Tesla Gen 3 charger board assembly as of mid-2024

[RobCote]

Hey folks. I just received a Tesla gen 3 charger logic board V3 and I have a couple (probably stupid) questions regarding the assembly. I've spent the last several days reading everything I could in the legacy charger threads but the overwhelming majority of the information applies to the gen 2 charger boards and the early gen 3 charger boards.

I think this is right, but the wifi adapter setup seems a little bit wonky, so I might have it wrong. It fits inside the recommended enclosure, so it should work, but I don't know if I have the configuration of connectors right.





I think the connector identified SWDPROG is obsolete now, but I'm not sure. As I understand it, programming is done OTA through the WiFi adapter. I'm 100% in over my head, but I'm doing my best to learn all of this because it's fascinating. I just want to be sure I put it together correctly, so I don't ruin the hardware.

Finally, are the two spare straight (8) pin connectors surplus hardware? I don't see anywhere that they belong, but I'm just realizing as I type this that they may be intended to connect to the PCBs inside the charger. And CONN2 is not included, but required for the device to communicate via CAN, yes? As well as installing (2) resistors at SJ1 and SJ2?

[RobCote]

I believe I got the board assembled correctly. The documentation is really good for older versions of this board. Not so much for the current format. I can hardly find a reference to Gen 3 V3 anywhere. Maybe I'm looking in the wrong places.

Anyways, I'm struggling and could use some help. When I apply 12VDC to the control board and charger, the charger modules flash once then remain off. The control board has one solid red LED and one flashing red LED. The WiFi network is available and reliable. I'm using a complete Model S 75kWh battery pack, which I've opened up to override the contactors. I can't get the charger to charge the battery, though. There's no 3V activate signal going to the modules (module pin 3).

Is there a particular order of operations necessary? I've powered up the contactors and charger/control board with 12V, then plug in the EVSE to the wall, then the charge port. I've tried a bunch of variations thereof, as well. I can never get anything more than a solitary flash of the module boards. I can't help but feel like I'm missing some crucial step, but I can't find anything that's relevant to the current format of the control board.

[RobCote]

Is this board already programmed with stm32 firmware? According to documentation for the control board for the Gen 2 Tesla charger, it's necessary to flash the firmware with a ST-LINK programmer. The documentation doesn't exist for a Gen 3 charger, so I'm not sure.

[tom91]

Log into the wifi and check which version is on the board.

[RobCote]

Log into the wifi and check which version is on the board.

I see version - 1.17.R under spot values.

[tom91]

If you are not seeing any lights on the charger modules it means they are not being powered on.

https://github.com/damienmaguire/Tesla-Gen3-Charger

Did you do this check for the CAN?


Using the web interface put it into manual mode, and pull input 1 high to enable charging. Make sure to refresh the spot values to see them changes.
If things are wired as they should you should have the LEDs come on, on the modules.

V3 boards runs the commercial firmware that is shared with the Gen 2 charger https://openinverter.org/wiki/Tesla_Mod ... N2_Charger

[RobCote]

I measure 119.8 Ohms across the can terminals at each module board with the control board connected to the charger, but not powered up.

The web interface says "start inverter in manual mode". I want to assume this is how to start the charger in manual mode as well, but assuming stuff around here is liable to get me killed so I just want to be clear.

[tom91]

I measure 119.8 Ohms across the can terminals at each module board with the control board connected to the charger, but not powered up.

Might not be an issue but please add 120 ohm the the last module in the series, as it can work like this but ideally needs 120 ohm on both ends.

One thing to check what is the resistance without the controller plugged in.

[RobCote]

With the controller unplugged, the CAN terminals at both modules are open loop.

Have I fubared the modules?

[tom91]

Have I fubared the modules?

No, please add a 120 ohm resistor across the module furthest down the wiring. This is just to be safe.

Have you measured to see if there are voltages at the pins as expected as per pinout? Do not worry about activate for now.

[tom91]

The web interface says "start inverter in manual mode". I want to assume this is how to start the charger in manual mode as well, but assuming stuff around here is liable to get me killed so I just want to be clear.

No this is not it, look at the wiki link I posted as a reference.

[RobCote]

No, please add a 120 ohm resistor across the module furthest down the wiring. This is just to be safe.

Have you measured to see if there are voltages at the pins as expected as per pinout? Do not worry about activate for now.

I'm a little unclear on how the resistor is installed. Does it bridge the CANL and CANH terminals, or is it inline with the wire coming into CANL or CANH terminal?

I've checked voltages at the charger module terminals. Here's what I measure (powered on, EVSE and HV battery connected):
- Pin 1: 11.3V
- Pin 2: 0.0V
- Pin 3: 42mV
- Pin 4: 48mV
- Pin 5: 1.39V
- Pin 6: 2.7V
- Pin 7: N/A
- Pin 8: 4.89V

[tom91]

120 ohm across/between CAN Low and CAN High on last module

Voltages look good. Why are you connecting the EVSE and HV? we are not there yet this just adds risks, remove these. (leaving ofcourse the 12V to the controller and the controller plugged into the modules)

Now put it in manual mode via the web interface.

https://openinverter.org/wiki/Tesla_Mod ... N2_Charger

section Commercial firmware usage manual

Parameter inputype to manual

Then apply 12V to input 1 and the module LEDs should come on also refreshing the spot values: State and Lasterr should change

[RobCote]

Oh interesting. I had read - somewhere, I can't recall where at this point - that the charger wanted to see low voltage to power on the control board, then HV battery, then EVSE, in that order, to come on and operate. I thought it would stay dormant until it saw those inputs. Anyway, I have tried it without those connected and got about the same result.

That wiki link is very helpful. I'd come across it before, but I didn't realize it was applicable to the gen 3 charger as well. Good to know.

Does the sequence of events matter as you outlined above? I joined the 12V supply and IN1 wires to one just to simplify my setup for testing purposes, so IN1 goes high at the same time as the board is powered up, which would be before changing input type to manual. Does that matter?

[tom91]

Did you follow the steps and what changed in the spot values when refreshing them.

Please have In1 disconnected when changing the setting, this way you can see the spot values change.

Do as I ask and not ask a bunch of random other questions. Stop trying to make sense of it yourself, I am trying to figure out what is wrong with your setup. Your long posts just means it takes a lot more time to read and creates more confusion.

[RobCote]

I did follow the steps as you outlined them. Sorry for being a pest and trying to jump ahead of you. Just thought it might save you some time if I could understand a bit more what's going on. I really appreciate you trying to help out and I can't overstate that.

I disconnected the HV battery and EVSE, connected 12V to the control board, checked parameters and inputype was set to manual already. I applied 12V to IN1 and refreshed spot values. State changed from off to waitstart, lasterr remained as NONE.

[tom91]

Did any of the leds come on or any other spot values change?

I did not write the latest version of code for the STM32 I wrote the version before.

[RobCote]

When I apply 12V and ground initially (just to pins 1 and 2 of the control board 20-pin connector), the module LEDs flash once and the two LEDs on the control board come on; one stays on, the other flashes constantly until power is removed. Aside from that, nothing lights up.

I didn't check any other spot values, but I'll try it again and take note of each before and after setting IN1 to high.

[johu]

Hope I don't interfere with Tom. WaitStart means its waiting for timedly to run out. Set timedly to 0

[tom91]

The Leds I am on are the LEDs on the charger modules not the Controller.

When State changed from off to Run, it should be enable the charger modules. If the leds do not light and none of the charge module spot values populate we need to check lower level signals.

This would mean measure the enable and active lines on the modules.

[tom91]

Hope I don't interfere with Tom. WaitStart means its waiting for timedly to run out. Set timedly to 0

As per wiki page:

timedly

The default value is -1 but this results in the charging waiting forever. You need to set this to 0 (or more) for the OBC to work.

[RobCote]

Spot values that change when IN1 is connected to 12V:
state: from Off to WaitStart
enable: from Off to On

That's it.

As regards checking voltage at the module boards, I did that earlier:
- Pin 1: 11.3V
- Pin 2: 0.0V
- Pin 3: 42mV
- Pin 4: 48mV
- Pin 5: 1.39V
- Pin 6: 2.7V
- Pin 7: N/A
- Pin 8: 4.89V

It seems like there should be more than milivolts at pin 4 when the module is on, but I never see that occur. I don't know what it means in the table you posted where it says "3V signal from DUE".

[RobCote]

As per wiki page:

THAT TURNED ON THE MODULES!!

[RobCote]

I'm not sure it's actually charging the battery, though. What's the safest way to check? Measure current at the conductor to the battery?

[tom91]

Spot values.

You setup your controller properly and check the EVSE values populate as expected?