Page 1 of 2

Unable to tune in open loop but works fine closed loop....

Posted: Sun Oct 09, 2016 4:56 pm
by simjosh
- serial number of the s300 inside the ECU
37750
- version of software being used (Help, About)
2.5.4.0
- the calibration (as an attachment)
attached
- a datalog of a problem (as an attachment)
attached
- what sort of vehicle you have
1995 Integra RS
- what sort of ECU you are using
P28 Chipped from HAMotorsports
- anything about the car which has been modified from stock which may affect the ECU (don't include what sort of rims you have nor stereo etc).
Built bottom end - JE Pistons 81.5mm 9.0:1cr, Eagle Rods, ID 1000cc Injectors, Greddy TD0518G Turbo, Full 3'' downpipe and exhaust


Im currently having issues tuning my vehicle. I've adjusted every setting I can think of with multiple combinations but nothing changes...
My AFR consistently bounce, its erratic with no setting having an effect on results. I've tried running all idle cells at the same values to reduce discrepienys between too large changes from cell to cell. This did not work either.
I accidentally set up a new calibration with closed loop enabled and viola it works beautifully... finally figure out the cal wasn't in open loop so i switch it and boom back to erratic AFRs. I reduced my short and long term fuel trims to 0 thinking that if closed loop was altering anything that would stop it. This worked I removed the ability for the closed loop operation to adjust fuel so I can tune... But I'm still noticing problems that don't make sense. If I download information from the unit I've gotten datalogs even though I've never enabled on-board datalogging and I also notice information missing, IE I have a -.61 wideband correction. (Yes I followed the recommended installation for my PLX Wideband)

I also seem to have larger then normal ECT and IAT numbers... I recall these figures intimately from Crome so I cannot guarantee Crome wasn't misrepresenting the values over Hondata.

I Have a red light on the unit that blinks and a solid green light once my computer is plugged in.

I also randomly have issues datalogging. Only fix to correct this is to unplug usb and plug it back in. No other combinations will solve this issue and it only happens 10-20% of the time.
Knock Level displays 5v on datalogs.. I don't have a knock sensor so I'm not sure if this is a standard value...

Im attaching 2 logs and 2 calibrations. The ONLY difference between the two is one has open loop selected and the other has closed loop wideband...

Thank you for any help you can provide!!

Re: Unable to tune in open loop but works fine closed loop....

Posted: Mon Oct 10, 2016 8:30 am
by Hondata
The short and long term trims are set to 0%, so closed loop will not do any fuel adjustments.

Injector dead times are low.

IAT could be broken - it stays at 126F.

Re: Unable to tune in open loop but works fine closed loop....

Posted: Mon Oct 10, 2016 8:51 am
by simjosh
Engine was heat soaked and completely warmed up.

Closed Loop intentionally set to 0.

If you watched open loop datalog you can see how my AFR's get erratic quickly. WHY?

With Closed Loop Fuel Trims set to 0 this so in theory reflect OPen Loop, YES?

If so why doesn't open loop work?

I can take longer datalog's from cold starts in open loop but you'll just see the same exact scenario which is impossible to tune out of. AFR's will roller coaster 10-18 12-15... Doesn't matter what map, what settings. If it's in open loop AFR's are erratic and un-tunable.

If I run in closed loop with 0 trims I can actually tune. Why does this work? Again I'm intentionally setting closed loop fuel trims to 0 to see if the AFRS would go erratic as they do in open loop? Is there another setting that should be adjusted to keep AFR's from running erratic open vs closed? The way I look at it Closed Loop 0 fuel trim IS open loop? That's my theory at least. If this is true why do AFR's not run erratically?

I've tried tuning this vehicle for over a month and cannot find one common denominator between switching off CL and on OL that makes this erratic behavior. The vehicle is un-tunable because of this erratic behavior. My tuner sent me a generic P74 basemap with injector's and map scalar changed and stated this should be moderately stable so that tuning can take place.

Unfortunately it's so erratic I cannot figure out what or where to start tuning. But if I throw closed loop 0 fuel trim I can actually start tuning. IS this a bug? This doesn't sound right to me. I'd assumed i was my vechile for the longest time but if it runs open loop 0 fuel trim then the computer shouldn't be making adjustments and this means there's something wrong with my software or board?

I can recreate a new base map with higher deadtimes (as suggested in deadtimes list provided by hondata) and starting it cold open loop so you can see how erratic it is.

Then I'll use the same basemap but Closed loop on a cold engine with 0 fuel trim. Vehicle responds normally and is at least somewhat tune able.

Any other suggestions for changing any settings? I know what the outcome will be as I've tried every variation with old RC injector's and new ID injector's but it'll ping pong regardless (unless CL with or without fuel trims)

Re: Unable to tune in open loop but works fine closed loop....

Posted: Mon Oct 10, 2016 4:31 pm
by Spunkster
Have you removed the resistors for using the D10 input?

Re: Unable to tune in open loop but works fine closed loop....

Posted: Mon Oct 10, 2016 4:50 pm
by simjosh
Spunkster wrote:Have you removed the resistors for using the D10 input?
Yessir. Would you like any other logs with just the open loop set? Engine warm, cold, ? Or any additional information?
Also wouldn't resisters affect it more in closed loop vs open?

Are there any other areas that closed loop affects in a major way out side of short/long term fuel trim? As I would like to reflect that change into open loop to see if afrs would settle down.

The longer it runs in open loop the more it progresses toward a wider lean ratio. In the Datalog I supplied (Datalog open loop), it was switched within the first 2-5 seconds and as it runs afr progressively expands. At first maybe 12-14, then 12-15, 11.5-15.5, 11-17 ish before I shut it down.

To me it feels as if open and closed are functioning backwards, yet closed loop does function correctly with fuel trims. Yet doesn't spike without them which is what I assumed open loop should look like...

Let me know if there is anything else you'd like to see.

Thank you for the quick response also!

Re: Unable to tune in open loop but works fine closed loop....

Posted: Tue Oct 11, 2016 9:23 am
by Spunkster
Where do you have the power and ground connected for the wideband?

Over half a volt of voltage offset indicates there is some grounding problem.

Re: Unable to tune in open loop but works fine closed loop....

Posted: Tue Oct 11, 2016 9:47 am
by simjosh
Power is connected to the fuse panel under dash and ground is attached to the thermostat.

I ask why this matters? You also asked about the D10 resistors.

What effect will o2 power have on open loop running?

I can understand if my closed loop ran like garbage.... But Closed loop 0 trim runs like a normal engine would if it was open loop. Closed loop with fuel trim also does a decent job of correcting fuel but I'm trying to tune it so that when I do go closed loop w/ trims it runs more efficiently...

Only other option I can take for Wideband is to tap into A 24 / A 25 for ground and power which I thought about doing but didn't want to cut into any more harness wires (since I already have for D10).

If you think this will have any difference in open loop running I can attach them to A24/25 as this would probably be my cleanest / most reliable source for power / ground.

Let me know what you think. As I'm not confident in driving the vehicle as is to work I'll have to wait until I get home to rewire the AFR gauge.
Would you like another datalog? Also I've made slight alterations to map per earlier recommendations and other posts on the board.
-Dead times set to hondata specifics / injector size 890 ( read in a couple threads this was the most reliable CC sizing )

I re-verified Timing (Dead on at 16*)

Anything else I should check or change?

Thanks for you continued assistance.

Josh

Re: Unable to tune in open loop but works fine closed loop....

Posted: Tue Oct 11, 2016 10:25 am
by Spunkster
If you are getting inconsistent AF readings in open loop then the only thing that can really affect that is your injector dead times.

You should leave the other connections as they are, but you really should not have to use that high of a voltage offset.

Re: Unable to tune in open loop but works fine closed loop....

Posted: Tue Oct 11, 2016 11:10 am
by simjosh
I've got absolutely no problem with my AFR gauge readings or reading correctly....
It's the actual ability to tune the entire car that's my issue. I'm still unsure why my wideband wiring is your focus when the car basically cannot run when open loop in Smanager options is selected..?
The AFR will bounce drastically at every RPM, throttle position, engine load... Nothing can cure this. Injector times will not, making every cell the same timing and same fuel. A reliable source suggested these alterations as he thought there was too much variation between cells and the car / ecu couldn't find it's happy medium to allow tuning...but selecting Closed loop solves it (no fuel trim)...

Why can it run closed loop no fuel trim without ping ponging everywhere? (especially since it's in an open loop condition regardless of the CL selection at cold start / idle, yes?) This is the same map to a T minus the selection in Smanager for closed vs open loop. What does selecting Closed loop in Smanager change/affect on the map or ecu besides short/long fuel trims?

What would cause fuel delivery to drastically bounce all over the place in Open loop?
----- If tuning a car and you select open loop and you get AFR readings bouncing from 10-18afr (regardless of rpm, load, or throttle position) how do you tune this? With injector dead times?

I'll run more logs today. I don't think I've allowed you to see the significance between the two accurately.
1 log from start up completely in open loop
1 log from start up complete in closed loop (no fuel trim)





Does the car idle on start up in Open or Closed loop? (I thought start up / idle and low load / low tps was an open loop operation no matter what...)

That's the biggest question I have. Because if a car is starting up idling in Open loop no matter what(?) why does me selecting Closed loop (no fuel trim) make such a large difference?

I'll be able to post a better logs with the extreme difference between the two today around 6pm central. I apologize if I've repeated myself or been unable to accurately describe the problem as I've seen it.

Thanks again for any and all assistance.

Re: Unable to tune in open loop but works fine closed loop....

Posted: Tue Oct 11, 2016 3:18 pm
by Spunkster
I have just tested this in my own car using current software and a PLX wideband that has been installed for many years. It runs as it should in closed loop, and when I change to open loop it no longer has a fuel trim and holds a pretty steady AF ratio. My voltage offset is -.08volts.

With the wideband input disconnected what does it do in open loop?

Re: Unable to tune in open loop but works fine closed loop....

Posted: Tue Oct 11, 2016 4:56 pm
by simjosh
Ok Spunkster, Ive loaded my new map and a datalog.

This map was idling pretty decently but still had some hiccups. No where NEAR as chaotic afr flux as before. + or - 1 at the extreme and .2 .5 as the norm...

One huge thing I'm noticing is that ELD is still receiving voltage. I had open loop and I selected D14 pin (currently D14 is disconnected from O2 sensor)

How would ELD be receiving voltage without any wires attached to D10?

Is this my problem? Faulty ECU?

I also noticed my K Level is 5v but I don't have a knock sensor? That is normal?
Thanks a ton!

Re: Unable to tune in open loop but works fine closed loop....

Posted: Wed Oct 12, 2016 6:57 am
by simjosh
https://www.hondata.com/forum/viewtopic.php?t=18296

http://www.hondata.com/forum/viewtopic.php?t=9906

Since I was able to refine my problem to D10 creating my surging / erratic running I found the two posts above. There were no essential fixes besides not using D10 that I could see..

I'm attaching a picture of my ECU. My hope is that you might be able to see missing components or something else that might confirm the issue with ECU hardware vs anything else. After tons of reading and some clarification my "P28" was re-chipped to actually simulate a P28 so it could have originated from any number of original ECU's and also is not new and more then likely capacitor's were not changed.

Let me know what you think!

Thanks a ton for helping me with this issue Spunkster!!!

Re: Unable to tune in open loop but works fine closed loop....

Posted: Wed Jun 21, 2017 7:40 pm
by forevertrj
I'm now having this issue with idle surging in open loop only, using D10 for my wideband input source. I'll upload cali and log when my tuning laptop recharges. I don't think the ecu has ever received new capacitors so I'll be changing those and also doing some reverse engineering on the ELD circuit to see what it touches. and try D12 or B6 inputs.

Re: Unable to tune in open loop but works fine closed loop....

Posted: Wed Jun 21, 2017 8:13 pm
by simjosh
forevertrj wrote:
Wed Jun 21, 2017 7:40 pm
I'm now having this issue with idle surging in open loop only, using D10 for my wideband input source. I'll upload cali and log when my tuning laptop recharges. I don't think the ecu has ever received new capacitors so I'll be changing those and also doing some reverse engineering on the ELD circuit to see what it touches. and try D12 or B6 inputs.
The only way for my stuttering to disappear completely was to cut the wideband wire. Grounded to thermo housing like it's supposed to be. I never tired obtaining power and ground off the ecu... So that could be something else to try. (You are getting no stuttering closed loop 0 trims huh? ) If yes, find out what receives, sends or gets altered when switching from closed to open loop...
Hondata could never give me a specific area to check wiring or ecu wise that open vs closed loop changes. I'll either have to buy a new ecu and or complete wire harness before I blame the chip.
Good luck with your issue. Unfortunately I feel that this issue is too random and a low priority (for Hondata) to be a concern since it'll run fine closed loop/ 0 trims...

Keep me updated if you figure it any solutions. I would greatly appreciate it! Closed loop 0 trim does work but I seem to have issues with ect/iat compensations and a few other minor things..

Re: Unable to tune in open loop but works fine closed loop....

Posted: Sat Jun 24, 2017 9:42 am
by forevertrj
I took a datalog of me changing the calibration from closed loop to open loop so I'm going to look at that today, in the past my car ran fine with open loop and the wide band on the Primary O2 input D14, and nothing on the ELD input D10, my ground is on the chassis by the fuse boxes, and my voltage difference to the ecu is 0.08volts when reading the output from my wideband to what shows up in the Smanager software. I'll try a few more things today and update the thread and also upload my datalog of live switching from closed loop to open loop. I never needed to set closed loop trims to 0 as the car is just in the driveway and not insured at the moment.