TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

HPWC IR Pics (more fun!)

Discussion in 'Model S: Battery & Charging' started by wk057, Jun 24, 2014.

  1. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    #1 wk057, Jun 24, 2014
    Last edited: Jun 24, 2014
    So, I self installed my HPWC not long ago, but before I had my second onboard charger so hadn't been able to run it at 80A until recently.

    I used 2 gauge copper for about a 35ft run (8 gauge for ground). 100 amp circuit for 80 amp charging. Used 3/4" steel conduit for the exposed portions and 3/4" plastic flex (90C rated) for the hidden portions.

    Since I heard some reports around the interwebs that the HPWC gets very hot, I hung around with the FLIR cam the first time using it at a full 80A to see how things went. Here are the results of more fun with the FLIR cam:


    FLIR0070.jpg FLIR0071.jpg FLIR0072.jpg
    FLIR0073.jpg FLIR0074.jpg FLIR0075.jpg
    FLIR0077.jpg FLIR0079.jpg FLIR0080.jpg
    FLIR0081.jpg FLIR0082.jpg FLIR0083.jpg
    FLIR0085.jpg FLIR0086.jpg FLIR0087.jpg
    FLIR0088.jpg FLIR0089.jpg FLIR0092.jpg
    FLIR0093.jpg FLIR0094.jpg FLIR0095.jpg
    FLIR0096.jpg FLIR0097.jpg FLIR0098.jpg


    The one pic that shows 182F is where I pulled the top of the HPWC cover off and peeked inside... I was beat from my ~570 mile drive, and, I was satisfied that with these readings that something bad was unlikely, and called it a day (was about 5AM...)

    The next day I installed a TED Pro energy monitoring device on both of my panels, along with some whole panel surge suppressors... then I took a trip about 100 miles east, then back. Returned pretty late and started to recharge. After about 90 minutes I took some more pics...


    FLIR0103.jpg FLIR0104.jpg FLIR0105.jpg
    FLIR0106.jpg FLIR0107.jpg FLIR0108.jpg
    FLIR0109.jpg FLIR0110.jpg FLIR0111.jpg
    FLIR0113.jpg FLIR0115.jpg

    The one of the two panels looks alarming at first, until you realize that the highest temp in that pic is 76F. Cam is just sensitive, which is awesome.

    I thought it was interesting that heat from the two internal chargers sneaked through to the interior a little too.

    Here are the two PDFs from the FLIR software also (they have the real-color images that go with the above):
    HPWC @ 80A.pdf
    HPWC @ 80A after 90 minutes.pdf

    It seems to do fine for the ~4 hours needed for a full charge. It does get pretty hot, but not dangerously hot. The wiring and conduit I used is rated to 90C (194F) and the wiring Tesla used for the HPWC is rated to 105C (221F). I didn't see any temps get near these. The exception being inside the HPWC where the components do seem to get very hot (182F noted). The cable as it exits the HPWC gets hot, but cools within a few feet of it. I'll investigate all of this further next time I have the time.

    See also my Supercharger IR pics thread :)
     
  2. pgiralt

    pgiralt Active Member

    Joined:
    Jun 16, 2013
    Messages:
    1,476
    Location:
    Cary, NC
    Thanks for these as well! Was that 182 on the Fuse inside the HPWC? It's hard to tell what I'm looking at on that one.
     
  3. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    Maybe this will help... looks like the fuse and the bus bars around it (pic of inside of HPWC shamelessly copied from somewhere else...)

    FLIR0086-help.jpg

    Unfortunately I didn't take more. I'll remove the cover fully next time and take some pics that way.
     
  4. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    Digging up my own thread, I had some time to FLIR the HPWC again a bit last night after a ~150 mile drive about 2 hours into charging, and I'm admittedly a little alarmed... enough that I will be contacting Tesla about it probably just to be safe.

    I went ahead and popped the cover off of the front of the HPWC during the charge and snapped some FLIR pics. The cover was barely touchable it was so hot. FLIR doesn't pick it up since it is reflective, however.

    The contactor at the top was over 250F. The wiring heading towards the external wiring to the car was nearly as hot. There is also a resistor near the contactor's power output from the control board that was also ~250F.

    The external cord itself was within tolerance sitting at 125F average (221F rated), but nearest the HPWC (outside) it was around 150F... not really safe to touch, but still within tolerance (externally at least).

    The charger handle was about 130F with the connector itself a bit hotter... pretty hot to the touch, but not hurtful.

    My 2 gauge copper input wiring seemed to be sapping a bit of heat from inside the unit it, heating the entire exposed conduit to about 110F.

    Ambient in all this is about 80F.

    2 gauge wiring near the breaker panel was only about 90F. Only a 4V drop from panel to the car dash readout (244V to 240V) at 80A charging. About 5W per foot of wire including the charging side cable, which is probably acceptable.

    What concerns me is the 250F contactor and nearby wiring conducting some of that heat. I could not find a temperature rating on that wiring, but I doubt it is more than 105C (221F), so I'd be worried that the insulation on this wiring would fail. I'm also concerned because in general I don't know why the contactor would get so hot if it were operating within designed limits.

    Going to download the pics from it and post shortly.

    - - - Updated - - -

    Here is the FLIR report for a few of the pics. HPWC Update
    The report has the real color images and a little data to make some sense of it.

    Here are some of the pics:
    FLIR0142.jpg FLIR0143.jpg
    FLIR0149.jpg FLIR0150.jpg
     
  5. pgiralt

    pgiralt Active Member

    Joined:
    Jun 16, 2013
    Messages:
    1,476
    Location:
    Cary, NC
    Let us know what Tesla says about this. The contactor reading does seem pretty high. Your measurements on the outside are roughly the same as mine after I got the new cord. The hottest part is definitely where the cable exits the HPWC.
     
  6. rdrcrmatt

    rdrcrmatt Member

    Joined:
    Jun 27, 2013
    Messages:
    551
    Location:
    Milwaukee
    DAMNIT!! I have a FLIR ONE on order and was going to do this with it when it came in!


    NICE POST!
     
  7. pgiralt

    pgiralt Active Member

    Joined:
    Jun 16, 2013
    Messages:
    1,476
    Location:
    Cary, NC

    Well, since you don't need it anymore, you can just send it to me :)
     
  8. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    So, I hadn't even had a chance to email Tesla yet about this.

    I just got a call from the service center, "We saw your post about the HPWC and wanted to follow up."

    Long story short I was told that the temps I was seeing inside the HPWC were a bit high and they're sending a replacement unit.

    Pretty awesome support if you ask me.
     
  9. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    And < 17 hours later I have a brand new HPWC here to replace this one with, and a paid return label for the old one. Tesla, you guys are amazing...

    2014-08-06 10.09.16.jpg

    - - - Updated - - -

    I'll work on getting this swapped out today probably. Tomorrow I'll be driving ~100 miles again, so, can test temps again then hopefully.
     
  10. bonnie

    bonnie Oil is for sissies.

    Joined:
    Feb 6, 2011
    Messages:
    14,241
    Location:
    Columbia River Gorge
    That's the kind of proactive service I've always experienced from Tesla. Glad they took care of you so quickly! They're an amazing company.
     
  11. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    So, looking at this new unit and comparing to the existing one, they seem pretty much identical physically. Same contactor.

    However, I got a look at the contactor label on the new one...

    2014-08-06 10.42.51.jpg

    A little digging leads me to this datasheet: http://curtisinstruments.com/index.cfm?fuseaction=cProducts.DownloadPDF&file=50157%5FSW80%5FAlbrightRevA%2Epdf

    This appears to be a 96V DC 100A rated contactor... with a 240VDC coil voltage.

    Now, I believe that this is *probably* fine for the load, 80A @ 240VAC. DC contactors are generally designed with larger air gaps, silver coated contacts, etc in an effort to prevent arcing in a circuit that has constant voltage. (In an AC circuit the voltage passes through 0 volts many times per second so this is less of an issue.) The problem is that I don't see this rating anywhere in the datasheet for AC... which could have consequences for some certifications and such since it doesn't appear to be, on paper, rated for the load. It should be able to easily handle it, however, at 240VAC. It just concerns me that it is not rated for this (from what I can tell).

    A concern from the datasheet: "If mounted vertically the contact studs must point upward." The HPWC does the opposite of this. :( The consequences, I would guess, would be less pressure on the contacts, resulting in higher resistance, which could account for the increased heat I was seeing prior. The coil even on the least efficient configuration of this contactors dissipates 20W, which wouldn't account for the temps I was seeing alone. The contactor is also rated for only a 50mV drop across the contacts... so 4W of heat at 80A.

    I will still install and test the replacement unit before contacting Tesla and inquiring about a potential design error...

    (Not sure if I should have made this a new thread...)
     
  12. banterer

    banterer Member

    Joined:
    Dec 26, 2012
    Messages:
    29
    Location:
    Los Angeles
    Hi there,

    I thought I'd chime in here. Last year I reported to Tesla that the handle on my HPWC was at 136º and that I thought it was a little warm. They did some checking and came back and told me that it was "within acceptable limits". I let it go because it didn't really burn me, it just concerned me.

    Last week I interrupted the charging when it was going full throttle 79/80 amps and the car was at a rated range of 211 miles. This time it was really hot so I got my laser thermometer and took some readings—145-158º or so depending on where I shone the little red dot. I called again and this time they said they would get in touch with the local SC. They called me back but nothing was really resolved. I called back a few days later and spoke to my regular service advisor. She told me to send her a photo and I said I would. Holding the thermometer and the camera was not easy since I still need another finger to actually shoot a picture so I did something better, I shot video. The temps were all varied but confirmed the 145-158 range. I sent her the video and she sent the video to engineering. Engineering then told her to have me take the car in for a check up because maybe there are some loose resistors in the charge port. 'That made no sense to me since it wasn't the car nor the port that was hot, it was the handle and the cable.

    They got the car today and told me that they took the charge port apart and that everything checks out (their handle got up to 98º) and that I should get in touch with the contractor who installed it. Oh, you mean Solar City? The contractor that Tesla recommends? Ok sure. They are thinking that it may be a problem with the install. I voiced my disagreement with that analysis but I'll go along with it. Who knows? Maybe. I also took the opportunity to tell her that I found her mentioning of the charger only having a "one year warranty" to be ridiculous when I purchased the charger at the same time as the car and purchased all the extended warranties they had to offer. How can you sell me a rechargeable item and its charger and warranty the item for 8 yrs and the charger for one? Especially since I am on record as having complained about the hot handle last year sometime?

    After we hung up, I decided to check the forum to see if anyone else has reported problems with heat and I found this and another thread. (Oh when I called Solar City, he told me that he had just gotten off the phone with a regional mgr up in Portland who had just fielded a call from someone else reporting the same thing who was referred to Solar City. He too found it odd that although they are sister companies Tesla appears to be treating them like a red-headed step-child.) After reading through many of the posts in this thread, I called my advisor back and let her know that this problem was NOT new and that many others have been experiencing problems, some have even had two cable replacements already. She was not surprised, she just told me that they have to follow procedure and that she would champion my cause to at the very least help me if I have to purchase another HPWC. I let her know in no uncertain terms that I did not intend on "buying" another anything.

    A few minutes later, as I continued to read the posts on the subject, she called me with good news. She went in and spoke to her manager and somehow convinced him to get me a new charger and have another electrician install it. I was grateful but can't help wondering why Tesla isn't taking looking at this as a potential design flaw.

    Jorge
     
  13. bonnie

    bonnie Oil is for sissies.

    Joined:
    Feb 6, 2011
    Messages:
    14,241
    Location:
    Columbia River Gorge
    Did you ever get the install checked as recommended? Or is the assumption it must be the HPC and they were just trying to not deal with it (which hasn't been my experience)?

    I was talking to someone earlier today about this and there was concern that it wasn't actually the HPC, but rather an improperly installed breaker causing the HPC to display these issues. Just because Tesla and Solar City share a board member, doesn't make them sister companies (that would require the same parent corporation). It sounds like the SolarCity person you talked to managed to make this Tesla's problem without actually checking their own work. (Nice deflection on their part, btw.) Might be Tesla, but you don't actually know that based on what is posted here. Might be SolarCity, too.
     
  14. banterer

    banterer Member

    Joined:
    Dec 26, 2012
    Messages:
    29
    Location:
    Los Angeles
    I called them (immediately) to set up an inspection. He took my number and told me he would check with his manager to see what he could do for me. I am still waiting to hear back from them but Tesla called me and told me that they are going to replace it with another electrician of their choosing even though I told them that I called Solar City to look at it.

    They (Tesla) said they'd call me tomorrow to set it up.
     
  15. GabrielB

    GabrielB MS-P85, Vette GS-Conv ;(

    Joined:
    Jun 16, 2014
    Messages:
    79
    Location:
    Bay Area, CA
    wk057,
    Awesome pictures, thank you big time for your research here.
    The Curtis/Albright SW82 Contactor has a suffix of -312. Per their web site FAQ:
    What is the "end part number" on each Albright DC contactor label called?
    This "suffix" number is called a "precise identity number" which identifies that specific Albright DC contactor number with a specific customer and contactor special features.
    So, I would think that this identifies Tesla as a unique customer, and the "upside down" configuration with compensated contactors, as the data sheet describes as an option. The -312 cannot be decoded from the data sheet, but the FAQ describes it as a unique proprietary custom item, presumably per customer purchase contract.

    HPWC P/N Rev:
    -A and –B at the end of the P/N on the cardboard Box indicates the unit was built with the smaller internal 100Amp slow blow fuses. Rev –C and above have a physically 2x larger 200Amp? fuse. The larger fuses run cooler and have less propensity to fail.
    Tesla has had a service campaign to replace the fuses on Rev –A and –B units. If you are in doubt, check with your service manager. I think Rev –C came out about July of 2013, (remember the temporary software charge limit of 60Amps?) so only the earlier HPWC units were involved.
    Why is this important? The smaller fusses could have been dumping heat into the Contactor terminals, (see Photo FLIR0086.jpg 6/20/2014 10:43:20AM) making the Contactor run even hotter. However, your last set of FLIR photos show the larger fuses and the contactor as hotter than the new larger fuses. A dramatic improvement.

    2AWG In & 6AWG Out:
    I can see in Photos FLIR0149.jpg 8/5/2014 5:37:12AM & FLIR0150 5:37:23AM that the contactor is still hot, however there is a dramatic demonstration of how hot the #6AWG rubber cord set side is compared to the #2AWG input side through the new larger fuses with thicker brackets. Mainly, the terminal lug connection to the output side wires. In fact, the jumper wire that goes from the contactor through the ferrite coil and to the terminal block to the 6AWG rubber cord, looks like it is #4AWG.
    So, the limiter now is the #6AWG rubber cord set.

    Metal Conduit Vs. Romex:
    I also wired my system with 2AWG copper wire, and I also noted that the 6AWG Rubber cord at the HPWC exit and at the Connector Handle were the hottest areas per my un-calibrated hands.
    A friend of mine installed his 2AWG wiring using metal Conduit. I used Romex in the wall without conduit. I still need to close the sheetrock, so the Romex is still exposed for inspection. I noticed that my friend’s 1” metal conduit entering the HPWC was very very hot to the touch. There was also 1-1/4” conduit used to reach a junction box close by. This conduit was slightly cooler. My 2AWG Romex entering the HPWC was barely warm. In both cases I was charging my empty Model S at 80Amps for a good hour or more. Both of our HPWCs have the new larger fuses.
    So I am wondering how much inductive heating is being added by the small diameter metal conduit? I do have some metal conduit , 5 feet of 1-1/2”, where the Romex (stripped jacket) exits the attic and goes to the main electrical panel. During the same charge session at night, it was cool to the touch. Bigger conduit is better.

    In closing, if I could, I’d like to replace my 25-foot #6AWG rubber cord set with a 10-foot #4AWG rubber cord set. Meanwhile, since I have plenty of time overnight, I’ve reduced my charge current from 80 to 60Amps. Lower currents and lower temperatures result in longer component life.

    If possible, I’d love to see more detailed FLIR’s showing the components and jumper wires and terminal lugs with the cover off. (CAREFULL). And the 6AWG rubber cord set from 12 inches outside to the inside terminal blocks. Also, a closer look at that 3k Ohm 5% power resistor.

    Cool work on a hot topic! Thanks!
     
  16. bonnie

    bonnie Oil is for sissies.

    Joined:
    Feb 6, 2011
    Messages:
    14,241
    Location:
    Columbia River Gorge
    Smart on Tesla's part. People don't like to say their own work was bad. If there is a problem with the wiring, they'll have to identify specifically what the problem is and then you'll have your answer. SolarCity is a good company (I own stock there, too), but companies are made up of people. And people make mistakes sometimes. Happens. Glad everyone is getting involved to figure out what went wrong for you.
     
  17. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    Tested out the replacement HPWC with pretty close results. Oddly enough the handle at the car was a bit cooler than my previous one, so, I'll go with it.

    So, I wanted to get some good measurements of voltages at multiple points along the way. At 80A charging this is what I saw (this was in the middle of the day, no ToU metering at this location... voltages are easily 10V higher at night):

    • 234V @ Model S Dash readout
    • 236.0V @ junction inside HPWC heading towards car
    • 236.4V @ 2AWG copper input to HPWC
    • 238.2V @ Panel

    This comes to about 4.2V @ 80A of resistive heating loss, or about 336 watts. Not negligible, but, my 2AWG run is roughly 30ft. The HPWC cable is 25ft, so, about 6.1 watts per foot average heat generation in the wiring. ~6.4W/ft in the HPWC->Car cable, ~5.76W/ft in my 2AWG cable... so, nearly the same resistive losses, which explains the warm conduit. No where near any unsafe temp, however.

    336W may not seem like much, but it does add up in an insulated cable over time. At these currents, though, it is expected there will be some loss to heat unless we're talking about some ridiculous gauge wiring.

    So, here is a baseline, a couple of minutes into charging @ 80A:

    FLIR0175.jpg FLIR0176.jpg FLIR0177.jpg FLIR0178.jpg

    I checked back 30 minutes later:
    FLIR0179.jpg FLIR0180.jpg FLIR0181.jpg FLIR0182.jpg FLIR0183.jpg FLIR0184.jpg FLIR0185.jpg FLIR0186.jpg FLIR0187.jpg

    And about 30 minutes later again:
    FLIR0188.jpg FLIR0189.jpg FLIR0190.jpg FLIR0191.jpg FLIR0192.jpg FLIR0193.jpg FLIR0194.jpg FLIR0195.jpg

    -----

    So, some notes.

    This time I left the protective plastic on the cover for the HPWC... I figure this would help with the FLIR cam a bit and not just register reflections. Seems to have worked, so, pics here with the cover on have the film on.
    I also kept the cover installed during charging. I removed it each time I was taking pictures. Also, this is the brand new replacement HPWC that Tesla service mailed to me. I started the switch last night and finished the install this morning.

    Next... that resistor scares me. Was almost 300F... and was already 275F at the start of the charge. It's just a resistor, but seems like it's under a pretty decent load for whatever reason...

    There is some wiring inside that is well above 105C... not particularly safe for continuous use. And these pics are only after about an hour of charging... I hadn't charged for a couple of days since I received the replacement HPWC, but hadn't driven much... just about 100 miles.

    In summary, my thoughts:
    • I think resistive losses in the home wiring is acceptable.
    • I think resistive losses in the HPWC->Car wiring is pretty close to acceptable, but probably would be an issue if a HPWC was under continuous use (for example, charging two Model S one after another...)
    • To make the handle and external wiring cooler would require heavier (gauge and weight) wiring
      • The wire is already pretty heavy, weight wise...
    • The internals probably should be a bit beefier.
      • At a minimum I'd say 2AWG wiring from the contactor to the junction to the charge cable.
        • Have this spot sink the resistive heat that will linger here somewhere...
      • The contactor itself is possibly underrated for this and might benefit from being a notch more heavy duty
      • Perhaps a small cooling fan?
        • Guess this would prevent weatherproofing, so maybe a design that includes some fins for a heat sink
    • It will suffice for now.

    I doubt we're going to see an overhaul anytime soon, so, it will suffice. I'm not afraid of it bursting into flames or anything, but, I am going to keep an eye on it.
    Long term, since the warranty on the HPWC is only 1 year, I'll probably gut it and make an openEVSE type control board to put in there, a heavier contactor, and some heavier internal wiring. Also probably shorten the external cable slightly to fit my needs. Probably a PITA all together, but would make me feel better long term. Besides, can do plenty of cool stuff when you can program the controller yourself.

    Long term I'd like to see Tesla completely redesign this thing and I'd buy an upgraded one in a heartbeat.
     
  18. qwk

    qwk Model S P2681

    Joined:
    Dec 19, 2008
    Messages:
    2,817
    It's no secret that Tesla went with the "bare-minimum"design philosophy on both the HPWC, and the UMC, obviously for cost reasons.
    IMHO, this was a horrible idea, because buyers of $100k cars would rather pay a bit more for a robust charging solution, rather than constantly having problems.
     
  19. Johan

    Johan Took a TSLA bear test. Came back negative.

    Joined:
    Feb 9, 2012
    Messages:
    6,890
    Location:
    Drammen, Norway
    #19 Johan, Aug 8, 2014
    Last edited: Aug 8, 2014
    Very interesting thread. In EU they didn't even try to make a HPWC (for >10 kW 3-phase AC charging) but left it to 3rd party OEMs. Smart move perhaps, since the Mennekes type 2 standard was already in place.

    Also the US version of 20/22 kW AC charging is at 240V one phase (80 A) while our's is at 400 V 3-phase (32A). With regards to heat I'd guess the EU solution is superior.
     
  20. wk057

    wk057 Senior Tinkerer

    Joined:
    Feb 23, 2014
    Messages:
    4,723
    Location:
    Hickory, NC, USA
    Yeah, there are very few things that would be in a household that would pull 80A @ 240V... if anything. So a 20kW load is a bit of a non-standard thing, IMO, for a residence at least.
     

Share This Page