Hi all,
I'm Nick (nickb24 on Github) from Canada and I'm just getting started with OVMS and my 2020 e-Golf. I was wondering if anyone knows why UDS wasn't implemented in the e-Golf? What I'm looking to explore is to leverage the work done on this project (http://obd-amigos.linuxtech.net/) to get parameters into OVMS. Interesting ones for me is getting battery cell voltages and getting an accurate SOH of the battery pack. But these require UDS.
Welcome, Nick.
There is no need to implement UDS specifically for the e-Golf. UDS is just an ISO-TP command standard, so is covered by the OVMS ISO-TP poller framework. You can even send UDS requests via the shell and via Javascript.
If you want to add UDS queries to the e-Golf module, just follow the poller usage examples of e.g. the e-Up or (easier to understand) the Ioniq vFL, the latter also includes battery cell queries.
If your question is why the e-Golf implementation didn't yet include ISO-TP/UDS polling, that needs to be answered by the e-Golf developers.
Regards, Michael
Am 17.10.26 um 16:10 schrieb nicholas--- via OvmsDev:
Hi all,
I'm Nick (nickb24 on Github) from Canada and I'm just getting started with OVMS and my 2020 e-Golf. I was wondering if anyone knows why UDS wasn't implemented in the e-Golf? What I'm looking to explore is to leverage the work done on this project (http://obd-amigos.linuxtech.net/) to get parameters into OVMS. Interesting ones for me is getting battery cell voltages and getting an accurate SOH of the battery pack. But these require UDS.
OvmsDev mailing list -- ovmsdev@lists.openvehicles.com To unsubscribe send an email to ovmsdev-leave@lists.openvehicles.com
-- Michael Balzer * Am Rahmen 5 * D-58313 Herdecke Fon 02330 9104094 * Handy 0176 20698926
Hi Nick,
I'm not sure why it wasn't implemented, most likely it just wasn't required. If you choose to add UDS polling please keep in mind that according to this video https://youtu.be/jBZjk1BobSQ (Timestamp 14:20) the car alarm goes off if you poll data while the car is off.
Also we kind of have state of health data, we have the current HV energy stored in kWh and the SoC %. If you are close to a full charge you can calculate SoH this way (EnergyStored divided by SoC = EnergyStored at 100% SoC, divide that by the net Design Capacity and you got your SoH). Not sure about the accuracy of that though. It may also require waiting for the cells to finish balancing and waking up the car again. So reading SoH via UDS is probably better, but I think the only metric you get via UDS is Max Energy Stored (so a 100% SoC value) and you still need to divide that by the net design capacity (which differs between pre/post Facelift models).
Sincerely, Jona (SCjona on GitHub)
Hi Jona,
(Not sure why this is a different thread than the other one I replied to. Maybe I just don't understand how these mailing lists work. Sorry.)
I'll just re-mention what I told Andreas on the other thread:
The reason for my UDS polling questions is because I want to implement some of the data from this project: http://obd-amigos.linuxtech.net/
I'm interested in reading the real battery SOH (like ODIS does) and possibly the battery cells. But I was wondering how best to do it in OVMS. This shouldn't be something that is done often (and isn't needed to either). I'm thinking only once per charge session maybe? Or should it be on a per-request basis, meaning the user needs to ask for it and then it will read it on the next charge for example?
I have already prepared the UDS read commands to implement to accomplish this, but I was wondering if there was a historical reason in the e-golf OVMS history as to why UDS wasn't used or if there was any 'lessons learned'.
Nick
Hi Nick,
(I have no idea how mailing lists work either, I find them rather confusing and easy to get lost/miss things, for example I did not even get Andreas reply in my inbox)
I think these requests are cheap enough to allow regular polling. I would maybe only poll SoH at the end of a charging cycle or at the start of ignition power (whatever is feasible) because that value won't change often/much.
I read on the other thread that you turned FCAN back on. As a target I'd say keep CPU usage below 75% while the car is running (drivable, not just ignition). Without hardware filter that was pretty much 100% CPU usage leading to increased crashes/hangs.
As for the history of the vehicle module: I got my e-Golf/OVMS in June/July this year. According to the Github Issue the implementation before I started contributing was from Erick @ewfuentes and Mikus @mikus-ancevs (https://github.com/openvehicles/Open-Vehicle-Monitoring-System-3/issues/690#...) so those would be the people to ask I suppose.
Sincerely, Jona
participants (4)
-
Michael Balzer -
Nicholas Berlingieri -
nicholas@berlingieri.com -
ovms@jonawagner.me