Tesla Charger Support Thread

[Kevin Sharpe]

Any other thoughts?

Give the board a good visual inspection looking for dry joints. You should also measure the value of ADVREF on Pin 75 and confirm that it is stable.

[rx7r3ev]

I think my luck run out and my situation just went from bad to worse. I've remove the board from the charger and tested the continuity between the socket eye and pin 78 - see attached picture.

This solder pins are very tiny and hard to differentiate with the naked eye. I could not hear a beep from the tester so I push the pin down and it made a clear beep. I might have also push the other pins next to it and screwed it up. Anyways I think I have messed up the IC solder/pins even thought the software will still upload but the serial would not print at all. I've tried loading with a bare code of 'setup()' and 'loop()' with a print but no luck.

Anyways, I think this the end of the road for this board. I'll get a new one from evbmw. I would like to thank Kevin and tom91 for the suggestions in tracking my issue.

[xp677]

I'm having an issue with a Gen3 charger - there is a row of 3x 1206 10R resistors in series next to the Live fuse on each module. On each of my modules, one of these resistors was blown off the board on each module, it was R1 on one, and R6 on the other.

I have replaced these resistors, however as soon as I connect AC, they blow again. The modules so have communication over CAN, they are reporting accurate AC voltage (CAN ID 0x207, byte 1), but the AC present bits (CAN ID 0x207, byte 2, 00001100) are 0 when AC is connected, I think it may be linked to those resistors?

I have requested 80V from the charger, and have connected a 100R, 500W resistor across the output to simulate a load. However, the charger does not produce an output voltage. I'm not sure if this is related to my resistors, or if I've just set the voltage too low.

A voltmeter across the blown resistor shows 230v.

Any idea what's causing this?

[Jack Bauer]

I have said this in quite a few places but here we go again :

Running a Tesla charger at much under 200v dc will cause it to explode. Yes I know the label says 50 to 450v but it lies. Yes I blew one up discovering this.

[Kevin Sharpe]

I have said this in quite a few places but here we go again :

Running a Tesla charger at much under 200v dc will cause it to explode. Yes I know the label says 50 to 450v but it lies. Yes I blew one up discovering this.

I've updated the wiki;

https://openinverter.org/wiki/GEN2

[xp677]

I have said this in quite a few places but here we go again :

Running a Tesla charger at much under 200v dc will cause it to explode. Yes I know the label says 50 to 450v but it lies. Yes I blew one up discovering this.

Thanks Damien, that's good to know. Fortunately for me, the charger doesn't seem to attempt to charge at all.

My issue with the burning out resistors happens as soon as I connect AC, even before powering up the chargers logic. I'll edit this post with a photo of the resistors in question.

I've updated the wiki;

https://openinverter.org/wiki/GEN2

That looks good, however my question was regarding the Gen3 charger. I assume the same is true for the Gen3? If so, I imagine that would need updating as well.

[Kevin Sharpe]

I've updated the wiki;

https://openinverter.org/wiki/GEN2

That looks good, however my question was regarding the Gen3 charger. I assume the same is true for the Gen3? If so, I imagine that would need updating as well.

AFAIK Gen 2 and Gen 3 power stages are very similar and I've therefore updated the Gen 3 wiki as well.

[Kevin Sharpe]

My issue with the burning out resistors happens as soon as I connect AC, even before powering up the chargers logic. I'll edit this post with a photo of the resistors in question.

I suggest you also provide details of the wiring (including grounds), whether it's a US or EU unit (photo of labels), how you've modified it (single/three phase, etc.), and the history of the donor vehicle (any sign of flooding?).

Did the resistors burn out the first time it was powered up? What was connected at that time?

[xp677]

Sure, to answer the questions:

This is a 48A NA Gen3 charger.

There were no signs of flooding, fire damage, etc on this charger. The charger was still sealed with it's warranty tag.

The charger contains two modules (and an empty bay). The original wiring had both modules in parallel - single phase.

I have modified it by connecting a controller to the 12v, 3v3, 5v, GND, CANH and CANL connections. The modules work as expected, when powered up, I can send and receive valid CAN data.

The current wiring setup is a 13A UK plug, with L and N connected to the L and N terminals of Module 1 (see photo). The Earth is bolted to an otherwise unused thread on the charger chassis.

The resistors were already burnt out when I received the charger. Since different resistors had burnt out on each module, it was possible to cross reference these to find the value (10 ohms, noted "100" label, and verified with a multimeter).

After replacing the resistors, I connected power to Module 1. There was no load connected at the time. The modules were unpowered (no 12v, 3v3 or 5v power). The recently-replaced resistor (R8) on Module 1 immediately blew.

After powering up the module, it detected 230(ish) V AC through CAN, however the "AC present" flag (as mentioned above) is still negative. My initial impression is that the resistors in question are part of this detection.

See below, Module 1, as it sits now, with my replaced R8 resistor blown,

Module 1

Module 2 has not been connected to AC power yet. See here, It's previously blown resistor (R1) has been replaced, I have little doubt that it will blow as soon as I connect power to the module.

Module 2

These are the resistors which I used as a replacement: https://uk.rs-online.com/web/p/surface- ... s/8071117/

There is 240V measured across the terminals of the late R8 on Module 1.

Hopefully this helps?

[Kevin Sharpe]

The resistors were already burnt out when I received the charger. Since different resistors had burnt out on each module, it was possible to cross reference these to find the value (10 ohms, noted "100" label, and verified with a multimeter).

After replacing the resistors, I connected power to Module 1. There was no load connected at the time. The modules were unpowered (no 12v, 3v3 or 5v power). The recently-replaced resistor (R8) on Module 1 immediately blew.

Any chance you can return the charger for a refund? If not then it sounds like you need to reverse engineer the charger power board to understand what's going on.

[xp677]

I'm not able to return the charger. I'll tear it down and begin reverse engineering it.

I'd hoped that this had been a more common issue with a simple solution!

These boards are tough to trace - tesla don't seem to like putting traces on the outer layers.

[Kevin Sharpe]

I'd hoped that this had been a more common issue with a simple solution!

Very few people are using Gen 3 chargers and those that are don't share much information.

I've seen a few blown Gen 2 chargers but no one has successfully repaired one AFAIK.

[xp677]

Do you know what had blown in the Gen2 chargers? From what Damien was saying earlier, it sounded more catastrophic than the issue I'm experiencing here.

I just plan to throw a larger value resistor in at first and see how that fares.

[bulletbug]

(Kindly, before someone tells me to read the charger thread....done that...twice in the last 24hours. )

I'm doing intial setup and testing of gen2 v4 charger. I'm in the US using EVSE that is known to be working since I use it with my Chevy Spark EV.

Battery pack I'm attempting to charge is a full pack from Volt/Ampera that I intend to charge to 400v.

Current Voltage on pack 380 v

Here are my settings:

Settings Menu
1 - Auto Enable : ON
2 - Modules Enabled : 123
3 - Can Mode : Off
4 - Port Type : 1
5 - Phase Wiring : 2
6 - DC Charge Voltage : 400V
7 - AC Current Limit : 10A
8 - CAN0 Speed : 500
9 - CAN1 Speed : 500
a - Can Debug : OFF
b - EVSE Debug : ON
t - termination voltage : 400V
q - To Quit Menu


I get this as a result when I attempt to start charging. All 3 phases light up, click a few times and then it shuts down. The DC volts show at 515v?



Proximity Status : Connected AC limit : 39600 /Cable Limit: 0 /Module Cur Request: 0 /DC total Cur:0.00 /DC Setpoint:400 /DC tVolt:400 /DC driven AC Cur Lim: 5000
8014 State: 1 Phases : 1 Modules Avtive : 3 ON D1 H
Phase 1 Feebback // AC present: 1 AC volt: 246 AC cur: 19.20 DC volt: 515 DC cur: 0.01 Inlet Targ: 60 Temp Lim Cur: 15 25 24 EN:1 Flt:0 Stat:1001101
Phase 2 Feebback // AC present: 1 AC volt: 249 AC cur: 27.73 DC volt: 515 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:1001101
Phase 3 Feebback // AC present: 1 AC volt: 247 AC cur: 19.20 DC volt: 519 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:101101

Proximity Status : Connected AC limit : 39600 /Cable Limit: 0 /Module Cur Request: 0 /DC total Cur:0.01 /DC Setpoint:400 /DC tVolt:400 /DC driven AC Cur Lim: 47512
8515 State: 1 Phases : 1 Modules Avtive : 3 ON D1 H
Phase 1 Feebback // AC present: 1 AC volt: 246 AC cur: 19.20 DC volt: 515 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:1001101
Phase 2 Feebback // AC present: 1 AC volt: 249 AC cur: 19.20 DC volt: 515 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:1001101
Phase 3 Feebback // AC present: 1 AC volt: 247 AC cur: 19.20 DC volt: 519 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:101101

Proximity Status : Connected AC limit : 39600 /Cable Limit: 0 /Module Cur Request: 0 /DC total Cur:0.00 /DC Setpoint:400 /DC tVolt:400 /DC driven AC Cur Lim: 5000
9016 State: 0 Phases : 1 Modules Avtive : 3 ON D1 H
Phase 1 Feebback // AC present: 1 AC volt: 245 AC cur: 10.67 DC volt: 515 DC cur: 0.01 Inlet Targ: 60 Temp Lim Cur: 15 25 24 EN:1 Flt:0 Stat:1001101
Phase 2 Feebback // AC present: 1 AC volt: 249 AC cur: 27.73 DC volt: 515 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:1001101
Phase 3 Feebback // AC present: 1 AC volt: 247 AC cur: 19.20 DC volt: 519 DC cur: 0.00 Inlet Targ: 60 Temp Lim Cur: 15 25 23 EN:1 Flt:0 Stat:111001

Proximity Status : Connected AC limit : 39600 /Cable Limit: 0 /Module Cur Request: 0 /DC total Cur:0.01 /DC Setpoint:400 /DC tVolt:400 /DC driven AC Cur Lim: 47974
9517 State: 0 Phases : 1 Modules Avtive : 3 OFF D1 H
Modules Turned OFF

Proximity Status : Connected AC limit : 39600 /Cable Limit: 0 /Module Cur Request: 0 /DC total Cur:0.01 /DC Setpoint:400 /DC tVolt:400 /DC driven AC Cur Lim: 47974
10018 State: 0 Phases : 1 Modules Avtive : 3 OFF D1 H
Modules Turned OFF

Proximity Status : Connected AC limit : 39600 /Cable Limit: 0 /Module Cur Request: 0 /DC total Cur:0.01 /DC Setpoint:400 /DC tVolt:400 /DC driven AC Cur Lim: 47974
10519 State: 0 Phases : 1 Modules Avtive : 3 OFF D1 H
Modules Turned OFF

Can anyone help point me to what I'm doing wrong?

[tom91]

Something seems off in terms of canbus to the phase modules then, unless they are broken.

Can you get a can capture of the internal canbus?

[bulletbug]

Thank you for the response. When the Canbus logging is turned on in the interface, does it write this out to a text file?

I don't currently have the Can hi and lo connected to anything or terminated in any way. I don't intend to use the canbus communication initially. Do I need to do anything to the leads coming from the b4 and b10 lines?

[tom91]

The internal canbus is what you need to log, not the external canbus.

[bulletbug]

Would the internal CanBus be picked up via Conn3 on the v4 board?

[Kevin Sharpe]

Would the internal CanBus be picked up via Conn3 on the v4 board?

Yes

https://github.com/damienmaguire/Tesla- ... ematic.pdf

[bulletbug]

I'm not able (yet) to get a canbus log, but would like to restate something from previous post that I think may help someone help me.

My pack voltage is 380 v as measured with my volt meter.

The serial debugging of the board reports the voltage as 515 v.

I saw this issue raised in this post - viewtopic.php?f=10&t=78&start=50#p2254

It was acknowledged as an odd thing by Damien and that he'd seen the same thing in one of his "chargers on the shelf". Did this get resolved? Can someone point me towards that solution?

[bulletbug]

I believe I found the solution to my "damn thing is showing pack voltage at ~515v when it's actually at 380v" which in turn was preventing the charger from releasing the happy juice into the battery.

It appears that the CAN message coming from sensors within each of my phases needed a different scaler/multiplier to convert from HEX to integer.

I tracked down where the CAN message was being converted from HEX (is that right) to an integer (human readable) which is currently located starting at line 820 in the code.

Line 822 (for phase 1) : dcvolt[0] = ((frame.data.bytes[3] << 8) + frame.data.bytes[2]) * 0.0105286; //we left shift 8 bits to make a 16bit uint.

Using some of that fancy 8th grade algebra I figured out that if I changed the .0105286 to .007804311. It changed my dcvolt variable (the variable used to observe the battery pack voltage) from the incorrect ~515v to the correct ~381.5 that I was seeing on my separate volt meter.

After compiling and uploading to the board again, Phase 1 now showed the correct voltage. YAY! Charging would still not start because I needed to make the same change to the other 2 phases.

I went back into the software and changed this same multiplier on line 827 (for phase 2) and line 832 (for phase 3).

Upload that and it WORKED! Using my AMP Clamp meter, I was able to see the amps going into the battery pack climb up to 5 amps.

Now I'm seeing that the serial monitor DC Amps (reported by the variable dccur) is inaccurate. When my Amp clamp meter is over the battery packs most negative terminal shows 5 amps the serial monitor is showing less than 0.03 on any single phase.

I'm bumping into temperature shut downs so I'll need to get my water cooling loop in place before I can figure out how to get the displayed dccur sorted out. Not sure it's even important other than on the serial monitor display.

[joromy]

Line 822 (for phase 1) : dcvolt[0] = ((frame.data.bytes[3] << 8) + frame.data.bytes[2]) * 0.0105286; //we left shift 8 bits to make a 16bit uint.

This is very good finding!!
I have the same voltage conversion with correct voltage reading, but with an European model.
Maybe we should make a list with prod version serial number....

[bulletbug]

I think there are other issues with this charger related to how the software is decoding the HEX from the power module CAN messages. I'm not a programmer by a long stretch so take anything I say with caution as it relates to changing code.

I set up a simple water pump and bucket to manage temps. Without it I think I was seeing shutdowns based on temp after the charger was putting out 5 amps and warming up. Now that I have confirmed water flow through the charger, I think I may be seeing shutdowns for it being to cold? The water isn't cold or hot at about 60 degrees F.

This makes me wonder if the temp sensors or at least the underlying CAN messages are based in Fahrenheit (in my charger) and Celsius in chargers from other parts of the world.

I'm guessing that their logic somewhere in this charger, either at the Damien board level or deeper in that reduces or prevents charging current when temps are extremely low in order to protect the battery. In all my testing today with both a 110v AC J1772 portable charger and a 240v 40 amp EVSE - I'm not able to get total current (as measured with Clamp Amp meter) above 5 amps.

BTW - I noticed the voltage input level is a setting in the code at line 61:

uint16_t ACvoltIN = 240; // AC input voltage 240VAC for EU/UK and 110VAC for US

I'm assuming in order to keep all the calculations on track...this needs to be manually edited, compiled and uploaded to the control board before switching between my higher and lower AC voltage chargers. I think this needs to be addressed in the setup instructions so that it will prevent other's from missing it like I did for a couple of hours.

Once I understand this, my intention is to help tune up the instructions and FAQ for the gen 2 charger.

[cancer038]

I don't have a gen 2 charging device yet. but i am following the subject. The gen2 installation guide should be updated by experienced people. with many illustrative images and videos. for people who will try it for the first time. Damien seems to be very busy.

I think there are other issues with this charger related to how the software is decoding the HEX from the power module CAN messages. I'm not a programmer by a long stretch so take anything I say with caution as it relates to changing code.

I set up a simple water pump and bucket to manage temps. Without it I think I was seeing shutdowns based on temp after the charger was putting out 5 amps and warming up. Now that I have confirmed water flow through the charger, I think I may be seeing shutdowns for it being to cold? The water isn't cold or hot at about 60 degrees F.

This makes me wonder if the temp sensors or at least the underlying CAN messages are based in Fahrenheit (in my charger) and Celsius in chargers from other parts of the world.

I'm guessing that their logic somewhere in this charger, either at the Damien board level or deeper in that reduces or prevents charging current when temps are extremely low in order to protect the battery. In all my testing today with both a 110v AC J1772 portable charger and a 240v 40 amp EVSE - I'm not able to get total current (as measured with Clamp Amp meter) above 5 amps.

BTW - I noticed the voltage input level is a setting in the code at line 61:

uint16_t ACvoltIN = 240; // AC input voltage 240VAC for EU/UK and 110VAC for US

I'm assuming in order to keep all the calculations on track...this needs to be manually edited, compiled and uploaded to the control board before switching between my higher and lower AC voltage chargers. I think this needs to be addressed in the setup instructions so that it will prevent other's from missing it like I did for a couple of hours.

Once I understand this, my intention is to help tune up the instructions and FAQ for the gen 2 charger.

[Jack Bauer]

Great work guys. I'll update the main code when I get back on my feet (stuck in bed with back pain past few days). Regards instructions etc I agree 100% and when people ask how they can help this is what I say but sadly few seem interested.