OVMS (Again)

[GreyPeter]

I have uploaded the latest firmware version to https://media.dragon2611.net/ovms/ovms3.3.003-499-gbe900b4a.bin All MG specific shell commands now need to be preceded by xmg e.g. xmg polls on Use xmg ? to see all commands for selected model. This is a requirement before I push it to the main code. There are a few other small fixes as well.

 

[jamessimpson]

Hi Peter, thanks for your posts and work to try and help us all.

Please can I ask advice? To be honest, I'm a bit confused....

I've got a ZS EV in the UK. I've just updated to your latest firmware version as above (10th June) and have my vehicle type set to AU/TH (after reading OVMS docs). But I'm still getting problems with the alarm going off when polling via home assistant if charging when the car is locked.

How should I proceed? Should I try again with UK/EU setting? I couldn't find out how to try the setting in the new firmware you referred to "where the polling is able to be selected as manual or automatic", to see if that makes a difference. Should I be setting up my polling settings in HA to only poll at certain times?

Thanks in advance!
James

 

[GreyPeter]

James you should be setting the vehicle to MG EV (UK/EU) unless your vehicle is the new facelift model, if so, try the MG ZS EV MK2 setting. Can you let me know which model you have.

 

[jamessimpson]

Dear Peter, thanks for your reply. It's a 2021 model, not the new facelift model. I'll switch it back to the MG EV (UK/EU) setting. Once I've done that would you expect it to be able to charge with the doors locked since I've updated to your latest firmware?

Thank you.

 

[dragon2611]

Dear Peter, thanks for your reply. It's a 2021 model, not the new facelift model. I'll switch it back to the MG EV (UK/EU) setting. Once I've done that would you expect it to be able to charge with the doors locked since I've updated to your latest firmware?

Thank you.

If it still sets the alarm off then @GreyPeter might have to implement the manual polling, but he problem with manual polling is you won't get the data when charging unless you start polling (which may set the alarm off).

I did wonder if there's a way to disable the alarm going off when something pokes at the OBD but I suspect it's on purpose to stop theft via obd2 hacks.

 

[GreyPeter]

Dear Peter, thanks for your reply. It's a 2021 model, not the new facelift model. I'll switch it back to the MG EV (UK/EU) setting. Once I've done that would you expect it to be able to charge with the doors locked since I've updated to your latest firmware?

Thank you.

You should have no problems with the alarm going off on the 2021 MG ZS EV now that you have selected the correct vehicle.

 

[jamessimpson]

You should have no problems with the alarm going off on the 2021 MG ZS EV now that you have selected the correct vehicle.

Dear Peter, just to say thanks very much, you've cured it!

Your help was much appreciated.

Cheers.

 

[GreyPeter]

Dear Peter, just to say thanks very much, you've cured it!

Your help was much appreciated.

Cheers.

Hi James,

Good to know that it is working.
Enjoy using OVMS.

Peter

 

[jamessimpson]

Hi James,

Good to know that it is working.
Enjoy using OVMS.

Peter

Please can I ask another question?

Any idea why I'm intermittently getting crazy figures out of the OVMS module for SoC and Range? It seems to be sporadic but is always the same figure (7159.5 %). When I unplug and reboot the module it goes away for a bit.....

 

[GreyPeter]

We did have a problem with the battery voltage going crazy sometimes and found the battery voltage reading went haywire as the bus turned off. I wonder if this is a symptom of the same thing. I think it usually happened when charging was stopped or perhaps just after the vehicle was turned off. I will investigate further but in the meantime can you try to work out if there is any rhyme or reason to it.

 

[jamessimpson]

OK thanks I will do some fiddling and see if I can establish a pattern.

 

[jamessimpson]

OK thanks I will do some fiddling and see if I can establish a pattern.

Hi Peter,

It did it again today. At approx 74% charge it switched to "done" status, and then shortly afterwards reported a charge state of 7159.5 %. It had been charging for a couple of days, on and off with solar, uneventfully. Looking at your previous post I looked back at the 12v status and there were some dips in this shortly beforehand. They look like erroneous blips rather than true readings, I have attached a screenshot of the timings and some server logs in case that would help?

 

[GreyPeter]

Hi Peter,

It did it again today. At approx 74% charge it switched to "done" status, and then shortly afterwards reported a charge state of 7159.5 %. It had been charging for a couple of days, on and off with solar, uneventfully. Looking at your previous post I looked back at the 12v status and there were some dips in this shortly beforehand. They look like erroneous blips rather than true readings, I have attached a screenshot of the timings and some server logs in case that would help?

Hi, I don’t believe the 12V should be jumping around like that. I believe you may have a loose or bad connection somewhere. It could be the CAN lead or possibly even the 12V Battery connections.

 

[jamessimpson]

OK thanks. I will do some investigating along those lines. I'll let you know how I get on!

 

[jamessimpson]

Hi, I don’t believe the 12V should be jumping around like that. I believe you may have a loose or bad connection somewhere. It could be the CAN lead or possibly even the 12V Battery connections.

Dear Peter,

You were right. By re-positioning the unit and uncoiling the CAN lead (and GPS/GSM leads) all the strange behaviour has gone away. So it was either a loose connection in the CAN cable which isn't manifest when straightened, or perhaps interference from the rats nest of cables that was there before?

Anyway, thanks again for your help,
James

 

[jamessimpson]

We did have a problem with the battery voltage going crazy sometimes and found the battery voltage reading went haywire as the bus turned off. I wonder if this is a symptom of the same thing. I think it usually happened when charging was stopped or perhaps just after the vehicle was turned off. I will investigate further but in the meantime can you try to work out if there is any rhyme or reason to it.

Dear Peter, did you have any luck investigating this issue? It went away for several months and I had put it down to a loose connection but it seems to have re-appeared. I have tried a new OBD lead to exclude a connection problem. As you state, it usually seems to be associated with charging, or shortly after locking the car. I have also made sure that the 12 V volt battery reading is accurately calibrated. None of the other metrics seem to be affected, just the SOC which jumps to exactly the same figure mentioned by another user above, 7159%. When I switch the ignition on, it then reverts back to the correct value.

I don’t suppose you have any ideas, please? Is it a bug? Thanks very muchy.

 

[dragon2611]

@jamessimpson Are recent releases of OVMS installing ok for you, the last 2 times I tried to do an OTA it softbricked the OVMSv3 unit, the first time was probably because I hadn't updated from the Beta I was previously using for a long time, but after the 2nd time I just never bothered to restore it and took it out.

 

[jamessimpson]

Yes they seem to be, running 3.3.004-4-gab8f16b7. I recently did a factory reset twice recently then OTA update afterwards. Sorry to hear yours isn't going so smoothly....

 

[jamessimpson]

It’s a strange behaviour either way as the values look about right when the car is on.

When it’s off the voltage reads over 1000v and the soc/soc raw sometimes goes to that silly reading.

Only happens sometimes if ovms tries to pool when the car is locked hasn’t done it since I calibrated the 12v and ovms goes to sleep when the cars off.

I've also had this strange behaviour with my MG ZS EV (UK). Occasionally I get 7159% SoC, sometimes it seems to coincide with charging or soon after locking. It always reverts to correct value when the car is woken up again. I have tried for the last year to identify the cause but it only seems to be sporadic....

 

[GreyPeter]

Hi, first of all I would suggest switching to Empty Vehicle before doing an OTA update. I believe there is a boot bug in the code that only occurs when booting into a new install. James, I will look into your SoC problem but can you install the latest OTA Edge code first. I think this problem has been fixed.