Inverter problem

[tytower]

I just bought a 1000 watt inverter off ebay. used it on my digital TV and the TV hums?
had a 500 Watt before that that ran the TV fine with no noise.

So I got out my little digital oscilliscope and had a look at the waveform . pretty jagged but with a big spike through it .
Guess thats whats causing the hum.

Can anybody suggest a fix?

I'll try get the videos of the oscilliscope on here somehow
https://mega.nz/#!608n3aiJ!5JayyRJ0ZxsGNBkNfRoosJKXYexyZHf4sAGsBiFCK9E
https://mega.nz/#!jwkzCChC!TpLMpwUy4l7sTRqVj3GgCpu95I4zTkVJtSM133rdUXY

[james_s]

Most lower cost inverters output a "modified sine wave" which is either a square wave with dead time inserted or a sort of square wave with multiple steps. Some types of equipment will hum or even malfunction when operated from one of these inverters, there is not much you can do other than use a different inverter. The good ones produce a true sine wave but they cost more. Beware random Chinese units not sold under a reputable brand.

[tytower]

This unit is sold as a true sine wave inverter . As such it should have a board that produces a true sine wave and the oscilloscope results when the unit is powered up but not switched on show that it is indeed showing a good sine wave .

When the unit is turned on however the wave gets raspy and has a big spike in it .

I was hoping somebody knowledgeable might have a look at those videos and point me in the right direction to save this unit .
Personally I think the output capacitor might be faulty or a mosfet or diode is breaking down.

I will attach some photos of its insides if I can

[tytower]

This continually comes up no matter what size and file type I try to upload??

Network Protocol Error

An error occurred during a connection to www.eevblog.com.

The page you are trying to view cannot be shown because an error in the network protocol was detected.

    Please contact the website owners to inform them of this problem.

[EEVblog]

Test

[EEVblog]

This continually comes up no matter what size and file type I try to upload??
Network Protocol Error
An error occurred during a connection to www.eevblog.com.
The page you are trying to view cannot be shown because an error in the network protocol was detected.
    Please contact the website owners to inform them of this problem.

No idea, never heard that one before   

[tytower]

Just retried after 24 hrs and if I do them one at a time laboriously thay are at least going up,

[MagicSmoker]

This unit is sold as a true sine wave inverter . As such it should have a board that produces a true sine wave and the oscilloscope results when the unit is powered up but not switched on show that it is indeed showing a good sine wave .

Lots of things sold on ebay don't do what they claim... AFAICT from your pictures this inverter just has a single LC output filter which will work best at making a sine wave if supplying a resistive-inductive load (e.g. - a motor). Computers, TVs and most other consumer electronics use switching power supplies present a capacitance-dominated load, however, which effectively messes up the C part of the LC filter.

However, oscilloscope screen shots would be far more useful than pictures of the inverter itself.

[james_s]

I would only buy a pure sine inverter from a reputable brand, there are a LOT of random Chinese inverters that don't even come close to meeting advertised specs.

[phester]

from what i can see there are two controller boards in the inverter. I have a "Chinese" inverter and it lived up to its claim of being pure sine wave but I would estimate at about 1/2 its nominal rated power. Small fans cylce alot for the small heatsinks at 200w halogen when it is a 500w nominal inverter.

If it has two controller boards then it should pure a sine wave inverter. One would be a DC stepup and the other a DC to ac pwm board inside. can you post pics of the small daughter boards. In mine one is an EGS 8010 board and the other is scrubbed of IC markers but it is a SPWM board.

[tytower]

Phester.. Thanks for the reply . Didn't think anyone was interested . I have been trying to learn how these inverters work  so I could fix the few I have kept for parts and maybe one day convert a Grid tied second hand unit to a non grid output with battery charger.
The grid tied units can be had cheaply as people update to bigger systems . no one wants the smaller units so they just get dumped.

I can't give you the pics you asked for because i did a deal with the seller for a full refund and postage back if I bought the model recommended by him. The cost was $175 as opposed to $148 so I jumped on that and returned it . The new one seems to be working well with no hums or glitches .

Stll got the one it replaced though . It worked well for about 12 months the there was just no output one day .
I switched it off then back on and got a loud crack followed by a second loud crack an instant later than the first.
I will put a few pictures up of it . I guess the A/C output finals blew so i replaced them and what looked like a blown diode too but it powers up from Battery but there is no "ON" light and i can't detect any controller chip voltage either .
Its like whatever does the switching just isn't doing it .

I looked around for some sort of circuit diagram to work from to have a go at fixing it but nothing I have found seems to fit what I am seeing . Perhaps you might be able to point me to a possible schematic for it ?

[tytower]

A couple more

[phester]

I am not an expert but it is obvious the driver fets on the IR2110 are blown. This is most likely an eg8010 board but some oem swap some of the pins. My guess is this is the driver for the final stage that you say you have changed. so a quick test would be to check that you generate High voltage on the large cap after turn on. This would prove that half the cct is working (DC stepup). Be careful as the HV on the cap will be present even when the board is powered off. You must short the cap out with the unit de-energized before you can safely poke around. If so then your issue is still with the IR2110 drivers and associated board and / or the final stage fets.