Chademo Controller Development

[JaniK]

where is it saying the swd error?

is your stm32 getting warm once the board is powered on?

It says SWD error here:

No luck updating from wifi here

My stm32 is as cold as the pcb next to it.

After last post, I did this:
Unplug the esp8266 from VCU
Connect st-link to swdprog on ISA FC VCU V2
Plug in st-link to PC
open st-link software
Select address 0x08000000
Open: https://github.com/jsphuebner/tumanako- ... loader.hex
Ctrl+P (target, program & verify)

Select address 0x08001000
Open: https://github.com/damienmaguire/Chadem ... hademo.hex
Ctrl+P (target, program & verify)
Ctrl+D (target, disconnect)

Unplug ONLY clock and data wire from VCU
Plug in the esp8266 to VCU
Plug back the st-link USB to PC. (Powering VCU from stlink 3.3v)

Connect phone/PC to esp8266 wifi

And now I have parameters!

Whee! it looks like it's working !

I don't know what went wrong the first round but then I was learning what do (so something went south) and now I just did what I wanted to do (and got lucky )

[JaniK]

Whee, the IVT-S spits out 0x512 - 0x528 messages on CAN to MCP2515, to Arduino nano, to Laptop

In theory, I should get this working since all parts say "hello" now!

Had to put CAN receiver to 1000kbps like someone on the wiki had to do too.

Now then, the software is years old,

@bratitude wondered if generic openinverter interface works, so did you mean only web interface or the stm32 program?

[P.S.Mangelsdorf]

I was finally able to test CHAdeMO charging again today, and there is definitely something wrong with my set up, but I can't for the life of me figure out the error.

I tested at three stations, two Delta chargers in the EVGo network, and one Electrify America. In all cases, I appear to get all the way to delivering power before there is an error. On the EA station, the station even got to the Charging screen, before the error happened.

Attached are all three CAN logs and below is a link to a screen record of the final EVGo attempt. I'd appreciate any help in figuring out what I've screwed up.

Also does anyone have and/or can direct me to a list of the CHAdeMO CAN error messages? There's nothing in the wiki.

[royhen99]

From the voltage readings, 373V it looks like the contactors are closing but in all the logs the requested current is 0A.

This is the only document I have found that list some error messages. Page 32 lists some reasons for termination of charging.
https://cache.industry.siemens.com/dl/f ... 0118AB.pdf

[P.S.Mangelsdorf]

From the voltage readings, 373V it looks like the contactors are closing but in all the logs the requested current is 0A.

Which is odd, because it was set to max 25A , with a max V of 388. I'll try later with max v of 400, which is actually the pack's max V, and see if it helps. The stations reported a connection issue, which seemed odd that late in the process.

[P.S.Mangelsdorf]

This is the only document I have found that list some error messages. Page 32 lists some reasons for termination of charging.
https://cache.industry.siemens.com/dl/f ... 0118AB.pdf

Looking at those, I think the issue has to be the 0A requested, because the station reports the shut down coming from the car side.

Does anyone know what direction the board wants to see current? i.e. right now it reports a negative current when charging using the onboard charger, and a positive when driving. Does the board want to see the reverse? I know the shunt can be changed to report the inverse.

[JaniK]

ISA Shunt CAN communications working with the ISA FC VCU V2 (STM32)

ISA_FC_VCU_V2_with_ISA_8.62V_RC-CARlipo.jpg (13.47 KiB) Viewed 3043 times

Using RC car lipo balance plug and dupont cables to test

(little offtopic, other board, LEAF VCU (DUE) doesn't show the voltage for some reason, but it also receives the 0x521 to 0x528 frames via CAN from ISA Shunt. But I mainly need the shunt for the chademo part, so I can sort that later.)

Next step here, is to wire the chademo plug and make a contactor box between chademo plug and battery.

[P.S.Mangelsdorf]

I tested again this evening with a lower pack voltage and higher udcthresh, same issue.

Is this (below) discrepancy (positive vs negative) possibly what's causing the shutdown? I looked through the screen record and the frame before the malfunction idccdm (I assume current commanded?) is 3A, so the car claims it is asking for current, albeit only a small amount, just before the malfunction.

[arber333]

I was finally able to test CHAdeMO charging again today, and there is definitely something wrong with my set up, but I can't for the life of me figure out the error.

I tested at three stations, two Delta chargers in the EVGo network, and one Electrify America. In all cases, I appear to get all the way to delivering power before there is an error. On the EA station, the station even got to the Charging screen, before the error ...

That looks like exactly what happens with Delta chargers and my setup. Chargers like ABB work fine.
That Siemens document made me think. Suggested CAN setup is 500kbaud and 2ms!!! What if we are not cycling telegrams fast enough?
I have a Leaf DUE board and i have to reprogram it.
I will test this soon...

[P.S.Mangelsdorf]

That looks like exactly what happens with Delta chargers and my setup. Chargers like ABB work fine.
That Siemens document made me think. Suggested CAN setup is 500kbaud and 2ms!!! What if we are not cycling telegrams fast enough?
I have a Leaf DUE board and i have to reprogram it.
I will test this soon...

I'll see if I can find an ABB charger around here to test on, that should confirm if I have an issue or if its a problem with Delta chargers (and presumably EA chargers too)

[royhen99]

Suggested CAN setup is 500kbaud and 2ms!!! What if we are not cycling telegrams fast enough?

The 2ms is for ethernet used between modules of the test set-up. As far as I know, which isn't very much, the CAN bus messages should be sent every 100ms.

[P.S.Mangelsdorf]

So I tested on an ABB charger yesterday, and it failed as well. I got the same message I've seen on other chargers "Make sure your car is turned off and try again" I'm not sure if this is a standard error message or if it says anything about what might be wrong. CAN log is attached.

This was at the only ABB charger that I can reach from my house, so I can't double check this elsewhere.

[royhen99]

You still have not requested any current. Have you tried setting cdmcureq to say 25A and charge limit to the maximum current e.g. 100A or 125A. I am not sure whether the Chademo current limit parameter is chargelimit or chgcurlim. What FW/HW are you using?

[P.S.Mangelsdorf]

You still have not requested any current. Have you tried setting cdmcureq to say 25A and charge limit to the maximum current e.g. 100A or 125A. I am not sure whether the Chademo current limit parameter is chargelimit or chgcurlim. What FW/HW are you using?

I'm using the ISA version of Damien's dedicated CHAdeMO controller. Its on FW 1.00R, which to my understanding is the only firmware available. This version has three settings, my values are below.

chargelimit 25A
soc limit 100%
udcthresh 400V

[royhen99]

After further study of the code I am not sure if this version of the code works ( I think development was switched to Zombieverter before it was fully tested ). As you say chargelimit is the target current, but the actual value is the minimum of chargelimit and chgcurlim. chgcurlim is 0 so the requested current will always be 0. cdmcureq is what is actually sent to the charger and includes ramping. chgcurlim needs to be set to a value above 0, e.g. 125A, which can only be done by modifing the code.

[P.S.Mangelsdorf]

After further study of the code I am not sure if this version of the code works ( I think development was switched to Zombieverter before it was fully tested ). As you say chargelimit is the target current, but the actual value is the minimum of chargelimit and chgcurlim. chgcurlim is 0 so the requested current will always be 0. cdmcureq is what is actually sent to the charger and includes ramping. chgcurlim needs to be set to a value above 0, e.g. 125A, which can only be done by modifing the code.

I appreciate the analysis. Would it make sense for the code to be changed so chgcurlim=chargelimit? It seems they are/should be the same thing.

[royhen99]

I have already modified the code but can only test on Bluepill board to check basic functions. I left chgcurlim as it is and changed chargelimit to chargetarget. chgcurlim is a safety limit so I have left that in place.

[P.S.Mangelsdorf]

I have already modified the code but can only test on Bluepill board to check basic functions. I left chgcurlim as it is and changed chargelimit to chargetarget. chgcurlim is a safety limit so I have left that in place.

Can you share the modified code? I am willing to test on my board. (Note: it'll be at least a few days before I can test because a different system failed yesterday and I can't fix it until the weekend)

[royhen99]

I will PM it to you. I can put in on here or GitHub after testing. I have set chgcurlim default to 125A, and chargetarget default to 25A.

[JaniK]

I will PM it to you. I can put in on here or GitHub after testing. I have set chgcurlim default to 125A, and chargetarget default to 25A.

Great that we have more people interested in getting these boards to life. Thank you!

I'm using the ISA version of Damien's dedicated CHAdeMO controller. Its on FW 1.00R, which to my understanding is the only firmware available. This version has three settings, my values are below.

chargelimit 25A
soc limit 100%
udcthresh 400V

I have the same firmware on my board.

Do you also have ISA FC VCU V2 board ?

I have ABB and Kempower chargers ~5minutes from home, I will be testing my controller on them, as soon as I have it fully assembled to connect to battery pack. (missing HV junction box, contactors and cables between pack and chademo)

[P.S.Mangelsdorf]

Do you also have ISA FC VCU V2 board ?

Not sure if I have v2 or v3, I'll open up the enclosure and check when I'm working on the car this weekend.

[JaniK]

Not sure if I have v2 or v3, I'll open up the enclosure and check when I'm working on the car this weekend.

No need to open for V2 /V3 check.

If it is STM32 thenyou see from outside too.

If you have 12 + 20 pin connectors it is V2

If you have one 35pin connector it is V3

Edit: they should be essentially the same unit, (V2/V3) with just different connector.

[P.S.Mangelsdorf]

No need to open for V2 /V3 check.

If it is STM32 thenyou see from outside too.

If you have 12 + 20 pin connectors it is V2

If you have one 35pin connector it is V3

Edit: they should be essentially the same unit, (V2/V3) with just different connector.

I've got V3, just checked.

(I have it in a different enclosure that has to be unscrewed - long story.)

[P.S.Mangelsdorf]

I will PM it to you. I can put in on here or GitHub after testing. I have set chgcurlim default to 125A, and chargetarget default to 25A.

It uploaded to my board cleanly, hoping to find time to test this weekend. Forecast doesn't look great, but we'll see.

[P.S.Mangelsdorf]

I was able to make a quick run to one of the EVGO Delta chargers, and it still failed. Based on Arber's experience, that may not be dispositive.

I wasn't able to get a log due to time constraints. I'm hoping to find time this week to test at all of the chargers again (ABB, EA, EVGO/Delta) and get logs from each. Unfortunately the weather here is still a mess (and I don't have windows), and work is busy.