Help with Burned out 3000 watt inverter

[Hisham343]

Hi everybody

I had 3000 watt power inverter for a while now input 12V through 200 Ah lead acid battery output 240V 50 Hz, I use the inverter to power up my apartment during the planned blackouts in my country, few days ago I burned out the inverter when I introduced the mains power to the outlet of the inverter while it was on, I got all the kinds of bangs and flashes, before shut down the power switch. When I opened the inverter I found that the nearest MOSFET was blown out as in the picture.

When I flipped the board I found a 16 pin SMD IC burned as well as seen in the picture.

What is amazing that all the part numbers were deliberately erased by the manufacturer, I will try to find the part number of the MOSFET but what I need help with is the IC, I thing its some kind of oscillator because its connected to the Gats of the MOSFETS.

Can you help me to identify that IC

 

[Hisham343]

Can it be a 4060

 

[(*steve*)]

R89 and R75 don't look too happy either

 

[(*steve*)]

I doubt it's a 4060.

Those mosfets are in pairs, and the chip drives the gates.

Can we get images of both sides of the board, from directly above/below so we can make out what the connections are to this chip and how it connects to the mosfets. It might give someone a clue.

 

[(*steve*)]

Oh, and sometimes a wet finger on the chip will reveal markings. You don't stand much chance on the mosfets though.

 

[Hisham343]

Thank you *steve* for your time
I tried the wet finger, but it seams that the data was erased in purpose, anyway here are the two pic you requested
pic one from above

Picture two from below

I am planing -if possible- to sacrifice pair of MOSFETS (the one burned out and his pair) and reconstruct an oscillator circuit based on DIP IC, any suggestions

 

[(*steve*)]

It looks to me that the burnt out mosfet is at the end where the AC comes out. I think it's fairly safe to assume that the 8 mosfets around that end of the board are arranged in an H-Bridge, converting the high voltage generates by the other end of the board into a useable AC.

My gut feeling is that the input end produces a simple high voltage DC, and the output end produces a modified square wave. I don't think it is using PWM to produce a sine wave output (there would likely be an output inductor and the drive to the gates would be much stiffer).

If you pull out the mosfet and it's twin, my first priority would be to determine if it's a P or N channel mosfet (or indeed, something else).

My next priority (well, my first priority) would be to test all the other mosfets for shorts between drain and source. Pairs are in parallel, so if either one of a pair is shorted, you'll be able to read it in-circuit. You;re looking at the resistance between drain and source (the ones with the beg fat traces coming from them), and measure it in both directions. What is the result?

 

[Hisham343]

Thank you again *steve*, this inverter is modified square wave and not using PWM (according to specifications), your totally right about the MOSFETS, and your theory about the function of the output end of the circuit is considerable, BUT what about the oscillator, which is considered the most important part of the output circuit, (as MOSFETS is easily replaceable or we can just remove the pair, WHAT do you think about the two ICs in the under surface of the board (including the burned one)????

 

[(*steve*)]

Great, if it's a modified square wave (as I surmised) then the chip that died *may* not be anything exotic. The issue becomes identifying it (and that's where I'm heading).

I need to take a closer look at those images.

 

[JMW]

Hi everybody

I had 3000 watt power inverter for a while now input 12V through 200 Ah lead acid battery output 240V 50 Hz, I use the inverter to power up my apartment during the planned blackouts in my country, few days ago I burned out the inverter when I introduced the mains power to the outlet of the inverter while it was on, I got all the kinds of bangs and flashes, before shut down the power switch. When I opened the inverter I found that the nearest MOSFET was blown out as in the picture.

When I flipped the board I found a 16 pin SMD IC burned as well as seen in the picture.

What is amazing that all the part numbers were deliberately erased by the manufacturer, I will try to find the part number of the MOSFET but what I need help with is the IC, I thing its some kind of oscillator because its connected to the Gats of the MOSFETS.

Can you help me to identify that IC

 

[JMW]

While it is unfortunate you damaged the supply, the far more important issue is that this occurred. For the inverter to be damaged upon restoration of power, you committed a very serious violation. You plugged the inverter into an outlet, to power your residence apartment, you were also powering the grid supplying your home. This could have damaged someone else's appliances and more importantly injured or killed someone working on the power lines. Before you repair this device, get a transfer switch and install it properly.

 

[Hisham343]

While it is unfortunate you damaged the supply, the far more important issue is that this occurred. For the inverter to be damaged upon restoration of power, you committed a very serious violation. You plugged the inverter into an outlet, to power your residence apartment, you were also powering the grid supplying your home. This could have damaged someone else's appliances and more importantly injured or killed someone working on the power lines. Before you repair this device, get a transfer switch and install it properly.

exactly JMW I knew that could have happine and i know its my fault i should have taken the precautions (safety first) and i already get the transfer switch and i will not install a new inverter before i fix the installation.

 

[Hisham343]

if we cannot identify the components can we:
first sacrifice a pair of mosfits with 25% reduction in power .
second cancel the original oscillator circuit and replace it with a pure sin wave PWM one
?????

 

[(*steve*)]

OK, I've taken a closer look, but I'm no closer in figuring out what that chip might be.

It may pay (at least for curiosity) to check the 8 fuses on the board.

It would be interesting to know where the gates for those 8 mosfets end up. One pair end up at pin 11 of the dead chip, but I can't trace the others.

 

[Hisham343]

Thanks *steve*
I will do all the measurements you asked for and replay to you ASAP

 

[Hisham343]

I have tested the input low voltage part of the circuit including the fuses and every thing seems to be ok, I removed both the blown up mosfet and the driver IC, I followed the connections to the gates of the mosfets to the IC which certainly the driver of the mosfets, more over i powered the inverter with 12V , 4 amp source, and the inverter went on with the green led lite !!!!!! when i measured the output it was 240V DC, I didnt know how to check the mosfets one by one while its in the circuit ???, I am planning to make a driver circuit better pure sine wave and let it drive the output mosfets, what do you think??? and what kind of circuits you suggest???

 

[(*steve*)]

This is actually good (but not what I might have recommended)

Clearly the dead mosfet and the chip, once removed allows the unit to try to operate.

Can you tell me which pins the gates are connected to?

 

[Hisham343]

Its pin 11 *steve* and ground at pin 7

 

[Arouse1973]

The IC could be a 4093 I have seen these driving smaller fets that drive the bad boys.
Adam

 

[Hisham343]

Thank you Adam (Arouse 1973), I made a quick comparison based on the data sheet of the 4093 and the connections in the inverter and I think your absolutely right, what do you think about the fets ( the bad boys) I have tested them and I think I have 3 more bad fets need to be replaced, from there look what do you thing the voltage and type.