Mitsubishi outlander charger and DC:DC

[nkiernan]

Alibro, I'm not too far from you I believe and I have an outlander charger on the shelf and a savvycan setup, so if you can't get the issue figured out soon it could be an option to test with the spare charger to rule out a damaged unit you might have. Sounds like it could be the control side hopefully

[Alibro]

Is it possible that EVSE somehow disconnected PP signal? That would cause the change you speak of. Or maybe overheating....
There is B5 = Statusbyte within 0x389 which can tell you the status for charger and there is one for DCDC in 0x377 B7 = Statusbyte

At the moment I can't rule anything out especially overheating as the car cutting out (while driving) happens more often after a charging session.
It is not possible that the PP signal disconnecting is the issue as I didn't have it connected until a few days ago, it didn't make any difference for me if it was connected or not. To try and rule this out I connected it to ground via a 220R resistor but it still happened.

Alibro, I'm not too far from you I believe and I have an outlander charger on the shelf and a savvycan setup, so if you can't get the issue figured out soon it could be an option to test with the spare charger to rule out a damaged unit you might have. Sounds like it could be the control side hopefully

That's very kind of you thanks. I'll keep working on it for a while and get back to you.
While it is much easier to blame a faulty unit I suspect the issue will be something I have done.

[Alibro]

The plot thickens!

First I want to thank you guys for your help and support with this but I think the charger is OK and not the problem.
Today I started a charge and a CAN Dump with SavvyCan, around 20 minutes later I climbed into the car and tried something which caused the car to cut out a couple of days ago while driving. I turned on the headlights.
Immediately I heard what sounded like the clunk of a contactor and the charge stopped, the coolant pump briefly changed tone but then went back to normal, I didn't see the change in low voltage but if a contactor was turned off briefly then this makes sense .
I tried turning the headlights off and on again to see what would happen but nothing so this is intermittent.

I have attached the CAN Dump anyway but I think I'm looking for a stray voltage, a poor ground or a bad connection.

I feel a redesign of my LV connection box coming on as the current tangled mass of wires makes diagnosing faults extremely difficult.

[Bigpie]

Contactors are opening

You can see the voltage suddenly jumps to the max, the charger errors out.

[tom91]

Your DCDC seems to be fine, you clearly have some very dodgy wiring/12V power faults.

[Alibro]

Yes, agree 100% with you guys, thanks for showing what is happening.

I have ordered a couple of 12V fuse blocks to try and tidy up my wiring.
https://ebay.us/WPSy1N

In the meantime I'll try to figure out what is happening.

[Bigpie]

Your 0x285 message looks funky as too, but it seems to work. Wiki suggests the 3rd byte should be 0xB6 to pull the EVSE into the correct state. Yours is 0xAA. Not sure if this is a full log, but some public EVSEs wont work if you skips state B and jump straight to C. The Pod Points at Tesco for example

[Alibro]

Your 0x285 message looks funky as too, but it seems to work. Wiki suggests the 3rd byte should be 0xB6 to pull the EVSE into the correct state. Yours is 0xAA. Not sure if this is a full log, but some public EVSEs wont work if you skips state B and jump straight to C. The Pod Points at Tesco for example

Funny you should mention PodPoint as I tried to charge there and it didn't work, when I get the cutting out issue sorted I'll look again at this thanks.

Talking about cutting out I discovered something big today.
I wanted to see which contactor was dropping when I turned on the lights, and today it happened every time whether I was charging or not.
I only needed to wait for a few minutes after turning the car on if I turned on the lights it happened.
So I put a meter across the neg contactor and a second across the pos then turned on the lights and discovered it is the pos contactor dropping. The thing is it's not coming back on as I thought. This confused me for a bit as the dc to dc was still operating and I still had 14V which was why I presumed it was bouncing off and on.
Turns out the precharge relay is still active and was keeping HV to the dc/dc through the precharge resistor and a very light cable.
I think that's another thing to change.

[Alibro]

I was about to say
"It looks like the issue is a faulty contactor."
I put a meter across the 12V supply for the positive contactor and when I turn on the headlights the contactor opens and stays open even though the 12V to it doesn't change."

However it wasn't quite as simple as this.
I used a different multimeter to watch the voltage and noticed it flicker a tiny bit so decided to try changing the supply source for the contactor from the ignition switch to direct to the battery and this helped to the point where I thought I had it sorted. I was about to put the lids back on when I did one last flash of the headlights and it failed again.
So after this I tried flashing the lights with the contactor already stuck open and noticed a faint sound of a relay clicking so I new something else was going on.
I got my daughter to mess with the lights while I watched what was happening and It turned out that the timer module I am using to drive the positive contactor (from the switched neg supplied by the VCU) didn't like the fact I had several other devices using it's positive wire and when I flashed the lights the relay was sometimes flicking off. The thing is it was so briefly it caused the contactor to open and lock open.

So after more time running individual wires to the various devices involved I tested it again and couldn't get it to fail. I'll try again tomorrow but hopefully the issue is resolved but even better it may have resolved my cutting out problem while driving.

Once again a huge thank you for the help in tracking down this issue. I don't know for sure if it is resolved yet but at least I think I know where the problem is.

[arber333]

I put a meter across the 12V supply for the positive contactor and when I turn on the headlights the contactor opens and stays open even though the 12V to it doesn't change."

You could try running contactor from negative side. Supply 12V from ignition directly from ignition and use timer relay to bring GND to contactor. Maybe GND is more stable...

[Bigpie]

I'd consider binning off the timer relays, just use the VCU directly to do the pre-charge sequence. Timed pre-charge isn't the best method, if precharge fails, blown resistor for example, then boom, it'll just switch on the main contactor regardless

[Alibro]

You could try running contactor from negative side. Supply 12V directly from ignition and use timer relay to bring GND to contactor. Maybe GND is more stable...

That's a great suggestion, however I'm using the neg output from the VCU to power the timer relay. I tried connecting the VCU directly to the contactor and it didn't work. I think the VCU didn't like seeing the 12V pos from the other side of the relay coils. How do you guys use the neg outputs from a VCU to drive contactors?

I'd consider binning off the timer relays, just use the VCU directly to do the pre-charge sequence. Timed pre-charge isn't the best method, if precharge fails, blown resistor for example, then boom, it'll just switch on the main contactor regardless

The timer relay is there for two reasons. The reason above (ie how to drive a contactor from a VCU) plus it's a kind of fail safe.
A few months ago I had the timer relay turned down to 0 seconds but I believe the VCU had hung with the post precharge outputs active (the VCU would hang sometimes if I had the laptop connected to it while it was in the car) so when I turned on the ignition it activated the pos contactor immediately and blew a 300A fuse. I spent a while going through what had happened and double checked the connections and this was the only answer I could come up with. My VCU is the one Damien made specifically for the Gen1 Leaf inverter. There are not many here are using it so I don't know if others have found wee quirks with it.

The other thing I considered was putting a capacitor across the power connectors of the timer module but I don't know how big it would need to be to make a difference. Is this worth considering?

[Bigpie]

The board uses an NCV8401, designed for the job, 12V goes to one side of the contactor coil and the other to the VCU.
I see it's written in Arduino framework, I think they restart when serial is connected, but this shouldn't be much of an issue, if the DC bus has been pre-charged and it restarts the bus voltage doesn't drop that fast.

[andybpowell]

I think you're going about it the wrong way implementing workarounds and hacks when you really need to rip out the timers and other workarounds and find out what the issue actually is re-do the the 12v and gnd cabling to the controller and contactors and start again with a known good supply otherwise you'll just be going around in circles and getting frustrated

[Alibro]

The board uses an NCV8401, designed for the job, 12V goes to one side of the contactor coil and the other to the VCU.

Thanks Jamie, this is good to know, the issue I had with driving the contactors was very early in the evolution of my LV connection box so it is very possible I had wires crossed somewhere. I can give this another go as I am not that keen on using $2 modules to control the pos contactor.

I see it's written in Arduino framework, I think they restart when serial is connected, but this shouldn't be much of an issue, if the DC bus has been pre-charged and it restarts the bus voltage doesn't drop that fast.

Around the time the fuse blew I was using my laptop to watch the pack voltage through the serial monitor so was often plugging and unplugging the USB cable.
At the same time I was having random issues where the relays and contactors would not power up when the ignition was turned on.
Turning the ignition off and on again was not enough so I had to disconnect the battery.
I also noticed a couple of times that when I turned the ignition off and on again the relays and contactors seemed to come on instantly instead of in sequence. I think I got away with this on those occasions because the inverter still had residual voltage but the time I blew the fuse the car had been off for 30 minutes or more.
These issues only happened after I disconnected the USB cable from the VCU and have not happened again since I got the voltage showing on the WiFi, as I no longer need to connect to it.

I think you're going about it the wrong way implementing workarounds and hacks when you really need to rip out the timers and other workarounds and find out what the issue actually is re-do the the 12v and gnd cabling to the controller and contactors and start again with a known good supply otherwise you'll just be going around in circles and getting frustrated

I agree 100% which is why I have ordered the fuse blocks and intend to redo the LV connection box as a priority. Tracing anything through the tangle of spaghetti is not easy.
When I started this I hadn't a clue what connections I would need or how the final layout should look so it just evolved. At first I had it reasonably tidy but as things got changed and added it got a bit out of hand.

[arber333]

I agree 100% which is why I have ordered the fuse blocks and intend to redo the LV connection box as a priority. Tracing anything through the tangle of spaghetti is not easy.

First i would advise you to make a latching relay circuit. The fact you are using timed relay to power your contactor can be tricky.
I allways want to have ability to latch contactor so that no mattery what VCU does HV wont be an issue.

You can do this by using two automotive SP relays (or single multipole) and wire them to receive power from ignition side with trigger pulse from timed relay. That way you will have control to unlatch the pair simply by removing ignition. This would be a stop gap measure to use the car interim until fuse block come...

This is how they look:
viewtopic.php?p=41189#p41189

[Alibro]

First i would advise you to make a latching relay circuit. The fact you are using timed relay to power your contactor can be tricky.
I allways want to have ability to latch contactor so that no mattery what VCU does HV wont be an issue.

You can do this by using two automotive SP relays (or single multipole) and wire them to receive power from ignition side with trigger pulse from timed relay. That way you will have control to unlatch the pair simply by removing ignition. This would be a stop gap measure to use the car interim until fuse block come...

This is how they look:
viewtopic.php?p=41189#p41189

That's a brilliant suggestion, I still like the idea of using the timer (driven by the VCU) as a fail safe (Initially anyway) but by using a good quality 12V relay the timer would not be responsible for holding the contactor closed.
In the meantime I've tested more in the garage and had the car out on the road and so far all is well.

Apologies guys I've kinda hijacked this thread with what started out as Mitsubishi Phev charger questions but evolved into generic LV connector box stuff. A lot of what was said here should really be in the Gen1 VCU thread so if you guys don't mind I'm going to copy much of what you have said there.

I cannot thank you guys enough for your help, the cutting out issue was getting me down but I guess I was lucky to see the same issue when charging.

[Alibro]

I'm very happy to report back that the charging issue has not reoccurred since rewiring 12V supply to the Positive contactor.
I have also connected it through a latching relay as suggested by Arber333.
And the car has not cut out while driving either so I guess the two issues were caused by the same fault.
The latching relay is being triggered by the timer module but is held closed by 12V directly from the ignition switch so even if the timer module glitches it will not cause the positive contactor to drop out.
Thanks guys, I owe you a pint.

Now I have the car driving and charging at home reliably and to get this thread back on course I need to figure out why it is not charging at a Tesco Podpoint. I haven't looked at the sketch on the Due yet but have fitted the 220R resistor I mentioned earlier.
Jamie mentioned earlier that message 0x285 looked odd so I'll take a look at the sketch I used and compare it to the Wiki.

Automating charge stopping would be nice too but without a BMS or shunt I'm relying on the charger or Inverter to indicate Voltage.
Maybe getting both BMS and shunt working should be higher up the to-do list.

[tom91]

without a BMS or shunt I

You need to have a BMS before continuing. Your build is quite a hairbrained experiment that can turn deadly quickly.

This whole chasing of faults only to realise your build has a massive wiring and reliability problem show this.

[Alibro]

You need to have a BMS before continuing. Your build is quite a hairbrained experiment that can turn deadly quickly.

This whole chasing of faults only to realise your build has a massive wiring and reliability problem show this.

The BMS will be installed soon, most of the wiring is in place already, I had it connected before the batteries were installed in the car and they were well balanced then.
They were new unused batteries so the chances of them going wildly out of balance after a few months are slim.

[Alibro]

You need to have a BMS before continuing. Your build is quite a hairbrained experiment that can turn deadly quickly.

This whole chasing of faults only to realise your build has a massive wiring and reliability problem show this.

Sorry mate I couldn't stop thinking about what you said and had to respond further. I had one small issue with the feed to a timer module and you come out with this

I know you are keen on BMS's but I think what you meant is I need to have a properly installed good quality BMS that is constantly or at least frequently monitored.
I wonder what percentage of DIY EV builders have proper fusing on every BMS wire, or are using the correct cable or have them correctly routed and secured or monitor the cell voltages on a regular basis? I suspect not that many so to give a blanket statement than any EV build that does not have a BMS is inherently more dangerous or in your words, hairbrained is nonsense.
My experience of BMS's is mostly from Ebike batteries where they often are the cause of issues rather than preventing them
Off the top of my head I know of two people who have had major safety issues caused by problems directly connected to their BMS one of which caused a fire almost destroying his car and the other killed battery modules causing months of headaches for his build. Had they not bothered with a BMS then neither issues would have happened.
Also I have probably spent over a hundred hours over the last three years researching and catering for a BMS in my build which I had always intended to complete once the car was up and running so I really don't appreciate your comment.

Maybe you should think before being so insulting on this forum

[tom91]

I had one small issue with the feed to a timer module

I am sorry I would most definitely class this as a small issue. The wiring and electrical approach taken in this build is what I call hairbrained, as in there are wires and relays every where. I am not on about a BMS, if you want to be the BMS that is on you as it is risk you can decide to take either way (BMS or no BMS)

12V next to Hv wiring ect, and we end up doing a wild goose chase on two issues which are likely the biggest ones you are currently finding.

[arber333]

I wonder what percentage of DIY EV builders have proper fusing on every BMS wire, or are using the correct cable or have them correctly routed and secured or monitor the cell voltages on a regular basis?
....
Had they not bothered with a BMS then neither issues would have happened.

I wouldnt go that far with "no BMS" argument... It doesnt show any meaningfull data other than you havent had an incident...yet.
It all goes well untill it doesnt, or when cells go out of balance inside your box due to different temperature or resistance etc... could happen to anyone. Having BMS hardwired to your ignition/charge logic will prevent you from driving off/charging with battery issue.

I had 100% success rating on my Kokam cells only using voltage monitor, no balancing. BUT first i needed to determine the voltage window to drive/charge within. Cells would run happily in that range with total confidence. I would change them up to 4.05V and discharge not lower than 3.2V. Cell difference would be quite high dV being even 0.05V! But this happened each time and i had cells under control.
I had set it up as hard limits. If i would breach them it would mean lengthy slow charge and balancing on the top.

This also meant i had lower SOC to use for my drive though and cell voltage graph looked like saw teeth.

About fuses you are correct. Those JK BMS fuses on the BMS side dont mean protection for the BMS cables. Usually problems happen on the thin cables.
Just yesterday i was repairing a BMS wiring in one small car. JK BMS was not letting the car start. I traced issue to a faulty wire in the middle of the pack. Problem was crimped wire...

[EV_Builder]

Most of the BMS issues are with FET based BMS's. Most EV ones are relais controlled ones.

Not using an BMS in any build IMHO shouldn't be safer if it is you got the wrong BMS .

[Alibro]

I don't think Johu would appreciate this thread being taken over with discussion of BMS's. For the record I agree they are important and plan to implement one soon so maybe we should get back on topic.

On the subject of Outlander chargers I called into a Tesco carpark last night and plugged in. After around 20 minutes I had added 1kW to my battery pack so it seems to be working. The only thing I changed from the previous failed attempt was reinstating the 220R resistor between PP and ground however I was at a different model of Pod Point so I still need to try again at the charger that failed last time. I will do that some time this week to confirm if other Pod Point chargers are happy with the CAN messages my Due is sending.