openinverter forum

johu wrote: ↑Sun Jun 13, 2021 2:49 pm First test drive today. 105 km SoC from 93% to 44%. 110 on cruise control, even took it to 180 km/h briefly So 200 km on one charge if comfortably possible.

Really nice, congratulations! I have 96S at 120Ah = 42kWh now in my Pug. Net energy is probably cca 36kWh and i can go for 200Km at 110km/h with system saying 0% SOF fuel, but there is 16% SOC reserve. Tested a month ago.

That is also very promising. Of course the Peugeot isn't as tall, maybe I need to compensate that by going just 100

Just went on a quick charge tour.
I settled to charge at the local Peugeot dealer and the charger can provide 115A despite 125A being specified on the label. I started at 42°C battery temperature and 23% SoC. After 20 minutes I had acquired 14.2 kWh, so 42.6 kW average charging speed The BMS allowed full 70 kW up till 50% then slowly throttled down and reached 43 kW at 60%. The battery had reached a toasty 52°C and I got a light 100 kW discharge limit which disappeared at 50 °C

Early forum release

check out my frunk
Should take the charger & cables

celeron55 wrote: ↑Fri Jun 05, 2020 7:54 pm Did you charge at any Veefil chargers? Those slim, round, plastic looking ones. I've attached the best photo I could find from my collection, but just search the name on the 'net for a better one.

I've never been able to make one of those reliably charge for me. I've tried tens if not hundreds of times and maybe succeeded randomly once and I have no idea why. The charger's non-existent error messages and barely existing user interface doesn't help.

I'd like to know if this has been a problem for your DIY CHAdeMO system.

I finally managed to solve this problem I was having with my CHAdeMO code. As always, the most persistent evasive problems are the most dumbest ones. I had misunderstood the usage of the 0x102 byte 5 bit 4 "Normal stop request before charging" flag. The short and 100% accurate guidance to this flag is: NEVER set it. Forget that it exists. It makes no sense and will just make your car not charge, with the charger acting like everything is ok but not charging, and you only have to have it momentarily set for that to happen. All this reads in the standard, I just wish I could read I guess.

I sometimes get ahead of myself thinking things should make sense and this is the worst one so far.

What I did was I started reading through Damien's latest stm32 CHAdeMO code with the intention of reading through the entire process, cross checking what happens between it and my code. And I didn't have to get far: Damien's code didn't even have a variable to control that particular flag and always sent it as 0, which was obvious by just looking at the CANbus send function. I removed the lines that touch that flag from my code and bam, it just works with every charger I could find.

EDIT: Also... it kind of blows my mind how many chargers charged just fine with me having that flag set.

wow, well that's good to know. I don't use that flag either.
Still strange that most chargers ignored it.

Oh no.... Actually I was going to drive down to Saarland today but after loading up the car with the former battery pack and checking ground clearance, I saw a noticeable imbalance from left to right.
And indeed, negligence got us.
I remember standing in the workshop with dad and wondering whether we should dismount the floor carpet to be able to counter the M10 bolts. And it seemed like we could tighten them sufficiently without countering.

We could not.

The trained eye sees a gap between the aluminium angle and chassis. And also the bolt is visisble with it clearly should not be.

So, drop the whole thing down again, do it properly and hopefully be done with it.

We call it human factor in aviation.
This is why we have detailed instructions for everything. And we are required to NOT be creative!

It is why i started with EV conversions in the first place...

And this is why we develop extraordinary hearing, smelling and vision to sense when baby is ill

Another workshop day and the battery should now be in for good! I've developed some routine in dropping in raising it so it took only a few hours. 6 hours total with me checking ALL pole connectors again, doing minor tweaks, dismantling all covers and finally cleaning up after myself.
Now it sits flush again, compare against the picture in the post above
It still leans to the left a bit, but now I know that's fine.
Why is my tool stuck? Cell phone cam came in very handy. Getting the tool in and out is still the most complicated part.

This might be a little blast from the past, but if I followed this thread correctly you got TÜV approval while still using a Meanwell power supply as DC/DC converter. I could not find an isolated Meanwell power supply with more than a few watts, so I assume you have used a non-isolated one? If so, how did you comply with the TÜV regulation that states that the HV circuit must be galvanically isolated, especially from the vehicle chassis? By disconnecting all ground wires and thus also isolating the LV circuit?

You didn't search properly Of course it was isolated. 12V, 62.5A. Don't remember the part number, it has a 750 in it. But it was now replaced with a more capable Ampera unit.

johu wrote: ↑Mon Jul 12, 2021 6:46 pm You didn't search properly Of course it was isolated. 12V, 62.5A. Don't remember the part number, it has a 750 in it.

Then it was most likely a RSP-750-12 or SP-750-12. It doesn‘t have some kind of “isolated“-label which I was searching for, but when looking at its block diagram it becomes clear that it is isolated. Thanks a lot.

Any power supply with an input of more than 48V is usually required to have an isolated output.

We've made it to Sweden!

Kassel to Kiel was very easy with the larger battery. One stop (too early) after 150 km and another one after 270 km. Both ended in power throttling because battery temp exceeded 50°C. But it didn't really matter. In Kiel we did a final charge before getting on the ferry, 20 kWh for free courtesy of the tax department. Thanks, good way to spend taxes
Consumption was 166 Wh/km. I had limited top speed to 110 and there was a lot of heavy traffic.

We then met bexander in Göteborg which was very fun

Consumption on E6 went up to 190 Wh/km while also driving 110 pretty much constantly. I'm thinking Swedish roads are quite rough and we had head wind. That same night an output cap of poor mans charger blew so now I depend fully on CHAdeMO stations. Found another free one at Tanumshede shopping center, famous from the videos of Teslabjörn.

celeron55 wrote: ↑Wed Jun 23, 2021 4:55 pm I finally managed to solve this problem I was having with my CHAdeMO code. As always, the most persistent evasive problems are the most dumbest ones. I had misunderstood the usage of the 0x102 byte 5 bit 4 "Normal stop request before charging" flag. The short and 100% accurate guidance to this flag is: NEVER set it. Forget that it exists. It makes no sense and will just make your car not charge, with the charger acting like everything is ok but not charging, and you only have to have it momentarily set for that to happen. All this reads in the standard, I just wish I could read I guess.

I also tried to use a Veefil charger today and it did not work. It started the sequence and displayed the correct SoC but then dropped right out again. I checked said flag, I never set it. I also tried CHAdeMO versions 0, 1 and 2 and disabled the voltage and current deviation flags. Nothing. At a later try I got the same behaviour plus error message "Err 11". Any ideas?

EDIT: I should add that we got into a thunderstorm today and there was a lot of spray. I did pull over and waited for the rain to stop, but still, could it be some isolation error?

Did you see the cable test voltage ramp up?

No CAN logging this time

You might be able to download a veefil error code list on the 'net. Maybe this, requires signing up though: https://ozleaf.proboards.com/thread/805 ... mmon-field

Here it is.
Err 11 means "Vehicle has failed to connect its battery pack in time". So another supposed contactor error or some bit forgotten/too late in CAN?
How do you handle byte 5, bit 3?

EDIT: I turn it on 10s after connector lock, but have made it a variable now.

Nice that you are on a longer trip

Did you get interested people around you when you are charging the e-touran?

I don't have almost any delays in my code. Once the charger reports there's more than 10A of current available I enable the charge enable flag and start waiting for seq2. Once seq2 activates I immediately close contactors.

Ok, will try 0 delay also then. In fact I have no control over the contactors, so when the charger chooses to close the relays, they close. Maybe a bit awkward if CAN says they are still open.

Yes E-Touran has drawn some attention but most conversations were sharing the pain about the infrastructure

Do you know which messages turned off the ESP warning? I'm broadcasting these messages (found a DBC file online, but it appears to be in German)