Hondata

Here is the setup:

ECU Serial# 24383
Smanager version 2.2.8.0
I believe it is a P28 ECU

1998 Integra GSR

Benson sleeved block
Benson balanced and micropolished crank
CP pistons and rings (9.25:1 compression, 83mm bore, 87mm stroke)
Pauter rods
ACL bearings
OEM (TypeR) oil pump
Blueprinted assembly
ATI Damper
ARP Head Studs
SuperTech SS Valves
SuperTech dual springs/Ti retainers
3 angle valve job
Skunk2 Tuner stage 1 cams
TR3030R Turbo/.63 A/R
Garret Intercooler core: 24” x 6.3” x 3”
Al charge piping: 2” hot / 2.5” cold
Full Race AC Manifold
Tial MV-R 44mm wastegate
3 inch Vband DP with closed loop dump
3 inch Thermal Turbo Exhaust
QTP E-cutout
Manifold/DP/turbine housing coated by Swain Tech
Skunk2 Pro Series Intake Manifold
Stock TB bored to 64mm
Tial 50mm BOV
Injector Dynamics 1000cc injectors
Walbro 400lph fuel pump
Walbro FPR
BKR7ES-11 gap set to 0.020
Hondata COP with Honda 30520-RRA-007 Coils
Hondata 4 Bar map sensor
PLX M300 Wideband


Been working on part throttle/idle tuning and when I switch from open loop to closed loop AFR goes from around 15.0 to low 12s (sometime high 11s) with no other changes. This is only at idle, I haven't tried to drive it in closed loop since this problem developed. I've tried running the wideband input on 3 different inputs, D10, D12, and D14 but it makes no difference. I looked over the wiring and it appears to be good. I have the wideband controller power wire tapped into the ECU power (in the OBDI-OBDII jumper) and grounded to the thermo housing (it has been set up like this for a long time with no issues). I do not have the stock narrow band O2 wired in.

This has been going on for over a month now and I'm running out of ideas. Has anyone had an issue like this before?

I tried changing the injector size values and dead times makes no difference. Battery voltage stays more or less the same, 13.6-13.9v, regardless of open/closed loop and fuel pressure is steady, around 43 psi. Makes no difference if I switch to closed loop voltage target or lambda target. I do notice the injector pulse width increasing but I don't understand why it would do that. Is this the cause or is something else causing this which in turn causes the rich condition. It makes no difference whether I have the O2 heater disabled or not.

Here is a datalog with me switching between open loop and closed loop. Starts in open loop. Around 1 minute I switch to closed loop (lambda target) then back to open loop around 2:01. I switch it back to closed loop but do voltage target this time at 2:41 and stay there through the end of the datalog.

I made no other changes to the calibration. I have added all three in case anyone wants to look: open, closed Lambda target, and closed voltage target)

For reference fuel pressure is datalogged on AN0. If you want to view this set AN0 to custom, pressure, psi. Offset is 0. Voltage 0.4, 4.4 Value 0, 10.34

Thanks

Try using the same closed loop settings seen in this calibration.

I let the car warm up then tried those settings but it didn't seem to help much. Datalog starts in open loop then switches to closed loop around 20 seconds. I can get more/longer datalogs if you need it.

Thanks for your help.

Are you using the same rate of change settings? What color is the wire fro the wideband that you have connected to D12? You can see that the voltage is actually different that is being read from the wideband controller when you make the change.

Spunkster wrote:Are you using the same rate of change settings? What color is the wire fro the wideband that you have connected to D12? You can see that the voltage is actually different that is being read from the wideband controller when you make the change.

Yes, I used:
10-5-4-3
70-100-90-80
70-100-90-80

I verified it is the white wire I am using. This is one of the old black box PLX M300 (not sure if the new ones use the same wire colors)

I did notice the voltage changes. I opened up a trouble ticket with PLX to see if they think it could be something with the wideband controller. Waiting to hear back.

PLX has stated that if the voltage changes between open/closed loop it could be a problem with the controller.

I am going to test it tonight and see how quickly the voltage changes when I switch modes. I'm trying to figure out if the change in voltage is the cause or is an effect of something else.

I also asked them if a bad O2 sensor could cause this. This sensor is less than 3 years old and the car has only ever run on pump gas. I can try swapping in the original sensor if needed. It was still working good when I swapped sensors. I just had the exhaust out so I figured it was a good time to swap it out.

So I'm still fighting this issue and I am not certain the wideband is the problem. I currently have the PLX wideband output (white wire) going to D12(EGRL). I know the documentation says in closed loop with wideband a narrowband input is not needed.

I don't have anything hooked up to the stock O2 input. I cut the wire when I moved things around and never hooked something up to it. Could this cause a problem? Voltage on stock O2 is 3.87v with an occasional .02v fluctuation. Could this be the cause of my issue. I was thinking of running the narrowband output from the M300 to the stock O2 input when I get the chance.

You do not have to have anything connected to the o2 input.

Spunkster wrote:You do not have to have anything connected to the o2 input.

Thanks. That is what I thought. I grabbing at straws at this point. I may hook up the narrow band output and see if it runs "normal" in closed loop narrow band. Then I guess I'll go through all the wiring again. This is very frustrating.

Ok this is getting really weird now.

I tried running the narrowband output to D14. In open loop the car now runs very lean wideband reading 18:1 to lean/air displayed on the display.

If I disconnect the narrowband the AFR drops to 14.7:1 It still runs very rich in closed loop (with or without the narrowband hooked up)

Any one have any ideas?

I'm going to post this new development to PLX too.

I'm looking at wiring under the car and I noticed the stock oil pressure sender is not connected, the plug must have popped off. I'm shocked this didn't throw a code but I don't see how this could be related to my rich in closed loop issue but stranger things have happened.

I still have the car up in the air and am going through the wiring plus I want to test the wideband sensor/controller so I won't start it up anytime soon to figure out if that was the issue.

Well I just tried a brand new wideband and the same exact thing happens. Open loop runs 14.7-15.0 in closed loop AFR drops to 11. I have triple checked the wiring and it is good.

I don't see how this can be anything other than a calibration issue which I have gone over with a fine tooth comb and found nothing that looks wrong or a problem with my S300. I'm leaning in favor of an S300 problem. Anyone from Hondata have thoughts on this?

The s300 itself does not affect the operation of the ECU, so the closed loop behavior will not change.

Are you still using the ID1000 injectors? The injector size and injector dead times are incorrect in that case. The dead times need to be accurate for decent closed loop operation.

Yeah, still ID1000

Right now I have the dead times at:
7.99 - 3.02
10.01 - 2.10
12.04 - 1.45
13.96 - 1.04
15.98 - 0.72

This is with 43 psi fuel pressure with vac connected. I can't post up the cal right now, I can later tonight.

I have the injector size set as 1000.

I've seen/read the posts about the ID 1000 injector sizing issues and tried to incorporate that knowledge into my cal file but I guess I'm just not doing enough. I've tried using 880 and 900 as a size while playing with dead times but nothing seems to work as well as the way it currently is, minus the rich closed loop issue. I guess I just need to work on it more. This is quite frustrating, I appreciate your patience.

The dead times are a little low at 14 & 16V. I'd use 12V=1.50ms, 14V=1.20ms, 16V=0.92ms