days ago. Had some minor issues initially, but you guys got me sorted out and all was well. I drove the car home from the shop, everything worked perfectly. All sensors reading exactly right, could barely tell it wasn't the stock ECU any more.

Went out for a quick errand, car ran correctly. On the way back, it suddenly wanted to go very lean. I have a wideband installed, but not connected to the kpro at all. Fortunately I was close to my house and was able to limp home. I was seeing 17-20 AFRs at idle and even under very light throttle it only dropped to 18.

I looked around the engine to make sure nothing came loose, can't see anything wrong. I have tried several times to clear the calibration and reload the base, but get same behavior. When first firing up it has a good AFR, then gradually gets leaner and leaner until it idles terribly and I shut it down. I attempted to disable closed loop once, didnt help, so went back to base map.

Some things to consider that could be affecting it:
I have a test pipe with the O2 bung backed out of the exhaust stream, so the secondary sensor isnt going to get a super accurate reading. I could see this causing issues, except it still ran lean on a completely fresh calibration and even with closed loop disabled. Also, it ran fine for a whole day before.

Also, the sensor readings in kmanager show the secondary sensor voltage stuck around 4.08-4.10. I'm not an expert by any means but that doesnt seem right to me. In browsing the earlier posts I saw where some other S2000s were having issues with the secondary sensor voltage, but it also seemed that the newer software was supposed to fix that.

I cant figure out anything that changed. It was running great, then just went lean. I didnt bother attaching the calibration, it is just the base one included. I can maybe datalog it, not sure what readings would be helpful.

Thanks,
Jason
KPro Serial# 5071
02 S2000
I/H/E
User avatar
Spunkster
Site Admin
Posts: 23878
Joined: Fri Sep 20, 2002 5:06 pm
Location: Hondata

Post by Spunkster »

You'll need to post a calibration and datalogs showing the problem.

http://www.hondata.com/forum/viewtopic.php?t=3629
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

formalities

Post by sinistrs2k »

calibration and datalog
You do not have the required permissions to view the files attached to this post.
KPro Serial# 5071
02 S2000
I/H/E
User avatar
Spunkster
Site Admin
Posts: 23878
Joined: Fri Sep 20, 2002 5:06 pm
Location: Hondata

Post by Spunkster »

That datalog really does not show anything. The car is not warmed up fully, there is no short or long term fuel trim, and there is no datalogged wideband reading.
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Post by sinistrs2k »

I didnt think it would have much of value. Like I said, the wideband is not connected to the kpro at all, so no way to log that. It wouldnt add that much more info anyway, it just shows it consistently getting leaner and leaner.

There arent any fuel trims because its a brand new upload, started from scratch. And it isnt warmed up because I didnt want to let it lug along at 18 AFR for 5 minutes to get there.

Yesterday, this exact same fresh upload ran perfectly fine. In the middle of the day it just changed all of a sudden and went lean. I just don't know what other things to look at that could be broken in such a way to cause this.

Thanks,
J
KPro Serial# 5071
02 S2000
I/H/E
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Post by sinistrs2k »

Just as an update, it appears that my primary O2 sensor might be failing. It might just be very coicidental timing that it failed the day after the kpro install.

Thanks,
J
KPro Serial# 5071
02 S2000
I/H/E
User avatar
Spunkster
Site Admin
Posts: 23878
Joined: Fri Sep 20, 2002 5:06 pm
Location: Hondata

Post by Spunkster »

THe stock S2k O2 sensor is a narrow band and does not give accurate readings to tune by.
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Post by sinistrs2k »

I understand that, I wasn't tuning with it.

What appears to have happened is that everything was fine initially, ran perfect the first day. Then the primary O2 sensor started to fail with a constantly low voltage. Once this happened, when it went into closed loop the ecu saw a constantly rich mixture and started pulling fuel. The short term trim was something like -27% after 30 seconds or so of idling.

To get things running again, I disabled closed loop and gradually corrected the fuel tables by hand until everything was back to normal. Fortunately I had a wideband installed so I could see what was actually going on. I havent replaced the primary O2 yet (expensive!), so it just throws a CEL after a minute or so (Primary O2 Low Voltage code). I'm just leaving it out of closed loop for now and trying to get the fuel table as close to 0% trim as possible. Hopefully once I replace the Primary O2 and can go back to using closed loop, it wont have to change things too much. I believe anything under 5% short term trim is supposed to be about as good as you can expect, right?

Thanks,
J
KPro Serial# 5071
02 S2000
I/H/E
Tsx536
Posts: 20
Joined: Fri Jan 02, 2009 12:40 am

Post by Tsx536 »

I would get that o2 sensor replaced sooner rather than later. How rich is your engine running right now? Not sure if you still have your CAT on, but if you run rich for long periods of time it's bad for you CAT. I learned the hard way when I once disconnected my O2 sensor to run open loop at all times.
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Post by sinistrs2k »

Actually I'm not running rich at all now, its running quite good. I'm probably half a point leaner than stock, getting some better power and mileage. I keep an eye on my wideband constantly while driving, so I know if something is too far one way or the other and make small corrections each day.

Also, havent had a CAT for many years, so nothing to worry about there.

Thanks,
J
KPro Serial# 5071
02 S2000
I/H/E
geoffrey70
Posts: 13
Joined: Tue Mar 03, 2009 7:18 pm

Post by geoffrey70 »

sinistrs2k wrote:Fortunately I had a wideband installed so I could see what was actually going on. I havent replaced the primary O2 yet (expensive!), so it just throws a CEL after a minute or so (Primary O2 Low Voltage code).
Just a quick question for you. Did your car have any symptoms when this started happening? My car has been acting funny since the second day after my tune and this has me wondering... I'm guessing it's my O2, but i have no wideband to confirm this.
Also, does your wideband have a narrowband output? You may be able to feed that into the car instead of replacing your stock sensor.
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Post by sinistrs2k »

I only had the kpro in for 1 day before this issue occurred. It was fine for that one day, then all of a sudden the primary sensor starting reading wrong and closed loop started pulling fuel to "fix" it. It literally was running perfectly one minute, stopped the car for 5 minutes, started it again and bam, instantly at 17:1 AFR at idle.

Yes, my wideband has a narrowband output, and if all else fails I may end up having to use that. I recently tried replacing the primary O2 sensor and it didnt help at all. The new one did the exact same thing.
I'm getting "low o2 sensor voltage" constantly. It sits at around .22-.24 volts most of the time. So to me, if the sensor is working correctly but the voltage is reading low, then either the voltage going in is too low or the ground is not at zero. Does that make sense? I'm not sure where the ground for this is exactly, but doing some quick checking at other points along the car I dont see any ground issues so far. So I need to figure out where the voltage input is coming from and check that I guess.

I'm really concerned that the voltage for the sensor is generated by something inside the ecu that may have gone bad. I just don't know the details of this circuit yet.
KPro Serial# 5071
02 S2000
I/H/E
geoffrey70
Posts: 13
Joined: Tue Mar 03, 2009 7:18 pm

Post by geoffrey70 »

sinistrs2k wrote:I only had the kpro in for 1 day before this issue occurred. It was fine for that one day, then all of a sudden the primary sensor starting reading wrong and closed loop started pulling fuel to "fix" it. It literally was running perfectly one minute, stopped the car for 5 minutes, started it again and bam, instantly at 17:1 AFR at idle.
Hmm, I should probably take mine back to the tuner and have the a/f checked and the problem worked out. My issue actually started about 3 hours after i picked up the car from the tuner(picked it up then went on a 70 mile trip). It started with my idle being extremely rough, then the car acting like it wanted to stall at redlights from the idle. Everytime i cold start the car it funs fine, until it starts reading short term fuel trim, then it sputters and doesn't want to go for about 5 seconds or so. It's wierd because sometimes it runs perfectly normal, and then others it runs like a POS car.. Haha
Are you on the stock cal, or are you tuned?
sinistrs2k
Posts: 27
Joined: Fri Dec 05, 2008 10:49 am

Post by sinistrs2k »

Ya, that actually sounds pretty similar to my issue, except mine never got better, it just stayed running terrible. This was all on stock KAL too.

I fixed it temporarily by locking it in open loop so it wouldnt apply any short term trims.

I just checked the honda manual for dealing with this trouble code and it says to check for a short in the O2 connector. Assuming the connector is good and there arent any shorts (which I don't expect there to be) then to replace the ECU. :(
KPro Serial# 5071
02 S2000
I/H/E
User avatar
Hondata
Site Admin
Posts: 10615
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Post by Hondata »

While the ECU uses the secondary o2 sensor input, the o2 sensor itself should be the primary sensor in the stock location. The secondary sensor should not be used in substitution for the primary sensor.

If you are having problems with closed loop, then the following will help:
- calibration
- datalog (in closed loop showing the problem)
- description of o2 sensor location and type
Hondata
Locked