Onboard charger capacitors failed

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
wxmatt said:
Upon further review, my piece looks exactly like a part of this "Waffle Plate."
Yes, as my crossed post indicated.

I'm guessing it's pretty likely that I'll be buying a new onboard charger.
The Mitsubishi people found that a brand new OBC cost of the order of US$4000, and that's for a 3.6 kW model. They found it far better value to find a wrecked OBC and install that. With the iMiEV, there is no VIN locking of the OBC; that may not be the case for Leafs.

There is a common fault with the Mitsubishi version of this charrger, where two small capacitors tend to blow up (literally shatter into pieces). If you do use an OBC from a wreck, it may be worth replacing the equivalent of those capacitors before installing. Use higher rated (e.g. 4 kV) parts. Mitsubishi USA actually extended the warranty on the OBC to something like 10 years, so several of them have been replaced under warranty. It may not be wise waiting for Nissan USA to follow suit :eek:
 
There are 5 metal standoffs made from hex stock--these are threaded on the end and used as hold-downs for the big board and go thru the mounting holes of the waffle plate. As coulomb mentioned they have a sequence of circled numbers, start at 1 and remove the screw or the standoff and follow the arrows (should be 11 fasteners to remove). The white plastic faston covers have a release mechanism for the locking fastons--slide the plastic up the wire to release the tab and they will easily come off, pulling on the wire and they will fight you.

This is only the second failure i have ever seen or heard about of some component in the plate and would be very interested to find out what failed. i spent a lot of time tracing the waffle plate circuits and have a schematic that can be used to check the diode drop voltages inside, but yours has definitely blown the epoxy right off.

The good thing about the Laef OBC board is that all the components for the PFC circuit and the secondary LV supplies are mounted on the top surface--nothing on the bottom side which is not reachable without desoldering the 72 joints to the waffle plate. Also no little blue snubber caps.

Good luck and let us know what you find,
kenny
 
Thank you, again to everyone who has responded and given their input. 🙏🏻

I wasn’t able to see much, because accurately de-soldering 72 pins is a little beyond what I have time for today (or beyond my expertise, if I'm honest), but I was able to detect the location of one component that failed.


Ejva5kIXkAYnsH4

I found more pieces of epoxy that look to be part of components… Five total.


Ejva5kHWsAEXNrb

I don’t know if it’s obvious on this picture but it’s a little more obvious when I look at it, but I think the component that’s in the lower left-hand corner is at least one of the componets that has been separated from the waffle plate.


Ejva5kHXcAcSQoC

Relative to the board on top, it would be the lower left-hand quadrant, assuming you were looking at the charger as it would vertically be mounted in the car.


Ejva5kYWAAI72Uu

Referencing this picture which was posted in a report from Oak Ridge national laboratory, I have circled the one componet I solidly can see the scar from. https://info.ornl.gov/sites/publications/files/Pub46377.pdf


As far as repairing is concerned, I have located a used 2011 charger with the part number 296A0-3NA0A. I know that the 3NA8A has superceeded all previous models, including my current, failed 3NA5A. I don’t doubt that I could probably swap in this used charger, which believe it or not comes with a six month warranty from the shop who pulled the part, but long-term, if newer revisions have solved this sort of problem, or others, I would rather go with the new part at twice the cost. My wife and I currently live only 6 miles from our work, and within the year will be moving within 1 mile of work, so as long as the other components holdup, we could use this for a commuting vehicle for (pretty much) ever.

Thoughts?
 
In the upper left corner is a component that appears to be the twin of the one I mentioned a few posts earlier. (It says L201A near it.) It appears to have three connections. Two of them have plastic insulation shaped onto them. The third is bare, in a way that suggests to me that it had plastic on it but it broke or melted off.

7Qobct5.jpg
 
Good job wxmatt, you have done an excellent job of sleuthing out the circuit board. Those 5 pieces that came off the waffle plate are from the PFC Boost section in the left lower quad, 3 FETs and 2 diodes. They are controlled by IC312.

i would guess that something caused the PFC to stop chopping and turn hard on, and the direct current thru the FETs just blew them off the board taking the diodes with them.

It takes a fairly powerful desoldering station to remove the plate. The part number is on a tag on the edge if you find one that you want to swap out, but i would guess the root cause is from a component on the board rather than the plate, so i wouldn't reuse it without finding the culprit.
 
LeftieBiker said:
In the upper left corner is a component that appears to be the twin of the one I mentioned a few posts earlier. (It says L201A near it.) It appears to have three connections. Two of them have plastic insulation shaped onto them. The third is bare, in a way that suggests to me that it had plastic on it but it broke or melted off.

I hadn’t noticed it until you pointed it out, but on the side of that component, something looks to be fractured, and it almost looks like where you see R108, there is a little soot, as though it burst out that direction.
Ejx3ZlsXsAEQung
 
nlspace said:
Good job wxmatt, you have done an excellent job of sleuthing out the circuit board. Those 5 pieces that came off the waffle plate are from the PFC Boost section in the left lower quad, 3 FETs and 2 diodes. They are controlled by IC312.

i would guess that something caused the PFC to stop chopping and turn hard on, and the direct current thru the FETs just blew them off the board taking the diodes with them.

It takes a fairly powerful desoldering station to remove the plate. The part number is on a tag on the edge if you find one that you want to swap out, but i would guess the root cause is from a component on the board rather than the plate, so i wouldn't reuse it without finding the culprit.

Also keep in mind that GE Wattstation EVSE's can fry the OBC due to high inrush current. If you have a GE Wattstation, throw it in the trash and get an EVSE before you blow your next charger. More info:
https://mynissanleaf.com/viewtopic.php?t=7996&start=10
 
Good catch to find that blown diode next to the capacitor. That gray material is an RTV staking compound to hold the cap to the board and reduce vibration.

The diode is inline on the HV(+) side to bypass the big PFC inductor coil (orange wires that attach to the board at the L201A and L201B faston tabs) and attach to the (+) of the big black 2700 uF capacitor that attaches to the Red and Black faston tabs.

It is a Shindengen part number D4F60, 4A 600 V diode, available at Mouser, et al. marked 4FV60.

The four 100k resistors in series comprise a bleed-down circuit for the big cap.

Does your big L201 inductor have continuity or might it be blown also?

Is the big 2700uF capacitor okay or might it be shorted?

DLqQdBP.jpg
 
nlspace said:
(1) Does your big L201 inductor have continuity or might it be blown also?

(2) Is the big 2700uF capacitor okay or might it be shorted?

1) I believe that the L201 inductor could be blown because I got a continuity reading of infinity when measuring it from the pins below the board.

2) I got a reading of near zero ohms on the large black capacitor, so (the non-electrical engineer in me believes that) it looks to be shorted.


I just purchased a used charger from an early 2011 LEAF off of eBay that I plan to put in the car, but I’m still very interested to see what all blew up in this one so that I can feel confident that there wasn’t something else on my end that caused the issue.


Lothsahn said:
Also keep in mind that GE Wattstation EVSE's can fry the OBC due to high inrush current.
My EVSE is an early model JuiceBox (15kw/60A circa 2014). Dumbly, my power cord ran under my garage door and for a number of years I didn’t have a block to protect it until about a year ago. I do have some concern that this could’ve contributed to the problem, but that’s purely speculation.
 
wxmatt said:
1) I believe that the L201 inductor could be blown because I got a continuity reading of infinity when measuring it from the pins below the board.

2) I got a reading of near zero ohms on the large black capacitor, so (the non-electrical engineer in me believes that) it looks to be shorted.


I just purchased a used charger from an early 2011 LEAF off of eBay that I plan to put in the car, but I’m still very interested to see what all blew up in this one so that I can feel confident that there wasn’t something else on my end that caused the issue.

My EVSE is an early model JuiceBox (15kw/60A circa 2014). Dumbly, my power cord ran under my garage door and for a number of years I didn’t have a block to protect it until about a year ago. I do have some concern that this could’ve contributed to the problem, but that’s purely speculation.

Measure for continuity between the two orange wires leading to the inductor. It should read a very low, close to zero Ohms if okay.

Measure the capacitor between the red and black wires. If it reads close to zero ohms then it is shorted. A shorted capacitor could cause the damage you have found so far. i would expect to see some evidence such as the vent plug would be blown out or bulging if it failed over time. For a sudden overvoltage punch-thru it might look normal externally, but would be shorted internally.

It seems that if the EVSE voltage was too high then i would expect something upstream would have been taken out. A lightning strike or sudden loss of power during charging could cause damage in the OBC, there are protection circuits but everything has limits.

There are 3 fuses on that board that need to be checked for continuity also. F101, 102, 103.

Measure for resistance between the solder junctions labelled "8" and "7" on the seventh row down on the left hand side.
 
nlspace said:
Good catch to find that blown diode next to the capacitor. That gray material is an RTV staking compound to hold the cap to the board and reduce vibration.

The diode is inline on the HV(+) side to bypass the big PFC inductor coil (orange wires that attach to the board at the L201A and L201B faston tabs) and attach to the (+) of the big black 2700 uF capacitor that attaches to the Red and Black faston tabs.

The four 100k resistors in series comprise a bleed-down circuit for the big cap.

Does your big L201 inductor have continuity or might it be blown also?

Is the big 2700uF capacitor okay or might it be shorted?

DLqQdBP.jpg

That diode is less likely to fail than the other diode, as it's just to reduce peak startup currents in the FET and sees less peak reverse voltage
than the other diode. If it opens, the FET drive signal may be designed to not occur. The coil has a lower chance of failure than the FET.
 
nlspace said:
Measure for continuity between the two orange wires leading to the inductor. It should read a very low, close to zero Ohms if okay.
Measured at zero ohms.


nlspace said:
Measure the capacitor between the red and black wires. If it reads close to zero ohms then it is shorted. A shorted capacitor could cause the damage you have found so far. i would expect to see some evidence such as the vent plug would be blown out or bulging if it failed over time. For a sudden overvoltage punch-thru it might look normal externally, but would be shorted internally.
Resistance gradually builds to 19.7 mega ohms.


nlspace said:
It seems that if the EVSE voltage was too high then i would expect something upstream would have been taken out. A lightning strike or sudden loss of power during charging could cause damage in the OBC, there are protection circuits but everything has limits.

I think it’s unlikely that lightning was an issue, because I’m a meteorologist and I don’t remember an instance of that, but we do have frequent power disruptions where we live.



nlspace said:
There are 3 fuses on that board that need to be checked for continuity also. F101, 102, 103.
F101 - no continuity
F102 - zero ohms
F103 - zero ohms



nlspace said:
Measure for resistance between the solder junctions labelled "8" and "7" on the seventh row down on the left hand side.
Zero ohms… Assuming that we are talking about two of the 72 pins that are soldered from the waffle plate to the bottom circuit board.

My wife and I were joking on our way back to work last night that I had “received my homework assignment”… and my 6 year old daughter is actually helping me to check this with a multimeter, so it has been an educational experience for both of us and I’m hoping that someone in the future uses what we are finding to diagnose their problem.
 
wxmatt said:
...
F101 - no continuity
F102 - zero ohms
F103 - zero ohms
...

Zero ohms… Assuming that we are talking about two of the 72 pins that are soldered from the waffle plate to the bottom circuit board.

My wife and I were joking on our way back to work last night that I had “received my homework assignment”… and my 6 year old daughter is actually helping me to check this with a multimeter, so it has been an educational experience for both of us and I’m hoping that someone in the future uses what we are finding to diagnose their problem.

Haha that's funny, i'm glad she was learning how to use a voltmeter. Maybe we can find something else to measure and check, like re-check Fuse F101, and check for continuity on the those black nickel-sized lollypops next to the fuses in the black potted area. Also check continuity across that little glass tube (lightning arrestor) next to that.

So the inductor looks good, and the capacitor too. And yes those two labelled pins on the solder junctions to the waffle plate, so the sense resistor inside the plate looks okay. The output fuse 102 is okay, same for F103.

But the Good News is that you have found another failed part, and that may actually have been the culprit to start this massive failure, namely the open fuse F101, which is the first fuse of the two in the AC Input voltage to the board. A sudden disconnect of the AC input is not good, it is the bad thing™.

If that fuse lets go in the middle of charging then all the magnetic field of the big coil will collapse and all that energy will induce a current in the coil, which, if it no longer has a return path, then as Miranda reads about not having an attorney, one will be created for you. That current will make a path thru whatever is in the way and wreak havoc all the way home.
 
nlspace said:
But the Good News is that you have found another failed part, and that may actually have been the culprit to start this massive failure, namely the open fuse F101, which is the first fuse of the two in the AC Input voltage to the board. A sudden disconnect of the AC input is not good, it is the bad thing™.

If that fuse lets go in the middle of charging then all the magnetic field of the big coil will collapse and all that energy will induce a current in the coil, which, if it no longer has a return path, then as Miranda reads about not having an attorney, one will be created for you. That current will make a path thru whatever is in the way and wreak havoc all the way home.

Which came first, the chicken or the egg?

If parts fail, and the AC current increases enough, the fuse had better blow to prevent a fire or other damage.
 
@matt: what is the age, voltage and condition of your 12V aux battery--has the car been sitting for an extended time during the covid shutdown, might the 12V aux have been weak or degraded?
Do you remember which EVVSE was being used when it failed? Maybe it was charging but never reached FULL?


Which came first?
F101 the first fuse in the path has a ceramic body and is blown. It is marked 250V 31.5A, and is directly followed by F103, with a glass body marked 250V T4A, a time-lag fuse that has pre-arcing time characteristic of 10msec at 100 A.
 
nlspace said:
Maybe we can find something else to measure and check, like re-check Fuse F101, and check for continuity on the those black nickel-sized lollypops next to the fuses in the black potted area. Also check continuity across that little glass tube (lightning arrestor) next to that.
Finally got a few moments where my daughter and I could look things over.
1) Fuse F101 - No Continuity
2) Lightning Arrestor - No Continuity

Lollypops:
Since I don't have letter identifiers, I'll identify them by orientation, since two are aligned, and one is perpendicular.
1) The two that are side-by-side both showed no continuity
2) The perpendicular black lollypop showed 0.215 mega ohms

nlspace said:
@matt: what is the age, voltage and condition of your 12V aux battery--has the car been sitting for an extended time during the covid shutdown, might the 12V aux have been weak or degraded?
Do you remember which EVSE was being used when it failed? Maybe it was charging but never reached FULL?

The 12v that was in it at the time was about 2 years old. I have a trickle charger for the 12v that I would attach occasionally, though it had been a few months since I had topped it off. The LEAF is our workhorse, so we drove it pretty much daily, even through COVID. The car was parked for about a week (while we were on vacation) prior to the charging issue, though it was not plugged in during that time and stored in the garage, which is abnormal, because it is usually parked in the slab at the side of our driveway. When we returned from vacation, I plugged the car into the 240v EVSE that evening. I do not remember if it started charging or not. I believe it likely did because I drove it on a few errands the next 2 days and still had about 60% SOC when I first realized it wasn't charging any more a couple of days later. I then parked the car, knowing I needed to conserve battery power since I didn't know where/how I was going to fix it. When I went to drive it about 5-6 days later, the 12v had died (about 14-16 days after initially leaving on vacation). I attempted to recharge the 12v with the trickle charger, but it kept kicking out an error. I took the battery by Walmart, and they replaced it with a new one (under the 3 year warranty), and that's the battery in the car at the moment. Unfortunately, because the battery was flat dead when I took it to Walmart, I did not get a voltage reading on it (or more accurately, it was likely close to zero).

We almost exclusively charge at home on the 240v Juicebox EVSE, and it would likely have been the one that was utilized when the failure occurred. Unfortunately, because the kids are ALWAYS a distraction, I'm not sure which time it failed (as mentioned above), so it's possible/likely that failure occurred without it reaching a full charge at some point in the 2-3 days after we returned from vacation.

I apologize for the glut of details, but I figured it was better to lay everything out so I didn't miss a critical detail.
 
No that's the kind of detail needed to troubleshoot this sort of thing. I was expecting that there might be some issues associated with the 12V aux and you have confirmed that there were.

One thing we know is that the 12V charging of the laef is insufficient to prevent sulfation of the cells; with 2 years of sulfation that battery was in a weak and worn out state. One theory is that left to its own daily discharge over vacation it was too weak to support HV charging with the EVSE, either to the LV supplies for the OBC and VCU, or for the Main Contactors and relays used during HV charging.

Another is that the fuse failed due to mechanical fatigue over time.

What is amazing to me is that the output stage and filter section with fuse survived without damage, but that's a good thing.
 
nlspace said:
I was expecting that there might be some issues associated with the 12V aux and you have confirmed that there were.

One thing we know is that the 12V charging of the laef is insufficient to prevent sulfation of the cells; with 2 years of sulfation that battery was in a weak and worn out state. One theory is that left to its own daily discharge over vacation it was too weak to support HV charging with the EVSE, either to the LV supplies for the OBC and VCU, or for the Main Contactors and relays used during HV charging.

Another is that the fuse failed due to mechanical fatigue over time.

What is amazing to me is that the output stage and filter section with fuse survived without damage, but that's a good thing.

So… In a fortunate and surprising turn of events… My onboard charger is going to get here early… Probably today. Considering that I have a new 12V battery, and based upon what we know so far, is there any reason to believe, that the car will likely cause a failure in the new (used) charger?
 
wxmatt said:
...
So… In a fortunate and surprising turn of events… My onboard charger is going to get here early… Probably today. Considering that I have a new 12V battery, and based upon what we know so far, is there any reason to believe, that the car will likely cause a failure in the new (used) charger?

i think you should be good to go. It was driving okay before you removed the OBC, right?

If you want to donate the circuit boards to support OBC troubleshooting i will pay the shipping costs.
 
nlspace said:
i think you should be good to go. It was driving okay before you removed the OBC, right?

If you want to donate the circuit boards to support OBC troubleshooting i will pay the shipping costs.

I got the replacement charger installed in the car tonight. It is presently charging using my 240v EVSE. I refilled the coolant, but I noticed that the replacement on board charger was rather warm. I imagine that’s to be expected, but I measured somewhere in the neighborhood of around 90 to 100°F. Checking Leafspy Pro, I am code-free!

I would be willing to donate the circuit boards. Considering all you have helped me with on this project, it’s the least I can do. I really appreciate all of yours responses and help over the last few weeks!
 
Back
Top