Closed Loop - L.Trim questions

[jun1or]

Edited:

1.) Does L.Trim affect fuel maps so that they are in place permanently until I pull the negative battery cable / ECU?

2.) I've seen a few negative L.Trimm percentages one day then 0 the next. Does it apply the change and then start back at 0 the next day with new adjusted fuel table(s)?

3.) Does simply disabling Closed Loop operation reset L.Trim? Or is it just disabled temporarily while open loop operation is running?

4.) Does uploading a map in Smanager reset L.Trim? For example, If I encounter -4% L.Trim during closed loop driving for week, can I just upload the same map from the week before to reset L.Trim or does it retain those settings?

Really what I want to know is:

How is L.Trim applied, and when is it reset (just pulling battery cable only?)

Mike

[fly_n]

The L.Trim figure is a standalone figure which is just included in the calculation to derive your final fuel figure.
It doesnt change the values you entered in the fuel tables.

L.Trim is held in memory until the ECU is reset,i'm not sure if toggling between open & closed loop resets the L.Trim figure or if uploading a new map will reset it either,i've never tried or bothered looking into that.
My guess is it doesnt.

When part throttle tuning,i usually just out of practice reset the ECU before switching back to closed loop & logging the outcome of my alterations,to make sure no L.Trim figures are messing with the results,even though u can easily just read whatever L.Trim figures are occuring in the logs.

If u experience fluctuating figures like u describe above,then maybe just a particular section of your part throttle area of the table needs attention.
Take note of any area of the map u were driving in to cause the L.Trim figure to start dropping away that might be different to how u were driving when it wasnt.

To save pulling the battery cable all the time just locate your ECU "back-up" fuse(not the main ECU fuse) in the engine bay fuse box(if its the CRX your talking about,then its the "Hazard" fuse) & just pop that out for 10secs while u have the ignition turned off to reset the ECU.

Or if u dont want to have to keep resetting your clock/headunit presets/etc then just pop the D-plug out of the ECU for 10secs while igntion off. Dont recommend to keep plug/unplug ECU plugs however,its not good practice for the sake of connection integrity.

Just remember to validate the correct operation of your O2 sensor before chasing your tail over poor closed loop operation.

[jun1or]

The L.Trim figure is a standalone figure which is just included in the calculation to derive your final fuel figure.
It doesnt change the values you entered in the fuel tables.

Thanks fly_n, your input is always helpful. I was always curious.

I'd really like to know if "uploading" in smanager reset's L.Trim, but trial and error will inevitably play a part in finding this out.

Thanks again.

Mike

[jun1or]

Stupid question - Does L.Trim value apply to the whole map(s) or just part? I would assume all of it.

The reason I asked the L.Trim questions in post #1 is because:

I keep enabling closed loop like a moron wanting to get good gas mileage on my hour of driving to work, then disable and do some part throttle tuning.

I witnessed -7% L.Trim develop (yikes) at a long idle session, quite rich. So if I went WOT to redline at that point, would I be yanking 7% fuel from that opposite end of the map (different map too obviouslly, high speed map).

Thanks again,

Mike

[fly_n]

Long term trim(& short term trim) only occur in closed loop operation & hence only in the closed loop portion of your fuel maps which is defined by the load value entered at "closed loop tab/conditions/max MAP for closed loop".

When u breach this load value &/or go WOT,ECU goes to open loop operation,therefore no L.trim occuring.

I notice in your other thread u say your WB readings are inaccurate,so no real point chasing a solution to your L.Trim figures till u get that sorted,even moreso if your using a combo narrowband/wideband sensor.Remember my comment about chasing your tail?

[jun1or]

Long term trim(& short term trim) only occur in closed loop operation & hence only in the closed loop portion of your fuel maps which is defined by the load value entered at "closed loop tab/conditions/max MAP for closed loop".

When u breach this load value &/or go WOT,ECU goes to open loop operation,therefore no L.trim occuring.

I notice in your other thread u say your WB readings are inaccurate,so no real point chasing a solution to your L.Trim figures till u get that sorted,even moreso if your using a combo narrowband/wideband sensor.Remember my comment about chasing your tail?

Bingo, thanks fly_n. Stupid me, I realized after I posted that WOT run would be open loop as defined by my transition point (stock).

Yep, chaising tail comment was good. I am, fortunately, using a narrowband AND a wideband as the AEM does not output narrowband signals.

You are correct, my Wideband is messed up - AEM UEGO. Changed it's grounding point yesterday, no luck. (although I only moved it from hood release bolt to steering column). Also cleaned battery cables and grounds to car and chassis.

Believe it or not I may have all these strange problems from an electrical issue. I logged an idle session demonstrating lots of unusual stuff going on:
Notice at 47sec, battery voltage fluctuation conincides with crappy idle, MAP stuttering, and S.Trim adjustments. (wideband is all over the place, I don't trust it) And yes, alternator control is disabled.

My conclusion - Wideband sensor AND/OR battery, or alternator is bad. I'll pull the wideband tonight and do the brake cleaner test. (http://www.youtube.com/watch?v=1mb4VmDd0ao)

Many thanks,

Mike

[fly_n]

Those are poor ground locations for the WB.
The ECU is referencing the main EFI ground location at the thermostat housing on the motor & your WB is referencing a location quite a few connections between copper/bolts/steel chassis/corrosion?/paint? away from the ECU.
Try a dedicated ground line more closely associated with this main EFI ground point without upsetting the integrity of the EFI ground point,ie dont just simply sandwich a lug between the EFI ground lug & the thermostat housing.

Did u try flat-lining the analog voltage output of the AEM to the ECU by disconnecting the sensor from the AEM gauge unit & seeing how it compares between ECU & gauge?

The battery fluctuations just coincide with the fall in alternator output as your idle keeps dropping away.

Note your ECT sensor hardly moves the whole couple of mins u have the engine running for,it should be rising steadily towards full operating temps,take a closer look at this sensor & its wiring & perhaps also your thermostat.

47secs also signals the ECU going from open to closed loop,notice the 10% fall in your ECT corrections & the introduction of some S.Trim?
Notice also the fluctuations in the MAP readings?

To my untrained eye,the problem exists in the form of either,
- ECT sensor &/or wiring,
- maybe just the fact that uve lost your way in the state of the tune due to your WB problems,or
-an air delivery problem like a vac leak/idle valve/throttle body issue,
-maybe even some combination of the above.

U could even have a compatibility problem associated with trying to run a P72 idle valve from a P06 ECU,what throttle body are u using by the way & does it have a FITV?

Dont waste your life chasing my suggestions,but hopefully they might promote yours or someone elses thinking on the issue....

[jun1or]

Those are poor ground locations for the WB.
The ECU is referencing the main EFI ground location at the thermostat housing on the motor & your WB is referencing a location quite a few connections between copper/bolts/steel chassis/corrosion?/paint? away from the ECU.
Try a dedicated ground line more closely associated with this main EFI ground point without upsetting the integrity of the EFI ground point,ie dont just simply sandwich a lug between the EFI ground lug & the thermostat housing.

I figured that was coming. Agreed.
It has worked for 3 months (used sensor!) no problem then boom - full lean readings all the time(matching the youtube video I sent about bad sensor), but I will change ground location regardless. Thanks.

Did u try flat-lining the analog voltage output of the AEM to the ECU by disconnecting the sensor from the AEM gauge unit & seeing how it compares between ECU & gauge?

Yes I have done this. I only have -0.1v offset with R136 and R138 cut on the ECU and it has matched perfectly until recently. I bet it is a bad sensor.

The battery fluctuations just coincide with the fall in alternator output as your idle keeps dropping away.

Makes sense now.

Note your ECT sensor hardly moves the whole couple of mins u have the engine running for,it should be rising steadily towards full operating temps,take a closer look at this sensor & its wiring & perhaps also your thermostat.

It was very cold outside here in lovely Wisconsin and my car was warm but not fully warmed up. I datalogged yesterday fully warmed up and ECT was 183.2F.

47secs also signals the ECU going from open to closed loop,notice the 10% fall in your ECT corrections & the introduction of some S.Trim?

Open to closed loop - that transition kinda makes sense now, so battery voltage fluctuations are just a byproduct of that given alternator output perhaps. How do you know for sure when closed loop starts?

Notice also the fluctuations in the MAP readings?
To my untrained eye,the problem exists in the form of either,
- ECT sensor &/or wiring,
- maybe just the fact that uve lost your way in the state of the tune due to your WB problems,

Now this is a likely scenario! My wideband is acting goofy, works good, then full lean, repeat.

-an air delivery problem like a vac leak/idle valve/throttle body issue,
-maybe even some combination of the above.

U could even have a compatibility problem associated with trying to run a P72 idle valve from a P06 ECU,what throttle body are u using by the way & does it have a FITV?

4 months ago I installed a new Skunk2 intake manifold and 70mm skunk2 throttle body. No FITV on that TB of course. IACV is ITR I believe, should work.

Dont waste your life chasing my suggestions,but hopefully they might promote yours or someone elses thinking on the issue....

You've been most helpful, thanks for your comments!

Mike

[fly_n]

I dont know the exact procedure or parameter of what triggers the ECU into closed loop after the open loop start-up/warm-up,but the major clue of it happening in your log is the S.trim becomes active right at the 47sec mark.

Also the ECT correction disappears at that exact point as well,which matches what is detailed in your fuel compensation tables for an open to closed loop transition.

Yep,sounds like u got faulty WB sensor problems rather than voltage offset probs.

Dont have any experience with cold weather,not of the Wisconsin variety anyway(the dataloggable,loud,low, track whore crx dont go on my annual 1000km ski trips lol) but with all thats going on inside a running engine equipped with a thermostat,u'd think it would still keep warming up past 60degC regardless of ambient air?
Ah, i dont know,but i'd keep my eye on that ECT....

If it were me,i'd probably,
-check the intake install for leaks,might even be intermittent due to temperature
-confirm good & proper operation of the throttle plate,ie no binding/proper closing/etc
-check base idle(IACV disconnected) & mechanical timing
-get WB sorted & reset idle valve slider to default middle position & then try find some sort of stable idle from there,maybe even try borrow a known good idle valve off someone to test as well while your at it.

[jun1or]

I dont know the exact procedure or parameter of what triggers the ECU into closed loop after the open loop start-up/warm-up,but the major clue of it happening in your log is the S.trim becomes active right at the 47sec mark.

Also the ECT correction disappears at that exact point as well,which matches what is detailed in your fuel compensation tables for an open to closed loop transition.

Yep,sounds like u got faulty WB sensor problems rather than voltage offset probs.

Dont have any experience with cold weather,not of the Wisconsin variety anyway(the dataloggable,loud,low, track whore crx dont go on my annual 1000km ski trips lol) but with all thats going on inside a running engine equipped with a thermostat,u'd think it would still keep warming up past 60degC regardless of ambient air?

It heats up to 84C every day, all day. That 60C pull wasn't fully warmed, sorry.

Ah, i dont know,but i'd keep my eye on that ECT....

If it were me,i'd probably,
-check the intake install for leaks,might even be intermittent due to temperature
-confirm good & proper operation of the throttle plate,ie no binding/proper closing/etc
-check base idle(IACV disconnected) & mechanical timing
-get WB sorted & reset idle valve slider to default middle position & then try find some sort of stable idle from there,maybe even try borrow a known good idle valve off someone to test as well while your at it.

Good list to go off of, thanks for spending the time to help! I'm going to set the idle and the timing just to be sure.

1000km is a long drive, holy $^*%!
I drove 2,092 Kilometers to live in Jackson Hole, Wyoming which is a big ski resort with 3800 meter mountain! (do they measure mountain in meters? I used google to convert), but to drive that just for a short trip seems like a lot of driving!

Mike

[jun1or]

http://www.youtube.com/watch?v=onbjzEOZtMs

The sensor was bad. The video above shows the faulty sensor going full lean on a truely rich mixture. Replaced sensor.

Mike

[jun1or]

Long term trim(& short term trim) only occur in closed loop operation & hence only in the closed loop portion of your fuel maps which is defined by the load value entered at "closed loop tab/conditions/max MAP for closed loop".

When u breach this load value &/or go WOT,ECU goes to open loop operation,therefore no L.trim occuring.

Update!

From Hondata tech article on Closed loop:

Under high load when the ECU stops running in closed loop the long term adjustment is still used so increasing fuel delivery via these means in not recommended unless the ECU is recalibrated or closed loop disabled.

http://www.hondata.com/techclosed.html

Mike

[fly_n]

lol,i always thought that should be the case until someone else told me otherwise,guess u should rely on your own instincts.....

[jun1or]

lol,i always thought that should be the case until someone else told me otherwise,guess u should rely on your own instincts.....

Ah no worries, I just found this myself. The tech articles are pretty good. I wish there were more!

Mike