Hello,

I was given this inverter for free, that if I can repair I can use as a backup power.

"Pure sine wave Inverter 3000W"
"Power that really moves"
"DC-AC"
text on the top of the case.

When i first opened it I did notice a burned YMP230N55.
I found on aliexpress that they offer YMP230N55 as a replacement for the more expensive IRF2804.
I bought 5x IRF2804, replaced 3x of them, the burned one, and one next to it, and the one closest to the right end of the board. I have 2 more IRF2804 to test with.

I turned on the board with the replaced FETs and we heard a sound that the inverter powered up actually for a second.
The same mosfet that was burned, smoked out immediately.

What can be causing this ? Any ideas by looking at the pictures ?

Thank you in advance

 

Need a schematic.

 

Need a schematic.

Unfortunately I was not even able to find the same Inverter on internet, there are some posts about "Power that really moves" inverters. Some people know the parts.

Should be something specific to burn exactly this mosfet, not any other one.

 

Hello,

I was given this inverter for free, that if I can repair I can use as a backup power.

"Pure sine wave Inverter 3000W"
"Power that really moves"
"DC-AC"
text on the top of the case.

When i first opened it I did notice a burned YMP230N55.
I found on aliexpress that they offer YMP230N55 as a replacement for the more expensive IRF2804.
I bought 5x IRF2804, replaced 3x of them, the burned one, and one next to it, and the one closest to the right end of the board. I have 2 more IRF2804 to test with.

I turned on the board with the replaced FETs and we heard a sound that the inverter powered up actually for a second.
The same mosfet that was burned, smoked out immediately.

What can be causing this ? Any ideas by looking at the pictures ?

Thank you in advance

befor replacing the mosfet u have to check the resistor and the transistor if thez are bad, i beliv that ur resistors are bad

 

Hello,

I was given this inverter for free, that if I can repair I can use as a backup power.

"Pure sine wave Inverter 3000W"
"Power that really moves"
"DC-AC"
text on the top of the case.

When i first opened it I did notice a burned YMP230N55.
I found on aliexpress that they offer YMP230N55 as a replacement for the more expensive IRF2804.
I bought 5x IRF2804, replaced 3x of them, the burned one, and one next to it, and the one closest to the right end of the board. I have 2 more IRF2804 to test with.

I turned on the board with the replaced FETs and we heard a sound that the inverter powered up actually for a second.
The same mosfet that was burned, smoked out immediately.

What can be causing this ? Any ideas by looking at the pictures ?

Thank you in advance

Simply replacing faulty components does not solve the problem. Components fail for a variety of reasons, and unless all the faults are removed, you will again have the same component failing.
As crutschow mentions in post #2, you will need a Schematic, either the original or by tracing from the PCB. Then we can start the troubleshooting.

 

Hello,

I was given this inverter for free, that if I can repair I can use as a backup power.

"Pure sine wave Inverter 3000W"
"Power that really moves"
"DC-AC"
text on the top of the case.

When i first opened it I did notice a burned YMP230N55.
I found on aliexpress that they offer YMP230N55 as a replacement for the more expensive IRF2804.
I bought 5x IRF2804, replaced 3x of them, the burned one, and one next to it, and the one closest to the right end of the board. I have 2 more IRF2804 to test with.

I turned on the board with the replaced FETs and we heard a sound that the inverter powered up actually for a second.
The same mosfet that was burned, smoked out immediately.

What can be causing this ? Any ideas by looking at the pictures ?

Thank you in advance

Hi,

The main failure in high power inverters is the main transistors, but unfortunately when they blow out they often take out the driver transistors, and that can be the driver chips or the driver transistors and the chips that drive them too. In these devices it is customary to change out the driver transistors as well as the main transistors, but also of course any chips that go bad with it. That means all of the transistors not just one or two, but you have to be prepared to test this properly afterwards or you run the risk of blowing out another set of parts.

There is also a test technique that requires running the input buss up little by little. If you apply the entire input all at once and something is wrong, it simply blows out the same parts again unless you got it fixed right. By running the input up slowly (volt by volt) you can measure input current and other things and see if something does not look right before anything blows out again. Since yours is 12vdc input you'd have to use a power supply on the input, but remember you only have to test (at first) with no load. If something is really wrong it usually will not work at any load, even with no load. Once you get that working, you can apply an increasing load little by little.
The types of loads that are often used to test these are "load cones" which are heater coils that plug in to sockets like porcelain light bulb sockets. If you get 500 watt load cones you can plug one it at a time for testing, or get 1000 watt load cones and do the same. You may or may not want to test at full load for very long though depending on the quality of the inverter it may not take the whole 3000 watts even though it may be spec'd at that power level. Light bulbs are not recommended because of their high nonlinearity, they present too much of a load before they get up to operating temperature.

It's unfortunate that transistors can be damaged internally but still work. That means they can blow again at any time even if the spec's are still within tolerance. That's one reason for changing all of them not just one or two. It's hard to assess which ones were damaged but still working for now.
The slow startup procedure is very important too.

When running up for the first time after repair, you will be running with no load at first. That means you can use a low amperage fuse on the input, just for that one test. This fuse could protect the transistors from blowing again if there is a problem as the fuse will blow first. Yes it's not always true what they say the transistor protects the fuse, that's not in all cases. However, you have to place the fuse between the DC power buss and the transistors not at the very input if there are significant input capacitors in there. It has to be after those caps or it won't do anything. That means altering the circuit slightly but not that much. If the fuse blows, that means one or more transistors are causing a problem or else the control signals are not right anymore.
I will warn you though some argue this test with the fuse is not adequate, but I've seen it work many times in the past. What is not argued ever is turning up the DC buss slowly while watching the input current and other things like transistor switching patterns.
When turning up slowly like this for the tests, you usually have to supply a separate power source for the logic. That's so the logic runs normally even with very low DC input buss voltage. That means you have to dig into the unit to figure out how to do this if you do not have a schematic. These things are complicated devices in general, but their wiring is usually not that complicated so you can figure out what lines to break and how to power up the logic separately.

Lastly, I might also add that sine converters have come down a lot in price in recent years. Much cheaper than they used to be. Maybe a new one would be simpler to deal with.