Im making coil plates next week for a k coil setup. Making them out of billet. Im making one for myself but a couple extra so the overall cost is cheaper.

 

I think (fingers crossed/knock on wood) I am getting a handle on this.

I went out for an extended drive today and the car felt much better. I made big adjustments to the rate of change tables.

After thinking about what you guys were saying and knowing that my injectors are 4 times larger than stock I thought maybe the stock rate of change numbers is the cause of the wild A/F swing. I can only guess the ECU is overcorrecting causing the wild A/F swings.

 

Upon reviewing the datalog I found the WB settings to be suspect. The PLX WB outputs a linear 0-5volts but I only see 3.83v (16.91AFR) in the datalog when you go into fuel deceleration. So you should see closer to 5v when you go into fuel cutoff with a corresponding 20AFR.

Your s300 is programmed for 10AFR at 0volt and 20AFR at 5 volts. Make sure you match the WB output to the s300 input and your engine will feel much better. Also to prevent WB voltage offset make sure the WB ground is grounded at the ECU ground. Find the right wire (often called Power ground) at the ECU plug and do not use a sensor ground.

 

Thanks again for all of your help!

The WB issue is due to the fact that the input I am using only sees voltage that high, I'm using D10. Input on B6 (Analog 2) gives bad values. I have fuel pressure going into D12. I guess I could run fuel pressure to one of the other Analogs now that I have V2 of S300.

Is there any reason not to run it to D14 (stock O2). I was thinking there was but I cannot remember.

The PLX controller is wired to ECU power/ground (right where the OEM harness plugs into the ECU too).

 

Your tune is set to use D14 and you are using D10?
I see 0volt from D12 in the datalog.

 

No it's not set to use D14. I don't know why it is showing that. I noticed it when I DL it. It is set correctly, D10.

To see the fuel pressure you would have to set up the D12 analog 1 (not the V2 analog inputs) with the correct voltage offset.

For some reason that will not save with the cal file.

I can supply you with the offset file and what you need to do, but it is prob more pain than it is worth. The fuel pressure is stable.

 

I am going to try moving the WB to D14. I searched and didn't find anything saying it won't work. If it doesn't I can just run it back the other way.

It has always bothered me that it won't show the full range.

 

I just went and checked the wiring. I got confused (this was after all years ago that I set it up). Wideband does go into D14. I was going to use D10 but then I switched to D14.

 

I knew you were using D14 since the O2 volts matches the AF curves.
Just fix the WB issue first.

 

I was messing around with the WB wiring a bit last night. D14 is capped at that voltage. D10/B6 gives bad data. That only leaves D12 which right now is set up for fuel pressure. I have to see if I can get fuel pressure into 1 of the 8 analog inputs.

I did a test running the WB into D12 last night and I was getting about the same readings. Voltage offset was similar too. I noticed Hondata says to run the WB controller ground to thermo housing. I installed my WB when I was still using PowerFC so it is grounded to ECU ground. I am going to move that tonight.

It seems the target lambda for the 2nd column keeps wanting to add fuel no matter how much I add. The only time during driving that I am in this area is on decel and for a quick moment on tip in at lower rpm. Should I ignore this or is this a sign I need to adjust something else? Maybe injector dead time or tip in?

 

Don't even consider making changes to the fuel map until you fix the WB issue first since your are not getting the right data from the WB. Changes to the fuel map may not be going in the right direction. From my experience, most WB devices analog output have less voltage offset when grounded to ECU ground.

In you tune you have -0.39v of WB voltage offset and ideally this should be closer to 0. This means if the digital side off the WB say 14AFR then you must see 14AFR displayed live in s300 ''display''.

 

I think I have it now. I put ground to thermo and offset close to zero. I also wired it into D12 and see values up to 20 in S300 display.

 

Now you can start tuning the fuel map...

 

Well I thought I was making some progress and then the stumbling returned and started getting worse. I tried swapping a known good TPS and that seemed to work for 2 days but now it's back.

I made zero changes from when I drove it yesterday morning (when it ran almost flawless). Last night I tried to take the car to work and it was just terrible. Part throttle AFR was 18-19 (again no changes from the morning) and virtually all of the cells in the estimated fuel change table in Hondata are 20% or higher.

I have posted the calibration and datalog in my Hondata thread. (http://www.hondata.com/forum/viewtopic.php?f=19&t=17333) Datalog shows the car going from very lean to running normal and then back to very lean in a span of a few minutes with no changes to conditions or calibration.

If anyone wants the cal/datalog but isn't registered on Hondata forum pm me. I realize I've posting this in 3 different places but I'm hoping someone will see it and have some ideas. I'm about at the end of my rope with this car.

 

Why don't you post the logs here?

 

I just tried, wouldn't let me me upload the files. I can try to host them somewhere. Have to wait until tomorrow tho.

Let me know if this will work:
Calibration: http://www.hondata.com/forum/download/file.php?id=6500

Datalog: http://www.hondata.com/forum/download/file.php?id=6499

 

Car is killing me. Took it for a drive earlier this afternoon (about an hour) and drove it to work tonight. Ran like a champ.

All I did this afternoon was check vac lines (mind you I didn't change anything, just looked) and disconnected/reconnected all the injector clips.

 

Same here.man so confused with the cpr non sense.

Mind you there are many other users of cpr with similar problems,

I am going to try new ecu dizzy and jumper harness.

 

I think that the trouble you are seeing is related to be either the injectors or the wiring between them and the ECU. I would check the resistance of each injector from the ECU plug to make sure they measure around 10ohms. If it all checks good, I would find another set of good known working injectors and test from there.

Did you ever tested with another ECU?

 

I did test another ECU, had the same issues.

I was starting to think it might be injector related. I'll check the resistance tomorrow.

I'll drive it to work tonight again and see if it acts up.

 

Also, when you do the resistance check, it is useful to wiggle the injector connector and the wires that could reveal an open or short circuit. Just take the time to do it right.

 

I tested wiring. 14 ohms for all 4. I made sure to wiggle connectors/wires. If this issue comes up again I will test another set of injectors.

Side note. I changed the dead times and injector size in Hondata based on what I found in several posts where Hondata themselves were saying both numbers are off at stock fuel pressures. One of the threads for reference: http://www.hondata.com/forum/viewtopic.php?f=12&t=14581

I did that and then tuned the idle in open loop. Had idle at 14.8-15 AFR. I switch to closed loop and idle AFR goes extremely rich, high 11. Car was warmed up at this point. This is counterintuitive to my understanding of closed loop. If open loop is tuned properly shouldn't closed loop only have to make small adjustments? I checked the various closed loop compensation values for the temp the car was at and all of the compensations had 0 for a value.

At first I thought maybe wideband sensor is suspect, but wideband reads fine in open loop. Wideband display is always within a tenth or 2 of what is displayed in Hondata (both closed and open loop). Any ideas?

 

In you calibration you are using Closed loop - wideband input (voltage target).
Change it to Closed loop - wideband input (lambda target), save shutdown and restart.

 

I'll try that. Thanks

 

Well I tried that and no difference. Still idles upper 11s with S trim pegged at -15 in closed loop. Open loop idles perfect.