Think I need a new PEM Fan... Anything else to check?

[gregd]

I still have the 2 separate fans and my speeds jump around all over the place. I would be happy if they were a straight on and off, but they seem to be variable, I have not even looked at my new fans to see if they have multi or variable speeds, I’m just happy they are working again

Eh, you probably don't want on-full-blast. Besides wasting power, they're pretty noisy when they get going.

Watching my OVMSv3 display, it looks like the PEM controls the fan to keep the temps right at about 39C. They start when the PEM reaches around 35C with about 4 volts, ramp up to 7v for the session, back down after charging, and turn off again at about 35C. Ambient here was 26C during the session.

 

[jfischer]

Thanks. I'm in the Foothills, about 30 mi NE of Sacramento. It'd be a long drive for you, but perhaps some live phone advice might be a starting place. Not there yet...

An update. I inhibited the APS, and opened The Cover to look at the fuses. They're 20 amp models on my car, and both check out good. So, that's not it.

I may try to stick my hand through the port hole in the trunk to see if I can wiggle the fan connector a bit. Not as a fix, but rather to see if rubbing off a little corrosion might be enough to prove where the fault lays. As I recall, however, the connector is pretty tightly anchored, so I doubt that I can get enough force on it to do anything.

Any other ideas?

You ALSO might want to check the TWO connectors on the underside next to the FAN. Just pull the diffuser rear underneath cover. 4 bolts and about 10 screws. Comes off easily.

Mine were both fried causing 1144 errors On my 2.5 version.

Replacing them with a better sized connectors to help eliminate this design flaw.

 

[jfischer]

@Bunnak yes, it certainly looks that way. I'm going to attempt the fuse thing first (under the cover), hoping that a history of trying to push the fan hard caused one of them (the PEM side) to fail, and that replacing it will let the motor run long enough to get it into a service situation. I haven't decided whether that will be at the Tesla SC or my own garage. Cost tells me to do it here, the hassle factor is undecided, unfortunately. A major hassle either way. I'm tempted to do it here, just because the past 3 years Annual Maintenance have resulted in a car that was less reliable than when it went in, so I fear I will need to be able to work on this myself sooner rather than later.

Question to anyone who has replaced the fan assembly: Is it just common screw driver + wrench sorts of bolts, or does one need to remove any of the suspension parts? I draw the line at the suspension, and that bar that goes across the unit looks scary. I also have no idea how I am going to raise the car up enough to get under it.

EDIT: Oh, and a PSA to anyone who is depending on OVMS to alert them to the dreaded 1144 / 1146 messages. At least on my Android version, they are suppressed because they are considered "diagnostic" messages. However valuable, especially to warn you before the car actually shuts down on the freeway, they're no longer presented in real time. You have to grab the logs and look manually.

As far as raising the car.
RinoRamps work great. Got them at homeDepot for 39.00
Plenty of room to work and test.

 

[gregd]

You ALSO might want to check the TWO connectors on the underside next to the FAN. Just pull the diffuser rear underneath cover. 4 bolts and about 10 screws. Comes off easily.

Mine were both fried causing 1144 errors On my 2.5 version.

View attachment 578760 View attachment 578762
Replacing them with a better sized connectors to help eliminate this design flaw.

Yep, a few posts back I suggest that, in fact, a failure such as yours can be the cause of the failure up at the PEM-end of the wiring harness. Those connectors don't seem to have the tightest grip on each other, from a contacts perspective. I can insert and remove a standard "F-1" blade into mine easily with two fingers. Not at all what one would expect from a connection capable of passing 10+ amps continuously without failing. Yours got hot enough to melt the plastic. Mine seem to be loose enough that they didn't even generate much heat. Further, that's the positive pin on yours that melted. As I detailed earlier, the two positive sides go through fuses and are immediately after tied to the positive rail of the APS supply. The regulation of the motor speed is on the negative side, with both supplies acting together. Because the two sides are paralleled on the motor side of those connectors, if one of the positive pins at the motor fails, ALL of the return current flows through the other side's positive pin on the PEM end of the harness, doubling its load. At that point, failure at the PEM connector is practically assured. It's easier to see this effect if you look at the diagram I posted, tracing what happens if your one pin got disconnected.

There are several poor engineering choices in how this part of the car is put together. Primary is the severely under-spec'd connector at the PEM, followed by your connector example at the motor. The design is also not fault tolerant, as a failure such as yours can precipitate another failure, on a significantly more expensive part. Without redesigning the PEM, at least the two sides should have been tied together at the PEM, with the motor being fed through a single, properly spec'd connector. That would prevent the cascading failures. Sadly, the SPAL motor actually comes with the right kind of connector already on it. I'm guessing that Tesla, to retain the original wiring harness, cut the original connector off the motor and spliced in the two that we have now, wiring them together on the motor side. {Sigh...}

 

[gregd]

As far as raising the car.
RinoRamps work great. Got them at homeDepot for 39.00
Plenty of room to work and test.

View attachment 578773 View attachment 578774

Looks VERY familiar. Except my car is Gray.

It looks like you have the blower assembly off, or are in the process of swapping it out. Did you run into any challenges? Anything to warn about?

 

[gregd]

Quick update...

I figured out that the white plastic cover on the female side of the connectors that attach the wiring harness to the blower motor under the car are simply snapped on. The cover is clipped in with 4 small plastic pegs, and a thin blade or screw driver wedged gently under the outside edge will remove it. Under the cover are the pair of female F connectors. A gentle but firm squeeze with some needle-nose pliers can tighten them up, substantially improving the connection.



Test fitting a male F spade showed the motor side of mine was especially loose. If nothing else, if you replace the pins on the PEM-end of the harness, check the blower-end for a tight fit. Per my earlier discussion, I believe the looseness of the motor driver side shifted the load to the PEM driver side, causing the PEM side's positive pin under the PEM to fry.

Doing a bit of digging, I think the female side housing is an AMP #344081-1. Mouser seems to have them in stock. My favored plan at this point is to put a new housing on the end of my test cable, which is already attached to the PEM, after parallel-connecting the two driver circuits under the car. That will prevent the load shift from occurring in the future. The end result of this is that I will have replaced the blower motor part of the original wiring harness, leaving the original one unused and unaltered (fried pins and all).

Now that I've "fixed" the blower-end connectors, I could splice my new PEM-end connector into the original harness, after doing the circuit paralleling, but it's kind of tight under there, and this way I don't have to touch the virgin connector that's attached to the PEM right now. I know they're "rated" for 25 cycles, but seeing how they are constructed, I wouldn't trust them more than once, or twice if you're really careful when removing them. Do NOT wiggle the connector when removing or installing it, as that could easily bend the "spring" (ish) part of the pin, causing it to not make good contact in the future.

 

[gregd]

Hi folks,

On the home stretch, nearing the finish line. The decision is to put a new set of blower-end connectors on my test cable, with the cross-connect of the two supplies as discussed earlier. This keeps the car as close to original as possible, but with the cross-connect to aid in keeping the two sides balanced. I'm leaving the PEM-end of the test cable as-is. I've been measuring the current from both supplies during the daily top-off charge, and it appears to be performing well. Overall, this final design minimizes any changes to the original wiring harness (the change is fully reversible), maintains use of the progress so far, and is the easiest to complete. Basically, I'm replacing the PEM-to-Blower part of the original wiring harness with a slightly better one, and with new connector parts.

Mouser delivered the new AMP connector housings and new 0.250 female F connectors (571-344081-1 and 571-344009-1-LP, respectively). Turns out that the connectors do not come with the white plastic end caps, nor the foam weather inserts that go around the wires. I have no idea what to order, so I'm re-using the end caps from the original connectors, and have "adapted" some foam earpieces for the task of keeping the weather out.



I also picked up a 1/2" split wire loom from the home improvement store for protection of the wiring harness. That slipped easily over the wires, and was anchored in place. The harness goes up the right hand side to near the back corner of the PEM, well away from anything moving. It's held in place on the upper end by a tie to the wiring that goes to the ABS fuse holder, and on the bottom by a tie to the corner access hole, and a cable clamp with a short screw through the bottom of the trunk. Since I will lose electrical monitoring access once the connectors are attached, I added another set of wires going down in parallel with the power lines. It terminates with a PowerPole connector at the PEM end, and joins the power feed at the cross-connect splice. That gives me access to monitor the voltage (but not current, sadly) at the motor end, and also provides a back-up way to feed power to the blower should the PEM-end connector fail yet again.



All the wires are spliced together under the car: two going to the pair of new AMP connectors, and three coming down from the PEM (two from the PEM, and one from the PowerPole). Kind of a bigger lump than I had intended, but I'm tired of looking for other ways to connect stuff. The wires are insulated with heat-shrink tubing (the kind with glue on the inside), and another layer of electrical tape. If I had this to do over again, I would stagger the two splices, to keep the overall lump smaller.



The new ends of the cable are then attached to the original motor-side connectors to feed the motor, and secured to the underside of the car with another cable clamp.



All this done, I let the car do its daily top-off charge, while monitoring the voltage from above. It completed the cycle in an identical manner to what it's been doing recently, so the new cable is behaving well.

YAY! Almost done.

 

[gregd]

Final step is kind of a different subject, but part of the overall repair. While I'm in there, I wanted to cover the blower with some sort of screening, to reduce the accumulation of debris inside the motor and PEM. I opted for the simplest approach. Window screen. Wrapped around the blower, and pulled tight at the ends, attached to the anti-sway bar. I know the bar moves a bit, but the anchor is towards the mounting side, and the screen is quite stretchy. It doesn't provide a perfect cover (there's a gap in the back), but hopefully this will keep most of the larger items out. (The ends of the screen were trimmed back a bunch after the photos were taken.)



I'm going to let the car sit for a day or two, to see how this all works, before buttoning the underside back up.

Your thoughts, as always, are welcome!

 

[gregd]

Final post (I hope!) in this saga. All seems fine, so I buttoned things up, gave the car a good loving wash, and took it out for a spin. It performed very well, with the VDS showing the PEM running perhaps (maybe wishful thinking) a little cooler than before. Ambient temp in the low-90's F, and most of the time the graph stayed two bars below Yellow, where it normally ran up against the yellow (though never in it) before. I ran a cable back to the PowerPole connector under the trunk lid, so I could watch the voltage being applied to the blower motor as I drove. It got up to about 11.4v (measured close to the motor) in normally spirited traffic, quickly dropping back a bit to the mid-10's when just cruising along. Glancing at OVMS, it reported PEM temps getting up to about 43. Nice. Pulled the logs, and nothing bad reported. Looks like the peak temp was 44.5C.

The Maintenance Required reminder came on when I first turned on the car, and I reset it. The car hasn't been driven much since last year, due to both the pandemic and it being out of service since early June. I'll figure out what I want to do next Spring, after I see what collects on the screen.

Thanks all for your patience and help in the process of working this issue through to completion, with special thanks to Peter at Gruber.