kiev
Posts: 925
Joined: Sun May 03, 2015 7:15 am
Location: The Heart o' Dixie
Contact: Website

Bottom Board Trace Notes for OBC Schematic

Sun Jul 29, 2018 8:25 am

Here are my tracing notes on the two solder rails, left and right, of the bottom board. My plan is to replace all this mess with a pretty schematic drawing using Kicad. i didn't find a round solder pad or via in the components list, so i used these square solder pads on the rough draft drawing and labeled them with what is printed on the board. i've numbered them from 1-32 and 1-33 for my convenience as if they were connector pins. The designation top and bottom refers to components on the board or below it on the waffle plate. BT is the blade terminals soldered to the board for connecting to the magnetics.

Image

Here are some measurements of the inductor characteristics and a sketch of the final output filter before it leaves the box on it's way to the battery pack. These are marked as "zebra" transformers but i couldn't find any data sheet or info about these part numbers on their website.

Image

Rough draft Kicad sketch with measurement notes from the waffle plate (rectifier diode and FET diode drops). Toward the top center is a dashed box labelled Top with 2 capacitors--those are the output filter snubber caps blown in Jay and sky's potted doghouse.

Image

Here is a sketch of all the easy stuff i've traced that matches up with the OBC block diagram that Mits put in the technical information manual. This is the high-power analog stuff that carries the beans. The low level digital logic and control will need to be traced to tie it all together--but it can be more difficult and time-consuming to figure out due to the microscopic size of the components...

Image

.cheers,
kiev = kenny's innovative electric vehicle

skylogger
Posts: 76
Joined: Sun Dec 14, 2014 3:54 am

Re: Troubleshooting and repair for On-board Charger, DC-DC Converter

Sun Jul 29, 2018 3:43 pm

Hi Kiev:
That is the most awsome reverse engineering I've seen. At the moment, I put the faulty charger back in my friends car, so they could still use the DC-DC Converter section to charge the 12v battery, and can use a Chademo quick charger to keep the car up and running.
I have access to a wrecked IMIEV that is also a 2010, but the part number on the working charger inside it is different, and the inside
PCBs are totally different. I think the Wrecked IMIEV was a beginning of year 2010 model, and my friends IMIEV is End of Year 2010 model.
I had put the project on hold for a week, because 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.
So now that I know that the communications problem was a red herring, Tomorrow, I plan to swap chargers, so my friend will have a known good charger put into there car (although older revision), to see if there are any other faults in the car. This will allow me to take the faulty charger from his car, and use the wrecked IMIEV that I have access to as a test bed. I'll do a quick test of soldering wires to the points that you gave me on the output of the doghouse and check the voltage there so I can confirm if the first section of the block diagram is working.

I used to play a bit with the older desktop PC's, and I think I remember that you never power up the ATX Switching power supply without it being plugged into the mother board, or a dummy load, or the power supply would blow up. I was wondering if this IMIEV Charger and DC-DC Converter might be simular. Do you think if AC is applied to the AC input of the charger section, but no loads applied to the 12vdc output or the HV 360VDC output, would this possibly damage anything. Also, if there is no load on the 360vdc output, do you think the charger could be tested for short periods of time without coolant running through it, like testing it on a bench instead of fitting back in the car?

Besides the AC, 12V, 360V Connectors going to the box, there is the 4th connector that allows the charger to communicate with the rest of the car. I would assume there are pins in this connector that have to be grounded or 12v applied, or some CANBUS communications required for the Charger or DC-DC Converter to be active. Would you have any info on what would be required at this connector to be dummied out so as to be able to have the Charger in enabled mode. I can then test voltages at the outputs of each block in the block diagram and see what sections are still working and which are dead fairly quickly?

Thanks again for all the great documentation and work you are doing, With your help, I might be able to bring live back into this charger.

coulomb
Posts: 167
Joined: Sun Jun 10, 2018 8:32 pm
Location: Brisbane, Australia

Re: Bottom Board Trace Notes for OBC Schematic

Sun Jul 29, 2018 6:05 pm

kiev wrote:Here are my tracing notes on the two solder rails, left and right, of the bottom board.

Great work; thanks for starting this! 8-)

Rough draft Kicad sketch with measurement notes from the waffle plate (rectifier diode and FET diode drops).

I think that most if not all of your 0.7 V+ diodes are phantoms; other ~0.37 V diodes in series joined by low resistance windings. I get that this is just rough notes; may as well record all the readings, and make sense of it later.

I see no IGBTs here, but my guess is that they're in the lower right quadrant, with the 1K resistors being gate pull-down resistors perhaps. There are 8 squares there; my guess is that these are IGBTs (the type without integral diodes) and the back diodes right next to the collector and emitter connections.

The other thing that occurred to me is that maybe there are two sets of full-bridge IGBTs (the more common type with on-die back diodes), each driving their transformer primaries separately. But I can't reconcile that with your tracing.

Toward the top center is a dashed box labelled Top with 2 capacitors--those are the output filter snubber caps blown in Jay and sky's potted doghouse.

Ah! So these are in fact across the DC output. I'm not immediately seeing a reason for these to blow.

coulomb
Posts: 167
Joined: Sun Jun 10, 2018 8:32 pm
Location: Brisbane, Australia

Re: Troubleshooting and repair for On-board Charger, DC-DC Converter

Sun Jul 29, 2018 6:24 pm

skylogger wrote: I used to play a bit with the older desktop PC's, and I think I remember that you never power up the ATX Switching power supply without it being plugged into the mother board, or a dummy load, or the power supply would blow up. I was wondering if this IMIEV Charger and DC-DC Converter might be simular.

I don't believe so. I operated a lot of older Elcon/TC chargers on the bench with no load and had no problems. Of course, the iMiEV chargers could be different, but my gut feeling is no. Switching the DC-DC (12 V charger) with no load might be a different story, but I'm pretty certain that merely applying power and no CAN signals, no switching will happen, and no damage will be done. Applying a small 12 V battery for charging on the bench is a lot easier and safer than a 360 VDC battery.

Also, if there is no load on the 360vdc output, do you think the charger could be tested for short periods of time without coolant running through it, like testing it on a bench instead of fitting back in the car?

Again, I did this all the time on the Elcon/TC chargers. I'd say the iMiEV chargers would be fine on the bench with no cooling.

I'm a great fan of running these chargers initially at least from a current limited power supply. [ Edit: Into the mains input. The DC-DC might similarly work with much lower voltage at the 360 V input. ] I settled on 52 V (two 26 V 3 A power supplies in series) for the Elcon/TC chargers. This was enough to get the power supply chip working, and if enabled, the PFC stage could boost the 52 V (more like 50 V after the input rectifier) to the ~385 V that the power stage wanted. There were in fact jumpers (really just pairs of pads labelled something like J8) that could be used to first switch the output stage at 50 V, then if that worked, switch at 385 V (some 64x the energy in the bus capacitors with 8 times the voltage). There was even a jumper to cause the output stage to switch even though the microcontroller wasn't yet commanding output, and it wasn't seeing a battery to charge. This would put some 120% of maximum voltage at the output if all was working. By working methodically through the procedure, you'd be able to test the power stages with gradually increasing power levels, minimising the chances of having things blow up on the bench. I would never apply mains power till the charger was back together (apart from the cover with 28 screws), so the 400 V parts were safely tucked away, and the MOSFETs had their heatsink clips on. These Elcon/TC chargers were only air cooled.

[ Edit: heatsinks on -> heatsink clips on ]
[ Edit: 120% of rated -> 120% of maximum ]
Last edited by coulomb on Mon Aug 06, 2018 2:50 am, edited 1 time in total.

kiev
Posts: 925
Joined: Sun May 03, 2015 7:15 am
Location: The Heart o' Dixie
Contact: Website

Re: Bottom Board Trace Notes for OBC Schematic

Sun Jul 29, 2018 9:01 pm

coulomb wrote:...

I see no IGBTs here, but my guess is that they're in the lower right quadrant, with the 1K resistors being gate pull-down resistors perhaps. There are 8 squares there; my guess is that these are IGBTs (the type without integral diodes) and the back diodes right next to the collector and emitter connections.

The other thing that occurred to me is that maybe there are two sets of full-bridge IGBTs (the more common type with on-die back diodes), each driving their transformer primaries separately. But I can't reconcile that with your tracing.


That is what i was thinking also even though i didn't draw it up that way in the block diagram, but it seems likely since there are 2 separate Boost transformers that there are 2 separate H-bridges in parallel.

Looking at the waffle plate, i think the AC bridge rectifier is in the upper left quadrant, the PFC switches and blocking diode are in the lower left, the Boost H-bridges are in the lower right, and the Boost bridge rectifiers are in the upper right.

i too am trying to figure out how to test a charger on the bench.
kiev = kenny's innovative electric vehicle

ed5000
Posts: 166
Joined: Mon Jul 27, 2015 9:22 pm
Location: Castro Valley, CA

Re: Troubleshooting and repair for On-board Charger, DC-DC Converter

Mon Jul 30, 2018 10:34 am

I have noticed in the past how warm the interior of my car gets while charging on L2 (I'd say about 90 degrees with the windows up) so I started using my home DIY upgraded Mitsubishi evse that someone posted about here last year where you change out a small power supply and the cord plug and and it works on 240 VAC. The only problem with the diy upgrade is without programming it charges at around 1800 watts. I assume that it's still programmed for 8 amps but since now it's running on 240 V, it's now working at twice the wattage. It's more like a L1 evse.

The interior now feels much cooler and yesterday I opened up the engine hatch to check the charger temperature and found it to be very cool, maybe 90 degrees to the touch. That was after about 2 hours of charging on an 80 degree afternoon. I'll find my ir temperature gun and take some temperatures this evening to be more accurate. I've been doing this with a time clock for about a year now. Ed
2012 ES, Cool Metalic Silver, no quick charge, no fog lights and no navi but a cold weather package for when it gets 13 below, wait - it never gets that cold here.
"Silver Shadow"

jray3
Posts: 1623
Joined: Tue Dec 06, 2011 1:05 am
Location: Tacoma area, WA
Contact: Website

Re: Troubleshooting and repair for On-board Charger, DC-DC Converter

Mon Jul 30, 2018 10:56 am

skylogger wrote:I have access to a wrecked IMIEV that is also a 2010, but the part number on the working charger inside it is different, and the inside
PCBs are totally different. I think the Wrecked IMIEV was a beginning of year 2010 model, and my friends IMIEV is End of Year 2010 model.

The junkyard charger that I replaced my failed unit with also had a very different model number, though it looked identical on the outside and under the lid. Original on my May 2012 car was labeled as follows
9499D991
Serial.00692 V100
ZHTP1529R 2012.04.20
nichion
bar code= DPT1529ABR00692V100

The replacement came from an older car with far fewer miles
9499C662 NAS
Serial. 00497 V101
ZHTP1546R 2011.11.01
nichion
bar code= DPT1546NAS00497V101

We also know there are at least two different versions of the charger internals for our cars that work, so assuming the battery was of the same capacity (16, not 12 kwh) and using the same cells, I'd say to trust your common sense more than the sticker on the charger and try that swap if it looks like a match...
2012 i-SE "MR BEAN" 102,600 miles
2000 Mazda Miata EV, 78 kW, 17 kWh
1983 Grumman Kurbwatt EV,170 kW, 32 kWh
1983 Mazda RX-7 EV 43 kW 10 kWh
1971 "Karmann Eclectric" EV 240 kW 19 kWh
1965 Karmann Ghia Cabriolet

kiev
Posts: 925
Joined: Sun May 03, 2015 7:15 am
Location: The Heart o' Dixie
Contact: Website

Solder Pads, Waffle Plate and Output Filter schematics

Mon Jul 30, 2018 7:23 pm

Here is a cleaned up drawing with what i think may be on the waffle plate, being shown in the central region between the two solder pad strips. The IGBTs are just a guess since the part numbers have been etched off the chips (in addition to being covered in epoxy). There is a small label on the side of the waffle plate with a separate part number, CZ1525RPWR012 2R04 No. 0044.

Image


Output Filter

Image
kiev = kenny's innovative electric vehicle

skylogger
Posts: 76
Joined: Sun Dec 14, 2014 3:54 am

Re: Troubleshooting and repair for On-board Charger, DC-DC Converter

Tue Jul 31, 2018 8:32 pm

Hi Kiev:
You've really made life a lot easier with all the documentation you've done for these chargers.
I've got the charger out of the can, on the test bench, and using a power cord with fuse from wall
socket to the AC input ring terminal screws on the top pcb.

I tested the ACTIVE and NEUTRAL test points you showed me on the white strips of pins from the waffle.
The output from the Doghouse is measuring 239vac which is same as the wall outlet (here in australia)
so that looks like all is good with the doghouse.

I've used your new cleaned up schematic to check output of the rectifier bridge, and I measure 247vdc there.
I've check the voltage across the three BIG electrolytic caps, and there is no voltage there. This makes me think
maybe the Diode D7 is blow open. It's a bit slow going, as I have to disconnect all the connections to the top PCB,
Do a solder connection to test points, then put top pcb back on and run a test.
i'll test Diode D7 with an ohm meter once I get a chance to get to it again.

If this diode is blown open, I guess the next step is working out what part number the diode would be.
Depending on how many other parts are bad in the waffle board, I was thinking of maybe getting a replacement diode
that could be heat sinked to the case somewhere else, and run wire from it over to the the pads on the white strip.
Sort of a bypass operation ranther than removing the waffle board, cutting through the epoxy and replacing the original part.
Not the sexiest way of doing it but might get it working that way. (This would only work for parts blown open not shorted)

kiev
Posts: 925
Joined: Sun May 03, 2015 7:15 am
Location: The Heart o' Dixie
Contact: Website

Re: Troubleshooting and repair for On-board Charger, DC-DC Converter

Tue Jul 31, 2018 9:11 pm

Howdy sky logger,

i think the diodes are labelled D? as i didn't have any reference designators visible for the components on the waffle plate. But you are probably referring to the diode between L2 and P1 on the left hand solder strip, as that is in the circuit path from the rectifier thru the big inductor (bottom right corner of box) and then on to the big electrolytic cap bank.

i'm not sure why the DC is only reading 247--i would expect to see ~340vdc with 239vac mains.

do you have anyway to measure the mains current into the box while you are making the voltage measurement?

i think the diode(s) you are chasing are in the lower left quadrant of the waffle plate.
kiev = kenny's innovative electric vehicle

Return to “Batteries and Battery Management”