A/C PID Question

[Big Mike]

Could someone explain the nature of the "A/C compensation spark" and it's related PID {GM.EST_ACB_DMA}. I'm under the impression that it only assist in spark adjustments to reduce torque as the A/C cycles on/off.

Doing a random log today using that PID I noticed it's very active both at a stationary idle as well as at speed without the A/C even being on. It's seems to make some rather major spark adjustments for reasons unknown...without the A/C even being on?

So is there something else that uses this spark adjustment too that I'm not aware of? :help2:

[Big Mike]

Just noticed the log didn't attach. :doh2:

[joecar]

Hi BM,

My understanding is it increases torque output when A/C cycles on and vice-versa.

Very interesting that it is active even when A/C is off...

I'll look at your log file later today.

[Big Mike]

Thanks for taking a look when you get a chance Joe. If you look at the log you can clearly see that the A/C spark correction is always reducing timing as intended of course to reduce shock load on the A/C compressor as it engages. I just can't understand why it would be active with the A/C off. Has me and every one else stumped at this point.

-Mike

[joecar]

...
If you look at the log you can clearly see that the A/C spark correction is always reducing timing as intended of course to reduce shock load on the A/C compressor as it engages.
...Ah, I get it, to reduce shock load, not the other way round, ok... :cheers:



The STATE pid shows A/C not being requested and not being commanded, but yet AC timing is being pulled every so often...:nixweiss:

Remove IDLEDES, add FTC and take another log.

[Big Mike]

Here's a log from today with the added FTC PID you asked for Joe. This run was done over the exact same route as the first one and from a cold start. The only difference was ambient temp was about 15* cooler than the last.

Almost have to wonder if the A/C correction PID isn't monitoring something else. At some points it acts similar to when monitoring burst knock yet it's very active at idle in P/N. This is clearly evident at the end of the log, when I went to P/N the A/C spark correction went crazy pulling some massive timing.

Also curious what you're looking for with the FTC's in relation to this seemingly A/C correction phenomenon.

[joecar]

...

Almost have to wonder if the A/C correction PID isn't monitoring something else. At some points it acts similar to when monitoring burst knock yet it's very active at idle in P/N...
That's what might be happening, the ACB pid might have multiple roles...

Remove O2S11, O2S21 and add EST_KRB_DMA, EST_CAT_DMA and take another log, if you have time.

By cycling thru the various EST_ pids maybe we might find one that matches EST_ACB_DMA during those situations.

I'm curious of what/why is happening.



Also curious what you're looking for with the FTC's in relation to this seemingly A/C correction phenomenon.The FTC has cells to indicate A/C on/off and PN/IG, I wanted to see if the FTC agreed with the STATE00 A/C status flags...
it seems to agree (i.e. FTC never hits the A/C on cells).

[Big Mike]

I have BKR set to pull no more than -2.5 deg at these ECT's so I wouldn't suspect that to be a contributor, however I hadn't thought of cat light off spark corrections so we may be onto something there. I'll add the new PIDS and get another log.

[Big Mike]

A new log with the requested KRB and CAT pids Joe. I threw EGR in there too just in case even though it isn't equipped with one but it looks like neither of these two pids are a contributor to the timing reduction. Still strikes me as odd that it's very active in P/N even on a cold start and looking at this long specifically it appears that it's also active almost exclusively when knock retard is, but it doesn't follow it exactly. Then it becomes very active again at the end when I went to P/N. Almost like it's an idle speed over/under adjustment, but then why would it be so active during non-idle too?

Stumped more now than ever. :sly:

[joecar]

I noticed this:
when VSS is non-zero, EST_ACB_DMA has activity when DFCO changes state (according to STATE05-DFCO)... i.e. ACB spikes when DFCO becomes active and then spikes again when DFCO becomes inactive; when you see ACB in a non-spike form then it's not DFCO; also when you see DFCO active ACB does not necessarily spike; I noticed this in 2 of your logs.

In P/N it does seem like an adjustment... I think that pid is being used for multiple roles.

Stumped... yes, it makes us very curious as to what it is trying to tell us. :book:

[5.7ute]

It might be worth logging Battery voltage & seeing if it has something to do with the alternator's load as well.

[Big Mike]

I'll take another log using most/all the spark related PIDs we haven't tried yet and see where we're at. I'll also include batt voltage and alternator field duty cycle too to cover your suggestion to 5.7ute. Maybe even idle spark as well. Thanks again for the continuned assistance guys.

[Big Mike]

Something else I noticed too and goes along with some of your comments Joe. If you look at the screenshot below, ACB is calling for spark reduction yet timing isn't being affected...until it is massively once I went to P/N. Almost like it's "prepping" for an IG to P/N change ready to apply timing adjustments once the P/N change actually takes place. Interesting to say the least, so I'll definitely add the idle spark PID to the next log.

[Big Mike]

Looks like trans torque reduction request is well active and sometimes follows ACB very closely...but other times it's not even close so I wouldn't suspect they're related. Also not sure why trans TR is active in P/N either but perhaps we'll leave that for another discussion. Don't see anything odd with batt voltage or alternator DC either that could be contributing.

I know this log contains well over 24 channels but I wanted to get them all at once and I don't suspect the precision was compromised much.

[joecar]

Lots of activity, interesting log.

No, the precision is fine, I don't think we lost any events.

[Big Mike]

Several of the trans tq reduction spikes were from me manually down shifting, it's normally not that busy. :grin:

I'm a little intrigued that I'm the only (or first) one ever to experience/notice/mention this encounter with ACB. Perhaps it's just something with this OS.
Just can't think of anything else to check but I sure would like to track it down still.

I originally found all this while checking the ACB activitly in an attempt to possibly reduce the timing reduction a little as the A/C clutch cycles through it routine. So I first wanted to see how active it was before I made any adjustments in the tq loss table related to it. Glad I discovered that the ACB PID isn't telling the whole story first.