PEM cleaning now higher temps?

[Rotarypower]

Ok need some help. I undertook the process of cleaning the entire PEM and motor air cooling system in the past couple of weeks. I was noticing increasing PEM temps while under moderate loads and it was obvious to me the system was loaded with debris. Temps are running high here in North Carolina so this was needed.

So, I removed the rear underbelly pan, removed single motor dual fan SPAL unit and blew it out with compressed air. No issues observed. Blew out ducts to PEM and motor, very little debris. Removed PEM. There was standing debris in the plastic duct that connects to the bottom of the PEM. Obviously I cleaned that. Started cleaning underside of PEM. Spent a good 45 minutes at least blowing out each heat sink with fins. Using compressed air, was amazed at the massive amounts of debris including leaves, pine needles, and fine dust coating all the heat sinks inside the PEM. Also removed the plastic cover on one of the heat sinks and again was amazed at the debris build up. Finally loosened the motor duct shroud and was able to slide it over to expose most of the top motor cooling fins. They were completely clogged with debris. Cleaned motor fins with vacuum and compressed air. Checked the blower motor connector on the PEM and no issues, so I didn’t take the approach of replacing female connector/pins.

So put everything together and went for a very slow drive. PEM temps quickly went over 40C and went as high as 46c, last blue bar. I never punched the throttle as I drove extremely conservatively. Ambient outside temp low 80s, at dusk, so no direct solar radiation on the car. Something is wrong.

Other data:
Motor temps stayed very low
Blower motor seems to run fine. It adjusts speeds normally as far as I can tell. Is it running at full RPM, not sure.
I do notice more air flow sound. Note sure if this is due to the thorough cleaning allowing more airflow through the PEM or could it be a sealing issue?
No errors. None, nada.
Assembly of duct to PEM wasn’t too difficult. Was able to line up the 4-5 small studs on side of PEM. And pull the duct towards the LH side of car. Clip snapped into place on the lateral side channel support.
Confirmed duct was still lined up with the studs.

Ideas? Thoughts? I am perplexed.

 

[PV-EV]

Ok need some help. I undertook the process of cleaning the entire PEM and motor air cooling system in the past couple of weeks. I was noticing increasing PEM temps while under moderate loads and it was obvious to me the system was loaded with debris. Temps are running high here in North Carolina so this was needed.

So, I removed the rear underbelly pan, removed single motor dual fan SPAL unit and blew it out with compressed air. No issues observed. Blew out ducts to PEM and motor, very little debris. Removed PEM. There was standing debris in the plastic duct that connects to the bottom of the PEM. Obviously I cleaned that. Started cleaning underside of PEM. Spent a good 45 minutes at least blowing out each heat sink with fins. Using compressed air, was amazed at the massive amounts of debris including leaves, pine needles, and fine dust coating all the heat sinks inside the PEM. Also removed the plastic cover on one of the heat sinks and again was amazed at the debris build up. Finally loosened the motor duct shroud and was able to slide it over to expose most of the top motor cooling fins. They were completely clogged with debris. Cleaned motor fins with vacuum and compressed air. Checked the blower motor connector on the PEM and no issues, so I didn’t take the approach of replacing female connector/pins.

So put everything together and went for a very slow drive. PEM temps quickly went over 40C and went as high as 46c, last blue bar. I never punched the throttle as I drove extremely conservatively. Ambient outside temp low 80s, at dusk, so no direct solar radiation on the car. Something is wrong.

Other data:
Motor temps stayed very low
Blower motor seems to run fine. It adjusts speeds normally as far as I can tell. Is it running at full RPM, not sure.
I do notice more air flow sound. Note sure if this is due to the thorough cleaning allowing more airflow through the PEM or could it be a sealing issue?
No errors. None, nada.
Assembly of duct to PEM wasn’t too difficult. Was able to line up the 4-5 small studs on side of PEM. And pull the duct towards the LH side of car. Clip snapped into place on the lateral side channel support.
Confirmed duct was still lined up with the studs.

Ideas? Thoughts? I am perplexed.

I’ve had trouble lining up the duct to the PEM. Seems like that would be most likely cause of your symptoms. Sounds like you got it right but might be worth another look to be sure it didn’t pop off when you lowered the PEM in place.

 

[Stefan T]

The hose is lose
put it back is possible do from wheel well

 

[X.l.r.8]

Not related but worth checking. 90% of race cars that overheat have no radiator fan shroud and have the biggest fans I have ever seen. As Stephan says, check to make sure there is no leak of air from blower to PEM as that cooling air will take the path of least resistance and be far less inefficient.

 

[ML Auto]

Like Stefan T said, check the hose. Both ends are only secured with a cable tie. Once the fan goes on and builds up a little pressure, the hose tends to pop off. Wrapping the ends with duct tape (the actual aluminum professional duct tape, not the cheap fabric stuff everybody has) and then the cable tie is a better solution.

 

[Rotarypower]

So, been really busy but today I started troubleshooting this issue again. First thing was to get some numbers....set the car up to charge at 240 V and 24 Amps while monitoring the PEM temperature. Seemed to settle in at 42C. Outside ambient temps are pretty hot right now, around 35C (95F). Blower seemed to run fine and seemed to change speed normally. Next I pulled the rear pan off and inspected the SPAL fan. Seemed ok. Then without the rear pan, I started the charge cycle again to see if maybe there was a restriction into the fans. My car does have the snow dam sheet metal box around the blower, and my first thoughts when I saw that design was that it looked to be incredibly restrictive. Basically, there are just a few areas around the top of the box where there are some gaps that allow air into the blower fan inlets. I simply wanted to see if an open back side of the car would improve the cooling. After starting the charge cycle again, the temperature again settled at 42C.

So I had ordered an endoscope on Amazon and now was a good opportunity to try it out. A thorough external inspection under the PEM and down into the area where the flexible hose runs to the blower, I could see both tie straps on either end of the hose with apparently no issues. The flex hose appears to be wrapped around the duct / plenum on either side. Could not see 360 around but everything appears normal with the hose. At this point decided to remove the blowers again. Only issue is that there is maybe 25-30mm of the original seal that has degraded, so that is something I will fix upon assembly. Stuck my hand up the PEM hose and confirmed it was properly attached. Then stuck my endoscope up the belly of the beast, ran it all the way up to the PEM inlet. I don’t see any tears, rips, holes, or gaps with the hose.

So, only other possibility are these....

1. These high temps I am seeing are normal for 35C (~ 95F) ambient temperatures?
2. PEM plastic duct or plenum that attached into the PEM air inlet isn’t on properly. However, see my endoscope image, it all looks good. Also, I was super careful during assembly and had really good tactile feel of the plenum when attaching it to the undeside of the PEM. I have got to think that this isn’t a likely cause but now I am second guessing myself.
3. SPAL fan isn’t working right? So, I blew it out with compressed air, and spun the blower with compressed air. Could this have damaged the internals such that it isn’t putting out the correct airflow? I observed the fans spinning when I took the rear pan off the car....they appeared to be working properly but how would I know? There is no tach on these fans, but I believe the car monitors current. I had no errors. Also, there are two identical connectors. In theory, I could have connected these backwards. But...I believe they are each carrying half the motor max current as this is a 12 Volt motor, and the two connectors appear to be wired parallel to the motor. So not sure even if I did swap the connectors, that it would change anything. Thoughts?

Can anyone else check PEM temps during same ambient temp and at a 240V / 24 Amp charge rate? I monitored the PEM removal during charging after going into the diagnostic menu and the Vitals screen.

Oh, one other thing. With the endoscope in the MOTOR plenum, I see what appears to be a gap, or crack in the rubber or plastic material? Anyone ever noticed this before? Not sure if this is just a seem or parting line, or two piece duct that is warped or deformed over time? Not sure how this is designed as you can’t see any of these details unless the motor is out of the car. Just wondering how serious this is. I have not had any motor temp issues.

 

[gregd]

Is the PEM Fan motor really 12v? There was an earlier thread that said the upgraded fan - the single motor with two blowers attached - was actually a 24v unit, and that the PEM harness had to be rewired to put the two outputs in series. The electrical engineer in me says that you can't do that unless for some reason the two outputs are both floating (not grounded), which is unlikely, but possible. I complained a while back about some alerts I was seeing, and the tech eventually figured out that my PEM wiring harness was "wrong." Whatever he fixed really did help the cooling. He didn't specify what was wrong with it, but I'm thinking the parallel / series thing might have been it. But only if the motor really is 24v.

That doesn't answer the question about what changed for you in the temps, since connectors don't spontaneously rewire themselves. Do verify the connector (the one at the far end of the PEM) for any signs that the contacts have burned. Perhaps a squirt of DeOxit might be in order, if you haven't done that already. Being in parallel, the motor might be running mostly on one set of contacts, which will limit the current. Burned contacts is the most common source of PEM cooling issues from what I've read.

All this said, the PEM temps do tend to run in the upper half of the range pretty much all the time, and that seems normal. A quick look at the logs from my car, it seems to like running the PEM at around 46C during charging. So, I guess after all this, what you're seeing when charging is normal. Hopefully what you've fixed with the cleaning is what happens under load while driving.

 

[Rotarypower]

Hi I can only say that I anticipated this question and did take a picture of the unit. See images attached. No I have not put any deoxidizer on the contacts. At this point I am somewhat afraid of removing the connector again for fear that I will create the infamous connector burnout failure. I was under the impression that the car did have a current monitor for the fan circuit and that this is where the errors come from that have indicated problems with the blower circuit.

 

[gregd]

Ha. Well, ok then. 12v it is; parallel circuits. That's settled. Thanks!

I guess the question now is whether you are having a problem or not. From my logs, 42C at the PEM is quite normal. Either that, or we both have issues.

 

[ML Auto]

So the updated fan assembly is 443 CFM, or 221.5 CFM per port (motor and PEM are cooled together), and the original dual fan setup is 319 CFM each. So actually a reduction of around 100 CFM to cool the PEM. That proves that the update only was to resolve the burned connector issue. The two fan setup draws around 20A each, the single fan about 28A from both connectors, which are now tied together, so 14A per circuit. I think the dual fan setup with better connectors is the way to go.

 

[gregd]

So the updated fan assembly is 443 CFM, or 221.5 CFM per port (motor and PEM are cooled together), and the original dual fan setup is 319 CFM each. So actually a reduction of around 100 CFM to cool the PEM. That proves that the update only was to resolve the burned connector issue. The two fan setup draws around 20A each, the single fan about 28A from both connectors, which are now tied together, so 14A per circuit. I think the dual fan setup with better connectors is the way to go.

You'd think the connector would have been the easier thing to fix, no? And, even with the upgrade, I had my connector contacts burn. Hmmpf.

Another side effect of the upgrade is that conditions where either the PEM or motor is stressed more highly than the other will result in cooling being very unbalanced in the reverse. Strong hill climbing in winter, for example, can result in the motor running warm, with the PEM utterly freezing because the air flow can't be individually regulated.

 

[PV-EV]

You'd think the connector would have been the easier thing to fix, no? And, even with the upgrade, I had my connector contacts burn.

The mating half of the problematic connector is soldered on a PCB inside the PEM. The next size bigger connector has a different pin spacing and physical size. Changing the connector would have required a new PCB in the PEM and a new cutout in the PEM housing along with the mating connector that goes to the fans. I looked into doing this after my connector failed and decided it wasn’t doable. I replaced the original soldered in connector with the same part and opted for regular connector maintenance.

As I recall, the original replacement of the twin fans with the single dual fan was an upgrade for Roadsters in snowy areas to prevent snow accumulations around the fans and included sheet metal that was called a “snow dam”

 

[gregd]

The mating half of the problematic connector is soldered on a PCB inside the PEM. The next size bigger connector has a different pin spacing and physical size. Changing the connector would have required a new PCB in the PEM and a new cutout in the PEM housing along with the mating connector that goes to the fans. I looked into doing this after my connector failed and decided it wasn’t doable. I replaced the original soldered in connector with the same part and opted for regular connector maintenance.

As I recall, the original replacement of the twin fans with the single dual fan was an upgrade for Roadsters in snowy areas to prevent snow accumulations around the fans and included sheet metal that was called a “snow dam”

Ah, ok. Makes sense. Still seems like they're going about it in the wrong way, but I understand the constraints now. My car has the single motor / dual fans setup, but I don't think it has the sheet metal snow dams, as I can see the blowers if I look in behind the rear wheels. I'm the second owner, but the original owner was in the SF Bay Area, a place not known for its snow. I don't know what the circumstances were for doing the upgrade.

You're probably right that living with the original connector is probably the best solution, with an awareness and attention to its weakness. A "MacGyver" approach might be to remove the connector from the PCB and solder a short pigtail to the pads. Replace the connector housing with a big rubber grommet to seal the hole in the PEM case, and place the larger connector at the end of the pigtail, with its mate on the end of the wiring harness. Then hope it all fits in the space below the PEM. If I recall, it just might...

 

[Rotarypower]

So to add to my saga.....today I decided to do my best to seal up the interface between the mounted blower assembly and the receiving ducts for both the PEM and motor. I remember @Roadster doing this and posting an awesome “how to” on one of the other blower cleaning threads last year. My approach was slightly different in that I decided to put the seal (1/4” EPDM highly compressible bulb seal from Home Depot) on the receiving duct side instead of the periphery of the mounting flange of the blower assembly. The image below is of my first failed attempt. Once assembled, I put the car on the charger to force the blower into operation and then I was able to ensure no leaks. I failed miserably. Despite my high attention to detail in making sure the seal was properly attached and outside the oval Inlet periphery, I apparently had a pretty massive short circuit of air that was blowing down under the PEM Inlet. This was utterly perplexing! After some tape measurements of the blower outlets vs. the PEM and motor duct inlets, I was even more confused. It appeared they just don’t match up! Next step to figure out this mystery was to create a cardboard template of the blower assembly mounting flange, of particular importance was the rectangular cutouts on the flange. These are the air outlets from the blower fans. Once I had this template, I could mount it to the Inlet side over the four studs to exactly understand what went wrong. Well, this is quite disappointing. Maybe this has been discussed on these forums? If so I have missed it. [Please see images and you will understand]. Total surprise and unfortunately indicates a grave issue with the design but also confirms the theory that this single motor dual fan setup was a cost reduction only, and done with great haste and severely compromised the airflow into these critical components. As a vehicle engineer who has designed cooling systems, this is unacceptable. I know Tesla went into cost redux mode after the 1.5 cars were sold, I am sure the single motor setup was a decent cost reduction over the previous two motor design. I can only guess that the Inlet ducts were not movable since this involved structural changes, and a compromise solution was used. The other major design flaw is the PEM fan outlet is biased all the way to the inboard side of the receiving duct. It is basically dead headed. The actual Inlet up to the PEM is all the way to the left side of the duct. Arghh! One could argue that they tested the solution and found that the airflow was not too severely compromised over the ideal design. We will never know.

Anyway, based on the template, I was able to fix my trajectory of the seal, and confirmed that I at least fixed the airflow short circuit at this location. Still not sure if I have an issue with PEM temps, I will drive the car some tomorrow. Only other thing to do is pull the PEM up and check that the duct is properly attached. Didn’t have time today for that.

(Oh and enjoy seeing my daughter’s duct tape)!

 

[X.l.r.8]

Ouch, that’s impressively bad. And there’s no way to augment the blower ducts to line up either.

 

[ion_1]

also confirms the theory that this single motor dual fan setup was a cost reduction only

I'm not quite sure about this point. The fans themselves are relatively low cost. The optional upgrade to single motor/dual fan required installation of a new sway bar and the manufacture of a mounting plate which added cost / effort especially on labor side. Granted, the matching of plenums/ducts seem to leave a bit to be desired based on your post, this was a nice analysis. Single motor dual fan units were originally offered as an upgrade to cars which had overheating problems. In the history of my car, the car had PEM heat problems. The single motor, single fan system was replaced 2X and it did not solve the problem. PEM eventually fried, the the single motor dual fan was installed as an upgraded in 2014 with PEM. I drive the car in 100F weather and have never seen the PEM reach yellow with the single motor dual fan. Mine has no snow dam.
I do agree with MLAUTO that based on the CFMs noted above, it doesn't make sense why it would be better unless for some strange reason the air intake into the fans is less restricted...

 

[Rotarypower]

I'm not quite sure about this point.

Yeah, me either. It’s just a theory. But yes I have read numerous posts that the single motor setup improved cooling. I’m not familiar with the original design, admittedly, so agree perhaps it had it’s own issues.

Bottom line, there is a lot to be desired regarding the design here. Unfortunately the dual fan setup can’t just be shifted to improve both sides, as the blowers are too close together. I wonder if SPAL has a different unit with an increased center to center distance between the blowers?

Later edit....nope. Our dual fan blower uses a 217.5mm c to c distance, that looks to be the largest spread in their catalogue.

 

[Mark77a]

Interesting observations, I will make a better seal next time motors off.

HOWEVER ..... I wonder if PEM's are really over heating ?
I've been away so car parked, and no charging for a few days - so all temperatures will have stabilised and yet PEM temp is 8 degC higher than ambient 32degC vs 24degC ambient ( 90degF vs 75degF) .

Maybe the sensor (or VDU) is giving a wrong reading ??

Also FWIW I find the PEM is at its coolest, the faster I drive, it sits at 3-4 bars in traffic but drops to 1 bar at speed.
I put this down to the power transistors having to 'chop less' at higher speeds / outputs.

(OK where I live in UK, the ambient temperature is generally lower normally 20-24deg c in summer with bursts, as now, to 30degC, my Garage is 24degC as above)

 

[gregd]

Perhaps the critical item is the connector, or rather, its limited current rating. With the dual motor dual fan setup, each motor had a separate set of pins. Even if the blower was rated higher in CFM, perhaps the PEM & connector wasn't able to provide all the current needed to actually deliver that CFM through the system. Perhaps the single motor dual fan setup, while rated lower, might have actually delivered more as a result?

 

[gregd]

Interesting observations, I will make a better seal next time motors off.

HOWEVER ..... I wonder if PEM's are really over heating ?
I've been away so car parked, and no charging for a few days - so all temperatures will have stabilised and yet PEM temp is 8 degC higher than ambient 32degC vs 24degC ambient ( 90degF vs 75degF) .

Maybe the sensor (or VDU) is giving a wrong reading ??

Also FWIW I find the PEM is at its coolest, the faster I drive, it sits at 3-4 bars in traffic but drops to 1 bar at speed.
I put this down to the power transistors having to 'chop less' at higher speeds / outputs.

View attachment 314794
(OK where I live in UK, the ambient temperature is generally lower normally 20-24deg c in summer with bursts, as now, to 30degC, my Garage is 24degC as above)

Doubt it's the sensors. The reason the PEM is so cool when you drive is that with the single motor / dual fan setup is that the car is probably needing to cool the motor more than the PEM, but can't control the two sides independently. So the PEM gets more air, because the motor needs more air. I've seen this too. Seems pretty normal.