The Troubleshooting and Repair for On-board Charger (OBC) Thread

Mitsubishi i-MiEV Forum

Help Support Mitsubishi i-MiEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
@ Kiev,

I am sorry you could not see the image (a photo of my scretch, 4MB size) in that link. I uploaded it again and Wetransfer generates this link that will be active for one week:
https://we.tl/t-X76eNfvwQD

Yes the intermitted faults are difficult to trace.
Just now it is again under charging process at 14A and the 3 voltages are stable.... :cry:
 
Hello everyone,

I’m joining this community with my broken OBC that is part of a Peugeot iOn from 2011. Thanks to the good documentation and large amount of information in this thread I was able to fix the blown snubber caps as well as the 20 A-fuse and, as a consequence, to get rid of the “red battery”- and “yellow car with exclamation mark”-warning lights on dash. My auxiliary battery is getting charged again jumping from 12.7 V to 14.4 V when the car is switched on and the “READY”-symbol appears. Also, the 20 A-fuse wasn't blown again when the car was started. So, I have the feeling to already have done a good part of the road to get the car completely fixed again and I can’t thank enough all the contributors to this thread who made this become possible.

Now I’m more or less at the same point where skylogger was in July 2018:
I ended up [...] replacing the two caps in the doghouse filter section, re-assemblied the whole thing and put back in the car. We replaced the 20amp fuse, filled up the coolant tank, and plugged back in the safety plug and connected up the 12v battery. Connected up the MUT III Analyser, and had two errors being reported concerning OBC and DC-DC. Ran the option to clear errors and re-test, and all errors were cleared. Checked the 12v battery before turning car on and voltage was about 12.3v. After turning the key and getting "READY" The Battery ICON and the HV fault Icons were no longer showing on the dash, and the 12v battery was now reading 14.4V so it appears to be charging. Connected the charging cord to the car and attempted to charge, but could see the charge light on dash blink, then the HV Fault came on the dash, then both went out.
Just like skylogger describes, when I put the AC-charger-plug into the car, the ventilation starts instantly for 2 or 3 seconds and on dash the “red plug”-symbol starts blinking. After 5 seconds the “yellow car with the exclamation mark”- symbol lights up. After another second the “yellow car with the exclamation mark”- symbol and the “red plug”-symbol turn durably off.

Back to skylogger’s case: In August 2018 had finally fixed the issue.
So I decided to bit the bullet, and re-install the charger back into the car and test it out.
Once I got through fitting it, I turned the key and car went to "READY" with no errors on Dash. I checked the 12v battery and could see it was sitting at 14.4v so it looked like it was charging ok, particularly since it was sitting for a week without a charge. I connected up the MUT III and scanned for errors. It not only listed the OBC Timeout Error and DC-DC Error from the previous time I turned on the car without the charger installed, but it also listed LIN errors for all of the CMUs which was a bit weird. I re-ran the DTC Scan with erasing all the errors first, and all errors were cleared. Then for the moment of truth, I turned off the car, plugged in the charge cable, and the charge Light lit up on the dash, cooling fan came on. and I could also hear the coolant pump circulating coolant. It was 5 bars down when I started, and after 20 minutes I could see that the fuel gauge had moved up one bar. So all is looking great so far.

As far as I see, from the documentation of the problem resolution that was posted in the meantime, two things could have caused his on-board-charging first to fail and then to work again:

One was a twisted resistor, due to which a circuit was open.
I was looking over the section on the bottom PCB next to CN1 Connector. I spotted a resistor that was a bit diagonal and when I checked from the pin of the IC to the VIA I found it was open. one side was not making contact to the pads.
I haven’t found a part obviously misplaced in my charger, but I have to admit that I haven’t checked that very thoroughly. So, I may need to get back into it again...

The other one was bad communication between the MUT III-clone he used and OBC, so that stored diagnostic codes may not have been erased from there.
I ran some test with MUT III. I could look at Hardware and Software revisions on BMU, CMU, MMU, But the MUT III would not communicate with the OBC. I only just tried this on two other working IMIEVs and have the same problem, so it looks like my problem with communications between MUT III and OBC are with the MUT III and not really a fault with the OBC. I'm using a MUT III Clone from china, and the MUT III Second Edition software that I acquired from Russia. I think the MUT III talks to all the ECUs on the same CANBUS, but the CANBUS That connects to the Diagnostic socket on the car is a different bus than what connects to the OBC.
I don’t have access to a MUT III, so I need to have the deletion of the stored diagnostic codes done at the Peugeot dealership. I have had stored diagnostic codes erased once already after having it made to the situation described in the beginning of this post, but it didn’t change anything on the charging behaviour. I haven’t assisted to that intervention at the dealership. So, this part of the process is kind of a black box for me: I don’t know the tool and I don’t know the process. Is it possible that a dealership has a tool (some alternative to the MUT III) that is not completely compatible with the car, so it wouldn’t be able to communicate with the OBC to erase stored diagnostic codes there? And, as a consequence, the remaining error codes would just trigger again the charging failure? Or can it happen that the user of the MUT III-alternative only erases a part of the diagnostic codes that would need to be erased (e.g. only the codes stored in one location), even if he could have access to the diagnostic codes stored in the OBC? Are there actually diagnostic codes stored in the OBC?
skylogger doesn’t get into details on which tool (the same as before?) he used in what manner (the same as before?) to clear the error codes on his final problem resolution.

So, that’s where I stand and the questions I have on my mind. Every help and input will be appreciated.
Nico
 
Sandrosan said:
@ Kiev,

I am sorry you could not see the image (a photo of my scretch, 4MB size) in that link. I uploaded it again and Wetransfer generates this link that will be active for one week:
https://we.tl/t-X76eNfvwQD

Yes the intermitted faults are difficult to trace.
Just now it is again under charging process at 14A and the 3 voltages are stable.... :cry:

Hi Kiev and all others,
Just to let you know that I think and hope I have fixed the problem making 2 jump wires: one directely from the +5V of C703 to the relay coil, and the other one from collector of TR301 and the other side of the relay coil. As the voltage directely on the coil was unstable, while the same voltage but from the + 5V of C703 and the collector of TR301 was stable. This means there was a bad contact on ribbon flat cable / its connectors / PCB traks. In my opinion to design a circuit which uses such ribbon flat cable for flowing some 50 / 70 mA necessary to the relay coil for so many hours during all the charging process was a sort of crime. Now the voltage directely on the coil is happily stable. :D
But to get it, belive me it took many days of data logging while charging process.
 
@Sandrosan, That is great news to hear-- all your hard work has paid off. Thanks for sharing your results.

i agree that the flat ribbon cable with such tiny traces and connector terminals is a poor design. i've had problems with similar type ribbons in other devices.

For comparison, the Nissan Laef OBC is made by the same vendor but uses a much more rugged connector and wiring harness to connect between the upper control board and the lower power board.

PhPamHP.jpg
 
@Nico

Welcome to the forum and glad you found some helpful information.

There are other areas of the OBC which can or have failed, but the 2 that you fixed are the most common. A reading of the stored DTCs would help determine what is the remaining issue.

The dealer's scan tool (MUT, Lexia, or Diagbox?) has all the capabilities and features to read and clear errors in all the control units, including the OBC, which does have its own set of simple codes.

Before clearing any codes, it would be good to get a reading of whatever codes are stored in order to have some idea where or what the problem might be. The DTCs and the simple OBC codes would help pinpoint which circuit might have issues.

The iCarSoft i909 scan tool may be an option for you to a low cost tool to read and clear DTCs. It is made for Mitsubishi and i don't know if it would work for the iOn, maybe someone else has tried and could let us know.
 
Thanks for your welcome, Kenny, and for your quick response.

I found a statement in the speakev.com forum concerning the use of the iCarSoft i909 on the Peugeot iOn or the Citroen C-Zero:
The question asked was "Does the i909 work on the PSA pair of the triplets?"
No the i909 is Jap models only (Mazda and Mitsubishi). Despite it being the same car it refuses to work (probably vin range check or similar). I tried the PSA version of the icarsoft (i970) on an ion but it didnt work well, the dash lit up like a christmas tree and the car wasnt happy, didnt give any useful information. I dont know if it was trying to talk "Peugeot" to what is really a Mitsubishi. There are more advanced icarsoft models available, presume one of those would work better.

Ok, so I bury my hope that the ongoing issue was just due to an incomplete deletion of the diagnostic codes. Would have been too easy anyway... :roll:

Before replacing the snubber caps and the MCU fuse the scan of diagnostic codes in the EV_ECU at the dealership had brought up these ones:
- P0A09: DCDC converter (1)
- P1AF8: On board charging stop processing time-out
- U111D: On board charger CAN time-out
- U1113: "Ignore Fault code" or "Remote CAN T/O/ Not equip" (code isn't listed in the workshop manual so I took the interpretation from this post and this post)

Of course it would be much more interesting to have the new, fresh diagnostic codes, from after my intervention. I'll try to bring them here quickly. I'll see if I'll invest in a scan tool or if I'll have the codes read out at the garage around the corner. If someone could tell about a good experience with a scan tool on a Peugeot iOn, I would be interested to know more about that, particularly which tool was used, of course.
 
Without the codes it will be more difficult to pinpoint, i don't know if one of the phone apps thru an OBDII dongle will read DTCs if that is possible.

But how far into the sequence does the system go before shutting down? do you hear the relays and contactors clacking? Does charging attempt to start for a brief period, i.e. the EVSE relay closes and AC power is applied to the OBC? This might give some clues.

First of all, Did you do a diode check of the waffle plate junctions? If that is damaged then of course it won't work.

Were you careful when removing and reconnecting the flat ribbon cable between the control board and the power board? That is very critical and should be checked. What about the connector on the upper board, was that mated properly?

There are low voltage power supplies created on the upper control board

Did you check thru the AC input section of your OBC for damaged capacitors, the AC relay, or the ceramic resistors? Did you check the HV DC output fuse of the OBC? while we have never seen it blow there is always the first time. There are numerous components (resistors, capacitors, diodes) that can be measured while on the board to verify that they are okay

Otherwise it is possible to solder some jumper wires on the board and run them out of the box so you can verify that AC is getting to the waffle plate. Once that is confirmed then you can move thru the various sections.
 
I've been at the garage earlier today and they have read out the diagnostic codes that remain in the car's memory now. I just took some time to find out a bit more about the codes they had found. The scan tool they used there was a Bosch ESI[tronic] 2.0, and that is not the same scan tool as the one that was used at the dealership before I did the replacement of the snubber caps. This is one aspect why the two set of records I got after the scans look a bit different (including code descriptions - I assume that despite of this, both scan tools will have correctly read out the stored codes). Also it may be that one of the scan tools was more or less compatible with the car causing less or more communication errors that would be only related to the communication between the car and the scan tool used and not really being issues of the car itself. Last aspect that may have had an influence on differences between the two scans is that at the dealership (before the replacement of the snubber caps) I had pointed to the issue I was facing and so they only scanned the EV-ECU system for diagnostic codes. Consequently the record they handed to me after the scan, listed only errors in the EV-ECU system. The garage where I was today did a scan of the total system and so other diagnostic codes came up. But these would probably also have been there at the first scan already, if at the dealership they also would have scanned the whole system.

So, that's the result of the today's scan:

---- EV_ECU ----
- U1113: CAN-communication with ABS/ESP controller, CAN-communication disturbed/implausible signal
The description of the error comes from the record I got. It's not exactly the same description as I had found before. This description
looks like a problem being a bit more serious.

- U111D: On board charger CAN time-out
I would assume that this is the issue that blocks the charging process.

---- ESP ----
- U1104: Steering wheel sensor time-out, SAS CAN time-out/ Not equipped (SAS = steering angle sensor?)
If this is a comfort feature it's possible the car isn't equipped with it.

---- Instrumental panel ----
- U1116: communication with controller keyless access, KOS CAN time-out /not equipped
This sounds like a comfort feature to me that enables you to open and start the car without actively doing something with the key. The
car certainly isn't equipped with such a feature, so that may be an explanation why this code is there.


---- Compressor & heater ----
- B1081: Control panel, Communication error

---- Central electronics ----
- U1111: Communication with info display, CAN time-out

---- Immobiliser ----
- U1111: Communication with info display, CAN time-out


I'm quite surprised about most of the errors listed. I hadn't noticed malfunctions that could be associated to them...


Concerning the questions in your last post:
Yes, there are clacking noises coming from the box on the EVSE-cable. But I couldn't say if AC power is applied.

I did the diode check of the waffle plate junctions. The measures were good, it all looked as it should.

I tried to be careful with the flat ribbon cable, but not finding how to loosen it, and still under the impression of opening some other connectors around the OBC box that needed a screw driver and some force, I broke a little piece of plastic. It's doesn't seem to have played a major role and wasn't in direct contact with the leads (see the following picture, broken part highlighted in red, then a picture before and after breaking it - last picture has a poor contrast, sorry for that...), but as this cable is so sensitive, I definitely want to check it. Is there a method how to test the transmission through this cable? Or would it be easier to directly replace it by a new cable of the same spec?
h7hyIMf.jpg


Did you check thru the AC input section of your OBC for damaged capacitors, the AC relay, or the ceramic resistors?
These are the parts that are in the same so called doghouse as the usually blown snubber caps, right? I checked these as far as possible. The resistors show a resistance of 9.7 Ohm on my multimeter and when I'm applying the 5V to the Diode D301 I hear the relay clacking and the resistance drops to 0 Ohm. Reproducing the test, the relay seemed to react reliably. What I haven't tested, as I didn't want to solder it out of the circuit (the possibility to access it is quite poor with the waffle plate still in place), was the 2.2 uF X2 capacitor in the doghouse. The fuse close to the doghouse area is good, too.

I hope this additional information gives you some good hints. For me it's only guessing... Because of the diagnostic codes I'd suppose something in the CAN communication, but wouldn't be able to determine if it could rather be an open line (broken cable or not entirely closed connector) or an element of the CAN not being powered and not participating to the communication due to that. And even if this guess was right I wouldn't know where to start searching and testing. So thank you for your help and sharing your expertise!
 
Concerning tiny connectors: FRC Connector Guide

All the "U" codes are CAN Buss related, and if real, could be related to loose connectors or terminals, pinched wire, etc.

But to have so many is more likely a weak 12V auxillary battery issue. A weak, old or worn out aux is trouble.

The black plastic in the ribbon connector is for a holding clip and likely not an issue, although i have a failed OBC from Sebastian that has intermittent faults with the LV power supply from the control board to the power board. And i now see that the clip of his ribbon connector is also quite broken--so maybe just that little bit of pressure is enough to throw off some of the signals across the cable? With a meter on continuity setting you could ring it out from the board to board using the little gold contacts on the back of the off-white connector body. This is done with car OFF and with the small CN101 connector disconnected (so no voltage on the boards). If all the pins show continuity from board to board, then the cable and connector is likely okay.

Any OBC fault will set codes both within itself and also sent to the EV-ECU, which is the master controller and keeper of all the faults that would prevent driving or charging. So reading just that is fine. i somewhat have doubts that the Bosch tool will read the DTCs of your EV-- Mitsubishi EV has its own set of codes and connection protocol not standard OBDII.
 
kiev said:
@Sandrosan, That is great news to hear-- all your hard work has paid off. Thanks for sharing your results.

i agree that the flat ribbon cable with such tiny traces and connector terminals is a poor design. i've had problems with similar type ribbons in other devices.

For comparison, the Nissan Laef OBC is made by the same vendor but uses a much more rugged connector and wiring harness to connect between the upper control board and the lower power board.

PhPamHP.jpg

Thanks Kiev,
Yes that connector in the OBC of Leaf is much more robust!!! I think iMiev has been a project designed too quickly to be the first EV car for mass production to get the prize as first EV car. Then with experience they improved it. Just really pity that Mitsubishi has decided to give it up to this project even without supply replacement of the battery pack with larger capacity as nowaday the technology offers. Probably the all group decided to focus on Leaf internationally and now on Renult Zoe, Twigo EV, Dacia Spring in EU at least.
We should write a very nice letter to Mistu to release the schematic diagrams and open the firmware and software (accpeting no responsability for any troubles eventuaally occurred) considering they abandoned this project, in this way who has the knowledge could upgrade the batt pack and easily repair this car nd the other 2 sisters. I have been working for a Japanese company and I could ask my collegues to pass this kind of nice letter in a good English, (not mine :) ; and also well translated Japanese... Playing violin sometimes we open the heart...
 
That is a great idea, i would be happy to collaborate with you on such a request letter.

It is such a pity that they gave up on such innovation. i'm not comfortable with the attitude that everything is just disposable electronics--when it breaks or gets old, just throw it in the trash...
 
Thanks for the link. I was better prepared facing the situation again today. I tested continuity from the little gold contacts of the connector’s base from the control board to the little gold contacts of the connector’s base from the power board. They are so small that’s it’s hard to say if you haven’t missed one or if you have touched more than one contact at a time with the probes. I did as well as I could and would say that cable and connectors are ok. It seems that many of the 50 leads of the connector are carrying parallel signals. I was expecting that one contact on one connector would show continuity to only one contact on the other connector. Instead of that, I got continuity signal for the 50 contacts of the second connector, by just touching 7 or 8 different contacts of the first connector. Example with random figures: touching contact 1 on connector A, I got continuity to contacts 1, 2, 3, 4 and 5 of connector B. I also got continuity to these contacts if I was touching contact 2, 3, 4 or 5 of connector A. Touching contact 6 on connector A, I got continuity to contacts 6, 7 and 8 of connector B. I also got continuity to these contacts if I was touching contact 7 or 8 of connector A. And so on...

After having done this test on the flat ribbon cable, I took the control board completely out of the box to have a closer look at a place that I had already spotted when I replaced the snubber caps. It’s the place of the control board that was just above the popped snubber caps (bottom side of the control board, facing the snubber caps). Here are two pictures. The second one is an enlargement of the area marked by the red square in the first picture.
DxJpCxE.jpg

dSofYXQ.jpg

There were black traces from the caps popping that I cleaned and after cleaning it, it didn’t look so bad. So, I had thought that it was worth trying if it was still working once I would have had repaired the obvious issue: the snubber caps. Now, as this is the most obvious eventual damage on the board, I guess it can’t be wrong to check the elements in this area next. Does someone know the resistance values for the resistors R720 and R721 and how I can test if the circuit between them and their neighbours in the circuit isn’t open?

Concerning the numerous “U” codes: I had an auxiliary battery that was probably in pretty bad shape, even if it hadn’t shown any visible sign of weakness. It’s very likely that it was the battery that the car got when it left the factory, so it would have been close to 10 years old. Reading through this thread I realised that replacing it could be a good idea. Unfortunately, I had no idea before that I was taking the risk to severely damage the car’s electronics by having a worn out 12 V battery... The battery that is now in the car is a completely new one. My first guess would be that these diagnostic codes are from the time before the battery change and that only the codes in the EV-ECU system have been deleted when I was at the Peugeot dealership. Quick chronology to make things clear:
1) Issue with “red battery” and “yellow car with !” symbols on the dash. 12 V battery not charging when car “ready”. Main battery not charging from AC. Went to Peugeot dealership and they got the diagnostic codes from EV-ECU.
2) Snubber caps repair and 12 V batter replacement. No warning symbols on dash any more and 12 V battery getting charged again. AC charging still not working. Hoping that AC charging failure was just due to some code remained in the system, I went back to the dealership and asked for deletion of diagnostic codes. On this point I have to speculate, but I’m pretty sure that only codes from EV-ECU were erased.
3) AC charging still not working. Asked for a new scan of diagnostic codes at a garage in my neighbourhood. There, a global scan was performed. That’s the scan with the many “U” codes.

I would suppose that if at (1) the scan had been global, the “U” codes from the other systems than EV-ECU would already have been found and if at (2) the codes of the global system would have been erased (and not only the EV-ECU codes), the “U” codes from the other systems wouldn’t have shown up at the scan at (3), as, with the new battery, the failures wouldn’t have occurred again.

I have searched some information about the Bosch tool used for the last scan. It’s a multi-brand scan tool working for 90 000 car models from 150 brands. It’s based on an online knowledge base. At the workshop the technician enters the VIN of the car and then, out of this online back-end, the front-end tool gets the information about how to deal with the car (connection protocol, available systems, ...). I’m not familiar at all with the diagnostic codes. Would you have expected something else than the codes that were on the record?

Even if my electronics skills probably won’t be enough to allow me to take advantage of published diagrams and opened firmware and software for the iMiEV, I fully support your letter initiative, sandrosan and kiev. The extractive way of life is not leading into a good future and we should be able and enabled to repair the products we buy and to make them last even beyond the time when the OEM has abandoned them. If you also want a French version of the letter to send another copy to Renault (as part of the Renault–Nissan–Mitsubishi Alliance) or to PSA Group (for the two sisters) I could participate on that.
 
R720 is a 10kΩ, R721 is a 1kΩ. When i measured resistance across these i got the 1k for R721, but i measured 5k across R720, which is in the bottom leg of a voltage divider connected to the ground plane. i suspect it is reading thru the IC509 and a parallel path to ground thru R576, which is also a 10kΩ.

Some of the ground and power signals are carried across on multiple wires on the ribbon cable. It looks like you ribbon is okay. It requires a very fine pointed tip on the meter leads and a magnifying glass to probe on these tiny circuits.

The DTCs of most interest will be of the "P" code types, e.g. P1A12 On Board Charrger Abnormal Stop. With the OEM tool at the dealer they can read the OBC codes for the specific fault. The OBC codes are simple 2-digit number codes such as 04-Control power supply voltage abnormal, or 27-AC Input current abnormal. The entire list is in the last post on the first page of this thread, aka Step 3: in the troubleshooting links post.

If the EV-ECU has a CAN buss problem with the OBC, it throws U111D.

So when trying to do AC charging, the process gets interrupted or fails to start--does the dash show any lights or symbols when this happens?
 
Thanks for the reference values of the resistors, Kenny.

It seems we're on a promising spot there... I measured the resistance through the 3 resistors and all are way off of what they should be:
R721: 167 Ω (should measure 1 kΩ)
R720: 17 Ω (should measure 5 kΩ)
R576: 15 Ω (should measure 5 kΩ)
I don’t know if something could happen that reduces the resistance of a resistor. I’d rather suppose that if a resistor is damaged it goes open circuit. But this would mean that somehow there must be something acting as additional resistors in parallel to the checked resistors. That raised my interest in the IC509.

Turning the board in all directions and looking at it from different perspectives I noticed that a pin of the IC509 seems to be broken: the pin next to the resistor R720. Not easy to see, even with magnifying glass. I thought until now, that it was just due to some reflection of light on the solder that it didn’t exactly look like the pins next to it (see 2nd picture in my last post or picture later in this post). But actually there’s a real interruption between the solder pad and the entry of the pin into the IC plastic body. The measured resistance between the solder pad and the part of the pin that sticks out of the IC’s body is 8.9 kΩ. Finding some helpful information in this post, I identified the IC as being a MB3793-45PNF. The broken lead is the one of the VCC pin (bin number 5). I first thought to try to just bridge the break to see if this would influence the resistance values in a positive way. But thinking about it, I found no physical reason why this should be the case. What seems more likely is that the IC is more severely damaged and is the troublemaker in this circuit.

I’ve measured some resistances from pin to pin for comparison. On the broken pin I measured from the solder pad (“VCC solder pad”) as well as from the little piece of lead coming out of the IC’s body (“VCC rest”). I added the pin numbers in brackets accordingly to the datasheet.
VCC solder pad (5) – VCC rest (5): 8.91 kΩ
VCC solder pad (5) – Ground (4): 3.7 Ω
VCC rest (5) – Ground (4): 8.91 kΩ
Ground (4) – Clock1 (8) and Ground (4) – Clock2 (7): 16 Ω
Ground (4) – INH (6): 18 Ω
VCC solder pad (5) – Reset (1) and VCC rest (5) – Reset (1) and Ground (4) – Reset (1): 39 MΩ
VCC rest (5) – Clock1 (8) and VCC rest (5) – Clock2 (7) and VCC rest (5) – INH (6): 8.91 kΩ

I also measured a resistance of 3.4 Ω between R720 (at the side with the “R720” inscription) and the C545 (at the side being under the “4” of the inscription “C545”).

For some other measurements I added the values measured in the following picture, as it was hard to describe for some, from where to where I had taken the measure. The colours of the lines and measures I added don’t have a meaning. It was just for better differentiation. Opening the picture in an new tab will allow to significantly enlarge it.
The first picture shows the concerned area of the control board as it should look like without being damaged (taken from another picture on this forum). The second picture is a picture of the same area on my board and the third is the same as the second but with the measured values added.
SAY7zM9.jpg


Without knowing the values I should have measured for these connections, every possible explanation is quite speculative. Very suspicious is this little resistance between the VCC solder pad and Ground... What may have happened to damage the control board in this manner, is that when the capacitor of the power board popped a high voltage has been searching its way to Ground, creating high discharges to the 3 points of the control board where the green coating (and the copper below) has been removed (two of these impacts are next to R720) and to the N00-connector (see pictures below, connector NOO being in the bottom part of the 1st and on the left side of the 2nd picture, that shows it in comparison to L00 connector). That area around the N00-connector was also pretty black before I cleaned it.
BEFfuvP.jpg

6menkrA.jpg


I guess that it’s very likely that IC509 is severely damaged and needs to be replaced. Concerning the resistors I’m not sure yet – if the discharge looked for a way towards ground, there’s no reason why it should have taken the way through one of the resistors, but the impact may have damaged at least two of the three mechanically. Also, I don’t know what to think about the low resistance between the solder pad of the VCC entry of the IC509 and Ground. With only 3. 7 Ω the resistance is very low and can’t be explained by what happens inside the IC as the lead is broken and the VCC solder pad has no contact to the IC (at least not through the VCC lead). Could it be a damage in the PCB itself? And if it is, how to find and repair it?

I’ve spent a couple of hours looking for where to buy a replacement for IC MB3793-45PNF. The MB3793-45PNF is obsolete since 2018 and none of the usual vendors does still sell it. So, I had been thinking about taking a very similar brother of it (the MB3793-45DPNF – the “D” in the part number stands for “open Drain”) and adding a pull up resistor (between VCC and Reset pin) to give him the same functional behaviour as the MB3793-45PNF has, with a CMOS logic implemented. A 100k resistor could be used as pull-up resistor in such a circuit. Should be working, no?

Ok thanks for clarifying the question about the codes. I probably should go to the dealership again for the next reading of DTCs. Or better would be to find a tool that would enable me to do the reading by myself, but I still don’t know what tool would work for certain on the iOn.

Concerning the charge process I had done a rough description of what happens in this post:
when I put the AC-charger-plug into the car, the ventilation starts instantly for 2 or 3 seconds and on dash the “red plug”-symbol starts blinking. After 5 seconds the “yellow car with the exclamation mark”- symbol lights up. After another second the “yellow car with the exclamation mark”- symbol and the “red plug”-symbol turn durably off.
I also have the following two short videos for further illustration. In the first you can see what happens on dash. On the second what happens on the EVSE. Unfortunately, I haven’t filmed the EVSE again after the interruption of charging. From memory I can’t say if the “fault”-light lights up or if just the “charge”-light goes off when the process stops. As far as I remember the “red plug”-dash-symbol’s normal behaviour is to be continuously on during charge and not to be blinking.
Video 1: https://i.imgur.com/eONlndi.mp4
Video 2: https://i.imgur.com/AXHYQEz.mp4
 
Great pictures, you have done some excellent board inspection and troubleshooting, and i think you have found the area that could be causing your issue.

Notice that there is a ball of metal on the pin 5 Vcc of the 3793; also that the ground traces have been vaporized around the vias. The molten metal ball may have been expelled by the snubber cap explosion and other metal splatter may have shorted or burned the traces and via lands. That's a good reason to wear eye protection when working around high energy boards and circuits. Elden accidentally shorted HV when probing on his DCDC board and that melted traces and blew parts off the board.

There may be subsurface damage to interior layers; inspect for discoloration and burned traces around the capacitors on the opposite side of the board above the 3793, and check their value or that they are not shorted. Your low voltage power supplies of the control board are suspect considering the splatter damage, e.g. the 5v supply may have been shorted to ground and been damaged.

Those resistors R720 and R576 are 10k and should measure that value off the board (and when you order some for repairs), not 5k. Same with R721, when you measure on the board there are sneak paths that are causing it not to read the right value. De-solder one end to open the circuit if you want to get a reading. Also you can use tweezers or a fine sewing needle to wiggle and lift the Vcc pin to see if it is a broken solder joint

i think the 3793 on our boards was made by Fujitsu in Japan, and possibly other vendors, so there may be other sources. i suspect yours is blown, and should be replaced as well as all the passive components connected around it.

After seeing this now i'm thinking it might be a good idea to install some sort of plastic splatter shield over the bottom side of the control board above the snubber caps.

To repair the board i would remove the ic and all the passive parts; the burnt traces and vias can be rebuilt using some 30AWG wire such as used for wire-wrap circuits. The solder mask can be scraped with an exacto blade to reveal bare copper on traces and lands, then the wire can be soldered and routed to replace the trace.

With the damaged parts removed, then the LV power supplies should be checked and verified before trying to repair the control board. If those are damaged then it would be easier to look for a used board to just replace yours. If the LV supplies are good, then replace the parts and it should work. No need to visit the dealer for any codes until this is all sorted out, and when fixed and working there will be no codes.
 
I checked the top side of the board and particularly its area above the damaged area on the bottom side. I found no signs of damage or short or other suspicious measurements. I’ve taken off of the board the IC509 and the C545. I was suspecting both of being responsible for the low resistance between 5V-Vcc and Ground. But my first measures after retracting them don’t seem to clearly confirm it. I’m still measuring only around 5-6 Ω between VCC and Ground. I can’t see how the 5 V-circuit gets to that area to supply the IC509. I only can follow the trace through the VIA to C704 and C707, then I’m losing the trace. But actually, I don’t know if following the 5 V-trace further into this direction would really lead to something. I still believe that the cause for the low resistance between VCC and ground is to be found in the damaged area.

I’ll have to replace my C545. Do you know the specifications of this capacitor?

Having taken off the IC509 and opened the circuit, R576 now measures 9.86 kΩ and R721 1 kΩ, so they both seem to be good. R720 will need to be replaced (measured 1.18 kΩ but should show around 10 kΩ, like R576 as they are in parallel). I know that I still haven’t taken into account other resistors that might be in parallel on some sneak paths. But as the values are close to the nominal value and as my SMD unsoldering skills and equipment aren’t that good, I’ll assume that the R576 and the R721 are ok and won’t touch them.

Here’s a new picture illustrating the latest state of the board.
h9pWguh.jpg


Yes, the 3793 was introduced by Fujitsu. But later on, Fujitsu sold the Microcontroller and Analog Business of Fujitsu Semiconductor to Spansion that later merged with Cypress Semiconductors keeping the name of Cypress Semiconductors. I looked that up on Wikipedia, as I hoped to find the IC more easily by knowing its history. The original IC with the negating Output (low when active) is now only sold by companies that are specialized in obsolete parts and usually trade big quantities in the B2B market. The best option I found for the original IC (MB3793-45PNF) was at WinSource Electronics, where they offer the minimal order quantity of 8 at the price of 6.25 $ each. So that makes 50 $ and the taxes and freight are on top. That’s why I’m thinking about to rather use a MB3793-45DPNF (less than 2 $ at Arrow Electronics) in combination with a pull-up resistor. MB3793-45DPNF is also hard to find. All other well-known B2C online electronic stores didn’t have it...

Many of the VIAs in the Ground part of the board seem to not be connected to anything on the other side of the board. What about the VIAs in the area marked in red in the picture below. Will it be important to rebuild a good connection at the place marked with “2”, so that the two Ground fields joined by the blue line have good connection again? Currently the Ground signal must transit through the VIA at the left in this red marked field. Does VIA “1” need the Ground signal to supply it to a component on the other side? There is a component sitting on the VIA on the other side, but I don’t know if it uses and is connected to the Ground signal coming from the bottom side of the board. Same is for VIA “3”. If Ground signal doesn’t need to be supplied, that would reduce reparation work...
F4KsZIJ.jpg


Well, many questions... I’ll continue to track the reason for the low resistance between VCC and Ground first. All other repair thoughts aren’t worth anything, if I can’t figure out to what it is due...
 
i can't tell from your photo, but one end of R721 should have a trace that connects to R720, the other end should connect to the ground plane. Are these both okay? Is the connection to the ground plane intact or damaged?

There are 2 12V supplies to the board, one is HATT (Hot All The Time) and the other is only active when Switched ON by the EV-ECU. And from these 2 are created several low voltage power supplies on the board, such that there are 2 5V supplies, one is HATT and the other is Switched. Your Vcc on IC509 is from the Switched 5V supply that is created by IC707 (PWM Buck Regulator) from the Switched 12V supply.

The 5V for Vcc comes thru the via at the end of the trace from C545--it's near the ground via that you labelled as "2". There appears to be discoloration around that 5V via as if maybe the ground plane and 5V via have melted together but it could just be residue or the photo.

Measure for continuity from the Vcc pin or the via to the metal tab of transistor TR710. If you hold the board such that you can read the designator for IC509, then IC508 is to its right, then diode D718 to the right of that, and then TR710 is to the next right. The metal tab is an quick and easy place to find a connection to the Switched 5V.

Now you are left with calculatus eliminatus--find parts where the 5v and are shorted, remove them from the board and check for shorts to eliminate them from the list of culprits.

Soldering trick: add some solder to the joints of the device you want to remove to increase the thermal mass and help the solder stay liquid longer, then you can heat all the pins on one side at the same time and use a wooden toothpick to pry it up or tweezers to pick it up, maybe alternating from side to side to work the pins up. Then wipe your soldering iron tip clean (damp sponge or paper towel) and use it to wick the excess off the pins and board, or use regular copper solder wick to remove excess from the lands, pins, etc. When it comes time to put the parts back on, it is easier if most all the excess has been removed first.

TR710 can be tested by lifting just the 2 small legs and leave the big tab soldered to the board--that would be a tough one to remove. The top leg is soldered to the ground plane, so heat that joint and pry the leg up a bit and wick out the solder. Does the short go away?
 
Back
Top