HVAC actuator technicals (not a routine question)

[Joe_67]

I know the web is full of info on GM HVAC actuators, and I've been through a lot of info and get them in a basic sense. But I'm looking for some "deeper" technicals if anyone knows (including specific conditions that set B-code DTCs).

The back story is long, so I'll only relate the whole thing if anyone really wants to hear it! But it doesn't change the Q.

The vehicle is an '02 Envoy with the auto AC.

As I understand it the actuators rely on a position signal which basically comes from a potentiometer on one of those (crappy, cheap, plastic) gears inside of the actuator. I'm not worried about the specifics in that respect - but something like a 5V ref is sent in and the pot moderates to it range between - call it something like .5V telling the BCM all the way one way and 4.5V all the way the other way.

But let's say that something makes it so that the actuator only does half it's travel. (Maybe an obstruction in the vent work that, during calibration, "tricks" it into thinking it hit a limit or some weirdness with the gears). Theoretically, now the signal voltage only goes like 2.5V to 4.5V. In common-sense terms the BCM should "know" that's a problem - not the full range of travel - and set a code. But maybe it doesn't - that's just a human programming thing with people putting "if-then" statements into the software. Maybe they never though of that one. So I guess the short question is, if an actuator is working fine but can't do its full range, is the BCM programmed to "know" that?

(Ok. Short part of the long story - mode actuator that was throwing B3770. Only blows floor. This is the classic mode actuator failure - defaults to floor to favor proper defrost operation. Mode actuator replaced. B3770 gone, and blows face, but actuator literally only goes 1/2 way to floor so won't go full floor (so then I don't get full defrost when needed). But I've now got now codes. Should the BCM "know" that it's not going the full range? Or maybe I'm just waiting for the code... System recalibrated (by HVAC-B fuse pull) like 4 times. All else is normal).

 

[Mooseman]

Maybe a crap new actuator? Only way to really know what the HVAC module is seeing is to use a scanner capable of reading HVAC data, a Tech 2 or a high end scanner like a $nap On.

Another possibility is the door is not able to move freely.

 

[azswiss]

Mode actuator replaced. B3770 gone, and blows face, but actuator literally only goes 1/2 way to floor so won't go full floor (so then I don't get full defrost when needed).

Sounds like the relative timing of the gears/cams might not be correct so you are not able to get full range of travel. Not sure how the BCM would respond to this situation.

 

[Joe_67]

Maybe a crap new actuator? Only way to really know what the HVAC module is seeing is to use a scanner capable of reading HVAC data, a Tech 2 or a high end scanner like a $nap On.

Another possibility is the door is not able to move freely.

LOL. It would be the 2nd crap new actuator. (Part of the "long" story. Oddly the 1st was AC Delco and never got anything from the face. Latest was a Four Seasons. Now I get face. At least the wife is happy!).

I do have a good local, long-time in operation A/C shop that I have to assume has the right scanning capability. I might take it there... but you know. They want me to pay them!

As for the door moving freely, I've found that ambiguous. It's 20ys old. Is it smooth as glass? Nope. It's hard to judge crammed when I'm crammed under the dash like that barely able to get a couple of fingers on it. Is it "sticky" enough to confound those wimpy little motors and their little <5V power supplies? Maybe!

 

[Mooseman]

One thing you could try is to take it out and see how much of a sweep it does when you change the controls. Doing a reset with it out, I have heard, that it could destroy the actuator because it takes it beyond its range but that could be for old ones with failing plastic gears.

Maybe try a cheap actuator since you've even had bad luck with an ACDelco. Quality of stuff has gone to the shitter lately but maybe the Chinese ones are from old stock. Another idea is if you have a U-pull yard to grab some there. Yeah, I know it's fun getting your inner Gumby out to work under there.

Another possibility is that the flapper door's weatherstripping has fallen off and allowing air through or it's partially off and blocking the door's movement.

 

[Joe_67]

One thing you could try is to take it out and see how much of a sweep it does when you change the controls. Doing a reset with it out, I have heard, that it could destroy the actuator because it takes it beyond its range but that could be for old ones with failing plastic gears.

Maybe try a cheap actuator since you've even had bad luck with an ACDelco. Quality of stuff has gone to the shitter lately but maybe the Chinese ones are from old stock. Another idea is if you have a U-pull yard to grab some there. Yeah, I know it's fun getting your inner Gumby out to work under there.

Another possibility is that the flapper door's weatherstripping has fallen off and allowing air through or it's partially off and blocking the door's movement.

I'm pretty sure resetting with it not attached will basically run it out of range - I think it relies on hard stops to determine the bottomed out door position so w/out the door bottoming out, it will just keep turning. It does have hard stops on the back of the case, but those are well outside of normal operating range. (And I think then creates "confusing" position signals for the BCM).

You can manipulate them on the bench though and bring them back into line if they've run out (

). I've played around with old ones on the bench.

The AC Delco one I have is labelled "TMI" and "Made in China" - no surprise there, and yeah the "OEM" part means less and less it seems. (Or even decent aftermarket - I once bought a pair of Moog tie rod ends at the same time. Both in Moog boxes. Functionally the same but completely different construction. One looked like the typical, solid Moog kind of thing I recognize. The other looked like cheap Chinese junk. I exchanged that one).

For now I think I've run out of energy for it, and it's working "well enough." FWIW, I also suspect something inside of the box, like the door weatherstripping is creating issues. For now, I'll just wait to see how long it takes the BCM to "think" there's a problem, if at all. Before I button it up though, I will at least look at the position signal and see what it's saying. Not today though. I'm about to get hammered by some fairly large thunderstorms... But I'll post what I learn about that when I get to it in the event that it ever helps anyone.

 

[Joe_67]

Well, the rain doesn't mean that I can't tinker around in the garage, and it seems that first actuator I bought was actually mis-assembled - the pot gear wasn't indexed correctly as far as I can tell.

I was just going to play with it - move it around with the 9V and look at what the pot resistance was doing. It was parked in floor position (all the way left if looking at front of actuator) which is what happens if the BCM detects an error and won't operate it any more. Well, the pot resistance was showing me infinite - no wonder it had the B3770. (This all started with a B3770, but that was from things going out of whack b/c of the typical cracked shaft gear on the original. But it did leave me confused when I installed the new one and nothing changed!).

Anyway, I went ahead and motored it all the other way, and at some point it started showing resistance. With some guidance from this thread: https://gmtnation.com/forums/threads/successful-re-index-of-mode-actuator-cheap-actuator-fix.3447/ - I *think* I may have re-indexed correctly.

Total range shown (when not in the no-mans-land of infinite resistance) was 0-9.8K ohms. So I just popped the shaft gear and worked the pot gear to park it right around the middle (around 5K ohms), and then popped the main shaft gear back in w/ the indexing marks centered. So, I *think* everything should just be middle of the road. I won't know for sure until the one I just installed fails ... because I am done with the gumby thing for now and not curious enough to do another r&r right now.

But I will still look at what the resistance & position voltage are doing on the newer one I just put in an report for anyone who might be wanting to know.