Thank you in advanced for anyone who figures this out.

I was a Commercial electrician for 7 years and currently an electronics hobbiest. I know how to diagnose most types of faults but this one has us stumped and this might be a long story.

We have 3 service trucks for the current company I work for. Before last year, we always used generators for our 120V power that we need to do our jobs. We came up with the idea of switching to solar with the assistance of alternator charging when driving. This would allow us not to use fuel on generators or idling trucks. Two of these trucks were already wired by me as far as outlets, panels and lights go. Never had any issues with generators except for typical generator break downs and fault but, nothing on the 120V side. Not even a trip breaker.

At first we had a hard time finding a large enough pure sine inverter to run the equipment that was not multiple of thousands of dollars. I found a company on ebay that builds large inverters by spec and by order. We ended up going with a 10KW pure sine wave inverter, duel 12VDC in, 120VAC out, North American standard plug. We do not use the outlets on the inverter, we only use the terminals. We first already had a ligit 350AH with BMS and a 150AH smart BMS LiFePO4 battery, wired in parallel for 12VDC 500AH. The inverter has duel 12VDC parallel inputs that both has to be hooked up for 10KW, which we all know it's a little over rated but more than what we need so I am sure we are not over running it on our equipment inrush or continouse running. We then have a smart BT charge controller to switch or combine between solar, alternator, and/or shore power charging, a solenoid for ignition isolation, breaker panel and automatic transfer switch on the HV side for shore and inverter power switching and a few solar panels. No need to list all the bus fuses and bus bars on LV and HV side.

We only bought one set of everything for testing. For three months there was zero issues, everything worked flawless. So we bought 3 more sets of the exact same components except the batteries we went with two 200AH LiFePO4 BMS batteries and these did not have a BT BMS since the charger has a smart BT display. The other trucks were not being used for a while so everything sat 6 months as I slowly peiced everything together for one of the other trucks.

That truck started getting used but no where near as much as the first, being my truck. I was told that everything worked great for 6 months without any issues. All of these trucks use the same rated equipment. All of the sudden one day, the tech starts the second truck, unplugs from shore power, goes inside the truck, turns on the inverter and immediately, the inverter starts smoking and he could see a bright red glow from inside it. He shut everything off and had to run the generator that we kept on the trucks as a backup. Note, the generators at this point now connect to the shore power and is transfered by the auto transfer switch, there is no way to backfeed or double feed, I made sure of that.

After he returned he called me, sent some pictures, explained what happened. I went through the existing work and the new work like crazy. I could not find any faults on the HV side or the LV side, no shorts or possibility for shorts. There is not enough equipment on the truck to were a combined in rush would overload the inverter. Transfer switch was working properly and everything isolated from HV to LV except chassis grounds. I disconnected the inverter, removed it, and took the cover off. I could not find any evidence of over heating, shorts, blown fuses, broken traces, nothing.. I connect a lead acid 12V battery to both inputs, there was 0 HV output, the LV voltage display was displaying the correct voltage but the HV voltage display was off. No alarms and fault light was off. These inverters have like 4 or 5 transformers, about 15 fuses, a few very large inductor, caps and alot of other support components.
They were still under warranty, I decided to just return it and let them deal with it. Unfortunately they were not being reasonable as they wanted to send me a new board which would require a massive solder iron. They do not know about my background in electronics nor did I want to be responsible for damaging the new board, so we filed a claim with the protection plan that we also purchased and shipped it off to them, I installed one of the other identical inverters just the way the other one was, double check for faults and function. Everything was working fine with no issues.

Fast forward a while, the inverter is still waiting to be repaired by the claim. The 2nd inverter in that truck lasted quite some time, I want to say months with about the same amount of use. The exact same thing happened again. This time he disconnected shore power first, started the truck, turned on the inverter and immediately started smoking with a large red glow coming from inside. This time he sent me a video and I could see what was going on. Pulled it out, took the cover off, the glow appeared to have come from one of the transformers, again nothing looks off except this time I can see the coating on one of the transformers has melted a little. It was so little thats it's almost not noticeable. Hooked another battery to it, almost same condition, correct LV display, no HV display, no faults. This time after connecting the battery with no load for diag, I can see the transformer that got hot start to smoke. All the transformers are identical and as i went back through the pictures from the first failure, I could see the winding's coating just barely melted as well from the transformer in the exact same position.

Edit: the windings were only somewhat melted just above the through hole solder joint, not in the actual winding.

At this point I have yet to find any faults with that truck, I have not excluding assembly defect yet for the first one was assembled much earlier than the other 3 where as the other 3 where assembled at the same time and has the exact same fault conditions.

Does anyone have any suggestion or has experience this before? I am not going to rule out that there is absolutly no issue with the install or componenats, there could always be a mistake somewhere that has been overlooked. I do not want to put another expensive inverter in this truck until I know 100% that it is not the equipment or wiring in the truck.
Even though we have these protection plans, I feel like if we have the same condition on a third inverter, questions from the protection plan will start to arise. My next step is to install the last inverter on the third truck which has zero existing HV. So it will be a fresh wire, conduit, equipment and componant install. If it does it again on this truck, I would suggest the same person assembled the three later inverters and its defects.

Any advice is appreciated, I apologize for this super long story but I am trying to be thorough to eliminate some questions and to paint the picture.

 

My 2C.

A quick WAQ from you story is that you might be having environmental issues if the electronics design is solid but running high power inverters on 12V battery banks is a questionable design choice IMO. Moisture, dirt, dust are killing the electronics with leakage paths. Even critters can eat wiring. What's the inverter IP rating or does it even have one. What's the provider warranty repair diagnosis?

 

My 2C.

A quick WAQ from you story is that you might be having environmental issues if the electronics design is solid. Moisture, dirt, dust are killing the electronics with leakage paths. Even critters can eat wiring. What's the inverter IP rating or does it even have one. What's the provider warranty repair diagnosis?

I don't know the IP rating or if they even have one. The location of the inverter is protected against against most environment. The outside gets a little dusty of course but the board on both were very clean. No indication of moisture. The truck in question is stored in a very clean climate controlled shop with pest control. Where as the original truck is stored outside, not protected well and still has no issues after over a year. I am confident that critters are not responsible nor is their any evidence of it, I have checked every inch of wire and conduit. Everything on the HV side is ran in EMT and only about 10 foot of MC is used. LV side is all encased in wire loom and no damage found.
As I mentioned the company does not want it to even be shipped backed to them, they want to send a new board and have me replace it in which I declined for liability reasons and shipped it to the protection agency which is still waiting for an answer, I have a feeling they are just going to replace the entire unit.

 

The Clean we see is often external. PCB substrates, transformer windings, wire to board connections, etc... are all places that can provide paths for HV leakage. Fans can jam dirt/gunk where it's hard to see with a simple visual inspection. Not saying it's the cause, only one possible cause to investigate or mitigate.

It could easily be simple surge related component failure, impossible to know without a detailed repair evaluation. You won't get more than pure guesses here from your story.

 

The Clean we see is often external. PCB substrates, transformer windings, wire to board connections, etc... are all places that can provide paths for HV leakage. Fans can jam dirt/gunk where it's hard to see with a simple visual inspection. Not saying it's the cause, only one possible cause to investigate or mitigate.

It could easily be simple surge related component failure, impossible to know without a detailed repair evaluation. You won't get more than pure guesses here from your story.

I was hoping a guess in something I have not already checked might lead to an answer. I get the environmental issue, just not looking into it due to combination of visual inspection and the original one goes through alot worse environment and usage conditions that these two. Thank you for the effort. If I get a diag report from claims I'll update.

 

I was hoping a guess in something I have not already checked might lead to an answer. I get the environmental issue, just not looking into it due to combination of visual inspection and the original one goes through alot worse environment and usage conditions that these two. Thank you for the effort. If I get a diag report from claims I'll update.

These things are frustrating and a good reason not to buy industrial quality products from other than from well know, long track record of reliability companies from places like eBay, save a nickel upfront and pay many dollars in lost productivity later.

Good luck.

 

It would be helpful if you shared some clear pictures of the damage for us to provide a better suggestion.

But a quick story of mine; which may or may not be applicable to your case.
I was also called to check out an inverter failure. To make a long story short, the manufacturer had used copper clad aluminum CCA wire in the transformer, presumably to save costs. Not only does CCA have a higher resistance than copper, if not attached correctly, the joints or splices tend to flow OVER TIME, causing loose or intermittent connections. With a high current, it will create a hot spot and also a fire. This is a recurring issue with CCA or pure Aluminum wire, although the former is more reliable it still requires an extremely good splice.

You did the right thing to refuse to repair the unit. If it fails again, and most likely it will, it would be YOUR responsibility this time.
I ignore what the sales contract mentions, but I would allow the company’s legal department or at the very least the purchasing manager to deal with it.

 

It would be helpful if you shared some clear pictures of the damage for us to provide a better suggestion.

But a quick story of mine; which may or may not be applicable to your case.
I was also called to check out an inverter failure. To make a long story short, the manufacturer had used copper clad aluminum CCA wire in the transformer, presumably to save costs. Not only does CCA have a higher resistance than copper, if not attached correctly, the joints or splices tend to flow OVER TIME, causing loose or intermittent connections. With a high current, it will create a hot spot and also a fire. This is a recurring issue with CCA or pure Aluminum wire, although the former is more reliable it still requires an extremely good splice.

You did the right thing to refuse to repair the unit. If it fails again, and most likely it will, it would be YOUR responsibility this time.
I ignore what the sales contract mentions, but I would allow the company’s legal department or at the very least the purchasing manager to deal with it.

Pics below, first one is full board, second is what all the transformers look like, third is only evidence of a thermal invent in both units on the same transformer.

 

Hard to say from the pictures, but it appears that the multiple strands soldered to the board have not been properly wetted by the solder. Just a thought.

Having said this, if you mentioned that you and your fellow worker saw “red glow” or at least heavy sparking, the damage should be far more noticeable. There would be charring on the transformer’s tape, discoloration of the board, smoke residue around the area, etc.
Strange!

 

Hard to say from the pictures, but it appears that the multiple strands soldered to the board have not been properly wetted by the solder. Just a thought.

Having said this, if you mentioned that you and your fellow worker saw “red glow” or at least heavy sparking, the damage should be far more noticeable. There would be charring on the transformer’s tape, discoloration of the board, smoke residue around the area, etc.
Strange!

I was expecting something more like this.

 

Hard to say from the pictures, but it appears that the multiple strands soldered to the board have not been properly wetted by the solder. Just a thought.

Having said this, if you mentioned that you and your fellow worker saw “red glow” or at least heavy sparking, the damage should be far more noticeable. There would be charring on the transformer’s tape, discoloration of the board, smoke residue around the area, etc.
Strange!

I know, one reason why I'm stumped on this. I'll try to get the video uploaded later.

 

I was expecting something more like this.
View attachment 341933
View attachment 341934

Nah, nothing like that, there is a big heat sink in the other side that coveres the entire board with just some support components under it but i don't see anything there either.

 

I know, one reason why I'm stumped on this. I'll try to get the video uploaded later.

Tried to upload the video and says the extention is not allowed so I'll post link to cloud share later.

 

I was expecting something more like this.
View attachment 341933
View attachment 341934

Exactly!
In my looooooong career in power conversion electronics, a power failure always leaves some significant, and I mean significant damage.

 

Nah, nothing like that, there is a big heat sink in the other side that coveres the entire board with just some support components under it but i don't see anything there either.

What do the power devices (MOSFET) legs, bodies and traces look like?
This is an example of smoked and sparked devices that were bolted to a heat-sink.


https://toshiba.semicon-storage.com...mosfets/400v-900v-mosfets/detail.2SK4207.html

 

Sp i have some updates. Now the inverter on my truck that has lasted more the twice as long now. I've noticed that when there is a low load, truck idling "charging off alternator" the LV display will reach 13.8 after some time. As soon as it reached 13.9, the HV side drops in and out, it makes the "power on beep" that it always does when turning it on, multiple times as the lights come back on multiple times, and start smoking out of the port like crazy. I turn it off, try and get a good look in the vent. I see nothing abnormal. Turn it back on, it will charge to 13.9 again, then rinse and repeat.
To combat this, I've been keeping an eye on it. As I use it, if I'm not using any heavy load, ie, microwave, vacume pump, compressor, heater, while only using my standby devices like lights, battery chargers, music, blower, ect.. as soon as I notice it reaches 13.8 on the lifepo4 or close to it, I drain my compressor down to kick on, it will drain the. Batteries down to about 13.2 and that gives me about 20 mins before it recovers to the 13.8 again. As long as I do that. Not a single issue. For the past 3 weeks that has been working great until last night I was struggling on a diag, usually doesn't happen, and completely forgot about it and it did it as soon as it reached 13.9.

This doesn't make sense to me because they are designed for 10-16 volt. Even if it was just considered 12 volt, any inverter should atleast be rated up to 15 or 16 volt because most alternators charge 14.1 to 14.6, electronic charges can charge as high as 15.2, that I've seen, and it's common sense to give up to 16 for some netting. Even if it wasn't designed for 14 and only designed for 12 to 13 or something dumb, it has a LV, HV cut out for over and under... so what the heck is the deal?

 

What it looks like is that one transistor was carrying the whole load. So I am suggesting that you invest in some fast-response fuses to quickly cut the power at the next failure. Size the fuses at just a bit more than the measured running current.