13 MS 85 can't go into drive, recent LDU rebuild, some clues from clicking sounds

[howardc64]

73k miles. Won't go into drive with a bunch of error codes (looks like HV loop disabled). LDU rebuild completed ~ 2weeks ago and driven 350 miles before this problem.

Error codes shows repeated attempts to connect HV (presumably to charge 12V and heat battery?) until 12V got too drained and wouldn't power up the screens. Connected battery charger to jump points behind nose cone and fired up the screen (then got the error codes)

LDU rebuild completed 1+ week ago (leaked rotor seal). Added 2-3L of coolant on reinstall, reservoir puked some out (didn't like being at max) and dumped on top of battery edge seam before adding drain hose to reroute.

Can hear some clicking where the contactors are but not the usual loud contactor sound. Much fainter. My understanding is there is a pre-charge relay that fires and wait for inverter ack before firing the big contactors. Perhaps its the pre-charge relay I hear? I think also hear multiple clicks in sequence (maybe retry?)

Charge port door opens, no light, can't plug in.

Many posts on battery heater failure causing a short on the HV loop preventing from going into drive (besides the obvious suspicion on the contactors or HV battery water ingress) Will try to disconnect HV parts like battery heater and AC compressor to see if culprit. Will pre-charge relay fire if this was the case? Will also inspect inverter underside to see if any coolant presence (none for a few days after LDU reinstall+drive cycle before belly aero pan was installed)

scanmytesla seems to get to all modules including inverter, battery heater temp sensors. So CANBUS looks fine. Battery voltage and BMS all seems to report healthy status.

Does the lighter than usual loud contactor click from the same area provide any clues? Does pre-charge relay even fire if there is a short in HV loop?

Tough one to chase down. No cost effective diag tools is a big barrier for DIYers operating in the blind. Like to avoid Tesla SC but may not be possible on this one (always a frustrating experience dealing with Tesla app service chat and for out of warranty cars)

Here are the pics

11/26 (sat) & 11/27 (sun) errors. 11/26 1:38pm is plugging in charger at home after last drive cycle. First attempt to plug mobile charger failed, 2nd attempt succeeded. 11/27 morning error starts ~4am. Looks like repeated attempt to connect HV for 12v bat recharge and/or battery thermal mgmt?

Both pic shows about same error list. First (blurry pic) include the 11/26 1:38pm charger plug error.



Error list and scanmytesla readings after powering up the car this evening (12v charged up and connected, HV fireman loop connect, 12v charger on post behind nose cone) Car powers up, step on brake will turn on head lights, drive info screen shows typical cold battery regen limit.

 

[howardc64]

Here are the contactor clicks. Activated it by triggering charger door open. Time stamp of contactor clicks and other sounds noted in the comments

Confirmed the sound is not from contactors inside HV Junction Box (gen1) under rear seat (didn't expect it to be, just confirming)



Bottom of LDU all dry, no leaks. Inverter connection & cover both dry. When coolant is in inverter housing, likely present at these connections (was present during LDU rotor seal leak rebuild. link to post + pic )



I believed confirmed HVIL is all good (Should be since HV battery contactor made clicking noises) Disconnected coolant heater from DCDC converter (gen1) to cause a break in HVIL and reconnect 12v battery + first responder loop. No HV battery contactor clicks. Expected since opening the HVIL circuit at the coolant heater connector is probably same as disconnecting first responders loop.

====

Question for the experts...

Reasonably to conclude HV battery is good?

- scanmytesla shows all modules and total pack voltage at expected values
- contactors/precharge tried to close

Likely fault on the HV circuit?

Okay to jumper HVIL on removed coolant heater, cabin heater, AC compressor connector to run the car to check if they are the HV circuit short source? (no Megohmmeter) I believe these don't have 60 Ohm resistor drops according to schematic ( link )

 

[howardc64]

Well, thats interesting. After removing ad plugging back in the coolant heater plug to cause a break to confirm HVIL circuit function. Contactors fires and car all back to normal after 12v+1st responder reconnect...

I have the updated yellow dots on all DCDC connectors per this TSB ( link ) where faulty connector prevented good contacts.



Replace coolant heater (seems to be high failure part) as precaution? What are failure rates on the gen1 DCDC converters?

For future reference, here is a sequence of 5 videos by a guy that replaced cabin heater and DCDC converter fuses for gen1 systems

1of5: Tesla Model S DC-DC Fuse+PTC Heater replacement (General overview, access and disassembly) - YouTube

 

[howardc64]

Here is a theory on what happened and maybe an explanation of high failure rate on the coolant heater as well as the 2 other HV devices on the DCDC converter (AC compressor/cabin heater) blowing fuses for no good reason. Note this is just a theory based on observations. Don't have the statistics that Tesla is not putting out. Here is the big long thread on AC compressor blowing gen1 DCDC converter fuses ( link )

HV line on pressed on connector

All 3 devices on gen1 DCDC converter is pressed on connector for HV supply. This alone is probably not the issue as my Prius AC compressors's HV supply is also a connector. But do note other higher current flow HV connections are bolted.

DCDC converter lives in high ambient moisture area

Gen1 is located behind passenger fender liner where water sprays directly on to. Liner is carpeted instead of plastic so its a pretty good moisture sponge. DCDC converter doesn't get direct moisture but this is where it and the 3 connectors live. Prius AC compressor HV connector lives near the ICE so got a nice heater to dry the environment. Also don't know the fuse margin difference between Prius and Tesla.

Connectors likely build up resistance over time

Given the above 2 scenarios, its likely connectors built up resistance over time. If the resultant voltage drop is above Tesla's HV short detection cutoff, then we get exactly what I observed. Precharge fire but fully HV contactors do not.

The resulting voltage drop and higher required current on cabin heater and AC compressor probably results in the blown fuse in DCDC convertor.

Coolant heater's ground is PWM

This HV source is switched for PWM. I'm not EE but perhaps more sensitive to resistance buildup? Hence the statistical higher "failure rate"?

Car was disconnected from HV for 2.5 months

During LDU rebuild, perhaps this accelerated the resistance build up? I live in rainy seattle although car parks in the garage. But generally more dampness. Curious if this issue is much less in the dryer regions like AZ.

Rare to hear precharge

I think the normal delay between precharge and contactor firing to supply full current is tiny and can't be heard. However, my inverter was disconnected from HV for 3 days and its cap would have been fairly discharged. After dis/reconnecting coolant heater connector followed by 12v+1st responder connect, heard light precharge click and a few seconds later the big pop of the main contactor. Presumably after inverter capacitor had charged up sufficiently to handle the big current flow.

I think if you listen carefully, sometimes contactors will be 2 rapid clicks. First light followed by a louder pop. (occurs often on door handle present) But it all depends on the state of inverter cap discharge. Teslas inverters will see HV often as car fires up HV frequently on its down for basic duties such as 12v charge up.

Theory of what happened in my "failure"

So what happened in my case is probably just scrapping for fresh metal contact by dis/reconnecting. Perhaps coolant heaters aren't really failing at these high rates and just the HV connector contacts?

Solution going forward?

Anyhow, this is just a theory but given gen1 DCDC converter location and pressed HV contacts for the 3 devices. I'll be pulling these connectors annually to spray contact cleaners on them going forward.

 

[howardc64]

More info for future DIYer

Battery Contactor Behavior

This BMS contactor activation sequence video probably explains my clicking sounds in post #2.

1.3.6- How to control contactors with a BMS - YouTube

My clicking sound is probably the following

- Ground contactor close (1 click)
- Precharge contactor close, test and fail, then open and a few retries (sequence of 2 clicks 8-9s apart)
- Ground contactor open (1 click) Car gives up.

Don't know why its so much louder when its successful.

Tesla focuses on HV and 12V rail isolation (Safety first)

Certainly make sense for safety. Don't want HV into rest of the car. Had a coolant heater replaced under warranty in 2017 due to intermittent isolation fault. Car was in service for other reasons. No isolation warning on car, no drivability issue, therefore presumably intermittent failure of only1 of the +- HV lines. So likely the car is monitoring for isolation faults. Only 1 of the 2 HV +- lines with low isolation doesn't hurt anything. But of course then you are half way to both lines in low isolation and shorted HV to the rest of the car. Here is a good explanation on HV isolation.

Megohmmeter - Loss of Isolation Diagnostics - YouTube

Non Drivability Component Failure Can Strand The Car

Since AC compressor, coolant heater, and cabin heater are all on HV, some of their failure modes can disable HV (and therefore the drive unit) and leave you stranded. Ideally have some simply bypass. Posted thread here to seek solution.

 

[brainhouston]

one thing i would add, is I usually connect Fireman's loop then 12v.
I once forgot n got a slew of HV/unable to drive errors n nothing about "hey fireman's loop is open" error...

also, probably unrelated here but sometimes u get HV errors after pre-heat, they do go away on its own tho

AP1 ONLY Please -- life after 2018.50.6

from my experience with .28 the alternate routes only show briefly if you're already in motion then default to one of them. if you navigate while stopped you'll see the alternate routes to choose from.

teslamotorsclub.com

 

[howardc64]

one thing i would add, is I usually connect Fireman's loop then 12v.
I once forgot n got a slew of HV/unable to drive errors n nothing about "hey fireman's loop is open" error...

This is interesting indeed. Service manual says disconnect is 12v->HV. Reconnect is reverse (which would be HV->12v but I've always wonder if thats a precise direction or "reverse" is loosely interpreted) but it just seems safer to me to do 12v->HV on reconnect. Glad to know you do HV->12V. I'll try that next time.

also, probably unrelated here but sometimes u get HV errors after pre-heat, they do go away on its own tho

AP1 ONLY Please -- life after 2018.50.6

from my experience with .28 the alternate routes only show briefly if you're already in motion then default to one of them. if you navigate while stopped you'll see the alternate routes to choose from.

teslamotorsclub.com

Thx. yeah I've read HV errors that goes away on its own (put in D, overnight etc) But mine stays for 2.5 days with countless disconnect/reconnect efforts. Only worked after I unplugged coolant heater, connect 12v->HV, no contactor noise, disconnect HV->12V (again just thought it was naturally safer but now double checked the service manual is opposite) plugged in coolant heater, connect 12v->HV and bam... big contactor fires after inverter cap charged up for a few seconds after 2.5days of no HV precharge. I distinctively remember hearing lighter single click, few seconds, typical loud contactor connect! and JOY haha.

 

[cleverscreenam]

Probably not related, but I've had a few instances of my S continuously clicking contactors while trying to start an AC charge. I remedied the problem by opening a door, causing the MCU to bootup, then it started charging after the large T logo switched to the UI. (Climate was turned off the whole time.)

 

[howardc64]

Probably not related, but I've had a few instances of my S continuously clicking contactors while trying to start an AC charge. I remedied the problem by opening a door, causing the MCU to bootup, then it started charging after the large T logo switched to the UI. (Climate was turned off the whole time.)

Haha yeah, gremlins. Are you gen1 DCDC?

Tesla's water drain path from windshield above DCDC seems high risk to DCDC's HV connectors. Regardless if routing path is perfect, not a design challenge risk to take on anyway. Gen2 DCDC relocated away to much safer area.

Will see if can fashion something to guide water away from the plugs and do annual plug pull and contact clean. If moisture gets into the seams between DCDC metal chassis and connector housing and there is insufficient seal in that gap, then cleaning is not going to help.



Here is a 12V battery replacement video showing water run off guide path in relation to gen1 DCDC and its HV connectors (lower left corner)

 

[cleverscreenam]

Yeah I'm 2014. Well snap, glad I read this because I cut a hole in that plastic piece to feed a 12v disconnect up under the wiper for me to easily (without opening the frunk) unplug every night to eliminate vampire drain. So now I have to fashion something to keep water away from the DCDC.

Can I take a minute to rant about how fragile Tesla's wheel-wells are? They should be armored and reinforced to take a beating from standing water, snow, ice, dirt, and road debris. Instead we have these pathetic flimsy fabric and plastic pieces that's all that stands between flying debris and sensitive electronics and vital coolant hoses. Mine have ripped out 4 times now, I've had to get creative with drilling new mounting points to replace the old plastic holes. $100,000 car held together with zip-ties, hopes and dreams, and now the tires rub the plastic a little while turning.

 

[howardc64]

Yeah I'm 2014. Well snap, glad I read this because I cut a hole in that plastic piece to feed a 12v disconnect up under the wiper for me to easily (without opening the frunk) unplug every night to eliminate vampire drain. So now I have to fashion something to keep water away from the DCDC.

Can I take a minute to rant about how fragile Tesla's wheel-wells are? They should be armored and reinforced to take a beating from standing water, snow, ice, dirt, and road debris. Instead we have these pathetic flimsy fabric and plastic pieces that's all that stands between flying debris and sensitive electronics and vital coolant hoses. Mine have ripped out 4 times now, I've had to get creative with drilling new mounting points to replace the old plastic holes. $100,000 car held together with zip-ties, hopes and dreams, and now the tires rub the plastic a little while turning.

Yeah that liner is pretty weak and mostly held together by push pins haha. I can see it taking a beating in east coast harsh winter potholed roads.

My observation is Tesla's aggressive vertical integration DNA (huge cost benefit of course, they do this in sourcing AND PR (Elon one man band show)) resulted in skipping reputable brand supplier's QC often. Quite amazing how many parts shows up from China no brand names on this car. German cars at least goes through reputable brands like Bosch and Continental to QC their Chinese manufacturing source (and pay higher margins of course). This car seems to have lots of parts skipping this. So aggressive custom design + direct sourcing from low cost source + own QC + silicon valley employee turn over probably got us wheel liner, MCU1 touch screen leaky goo, gen1 TPMS, etc. I heard the LDU rotor was sourced out of China with 20% failure rates at factory... Probably make sense for a huge chunk of copper material but there are higher tech stuff like shaft hardness treatment etc..

 

[vcor]

...My clicking sound is probably the following

- Ground contactor close (1 click)
- Precharge contactor close, test and fail, then open and a few retries (sequence of 2 clicks 8-9s apart)
- Ground contactor open (1 click) Car gives up.

I'm about 99.9% sure the HV is never connected to vehicle ground. The HV is always isolated and allows the system to detect leakage faults to the ground (i.e. water in the battery for example). I think you meant to say the contactor on the Negative HV bus. Thanks for the other details and links.

 

[howardc64]

I'm about 99.9% sure the HV is never connected to vehicle ground. The HV is always isolated and allows the system to detect leakage faults to the ground (i.e. water in the battery for example). I think you meant to say the contactor on the Negative HV bus. Thanks for the other details and links.

You are correct, I meant HV negative contactor close. Not the 12V ground. Should have been more clear when I say "ground" given the 2 completely different circuits haha.

Anyhow, I believe all contactors including precharge on are the HV rail fpr main connetion input/output. Only their solenoid controls are on 12v rail.

 

[howardc64]

BTW, here is my DCDC in high rez, check out the faint water drop marks all over.



I think I might not have tucked under the runoff channel properly like this



and water pooled and instead of dumping into the wheel well, went backwards dumping over DCDC. Definitely an easy mistake when getting to 12V and 1st responders loop.

The under flap (part of cover for air intake that needs to pull up to get to towards 12V battery) can be easily pulled out (held by 1 push pin) without pulling the top flap. Top flap is secured by 2x tight XMAS trees that are pretty hard to remove. They weren't previously remove just before taking the above picture so I definitely had the flap order wrong.

And I guess the PWM HV - on the coolant heater might just be a lot more sensitive to corrosion than the continuous HV - on AC compressor and cabin heater.

 

[howardc64]

Does anyone know if the windshield runoff water that routes behind the battery and behind the gen1 DCDC eventually dumps on the battery pack close to the edges? See the run off exit on the top middle of this video

Obviously if so... bad news and need to find a reroute solution.

 

[howardc64]

Looks like windshield runoff doesn't dump on top of the battery pack but don't know if there are some path that leads it towards sitting on a battery pack rivot/screw/seam. Man, what a great place to dump with HV and other wiring in the area.

Seems desirable to add a pipe to forward the runoff to clear the bottom of the car.

Just pics found online. Only added the green arrow on first pic to show where dumped runoff comes down.

 

[howardc64]

20min drive in light drizzle rain, wipers on. Everything was buttoned up properly during the drive.

• passenger wiper pools water in corner of windshield. Water run off guide at the very edge has no sealant. Just moulded fit, no way to keep back water running unguided below the hood.
• cabin airbox dry, run off dry
• windshield run off to wheel well path working
• water drops on DCDC, firewall behind, and connectors! Need to confirm not from hood opening / flap opening...

So found a new path for unguided water run off and need to confirm how water is getting to DCDC, firewall behind, and HV connectors. Perhaps need to dry off hood so it has no run off when opening.

Seems likely root cause of so many HV issues in this area and probably why Tesla relocated it to dryer location in Gen2 DCDC.

 

[cleverscreenam]

I just stuffed the whole area from the 12v to the air filter with a large tire plastic bag

 

[howardc64]

I just stuffed the whole area from the 12v to the air filter with a large tire plastic bag

haha, good idea, been thinking what to do (put a plastic folder over DCDC+connectors?) Been thinking about thermal dissipation but DCDC has coolant flow through so probably okay. I guess just don't wrap it in a bag and tie off so it can breathe .

But wait... there is more!

Took out the window cowl and man, not much between the water coming down your windshield and electronics in frunk. But weakest area is definitely passenger corner as wiper swipes water all down that corner. Has 2 water penetration spots


Now need to decide what is best remedy... But this is leak from windshield, probably more leak from wheel liner path. Need to find that after this dam is repaired.

 

[cleverscreenam]

Haha i saw the gray front clip and said "your Prius is showing!" I've owned 5, I'd recognize them with my eyes closed