trace engineering 1200W faultfinding (230v 50Hz)

[mab]

hi all,

this is probably a long shot, but can anyone suggest where I might find a schematic drawing (or anything that will help trace faults) for the trace 1200w msw inverter?

I've done a few searches but I suspect they're too old for there to be much info online.

got it as spares/repair so I've no idea what's wrong with it yet (figured it was worth 0.99p for the transformer alone), but I thought it might be worth having a go at fixing it.

[karlb]

First thing would be to check the power mosfets with a multimeter. They are the things that tend to blow first. If the two outer legs show as a short or a very low resistance then the mosfet is blown so you should replace it.
Next thing would be to check the AC and DC fuse then Diodes, capacitors and transformer

[mab]

Thanks - I've been doing some investigating in the meantime;

just to confuse the issue I actually got two units (rather than one as I said in the 1st post)

the 1st one powers up and buzzes softly (cont green light) but there's no ac on primary side so I suspect - as you suggest - the fets are blown on that one.

the 2nd unit powers up and generates ac but gets very loud, draws a lot of amps and shuts down (fast flashing green light) and repeats cycle - which I'm guessing could be any one of:- blown fets on one side (or on shorting coil), or shorted trans, or voltage-feedback is open circuit.

I guess I can figure it out eventually but a schematic would save a huge amount of time - or even an explanation of how it works (what are the two small transformers on the pcb for? I guess one provides voltage feedback but the other?)

[boB]

One of the little transformers would be for voltage feedback and regulation.

The other small one should be a current sense transformer for output current
and charging current limiting.  This one should have a one or two turn
larger wire running in series with the big transformer secondary.

On those old MSW inverters, there is usually another winding on the large power
transformer for "shorting" when the MSW is at 0 volts to take care of reactive
loads.  The clamp winding.

boB

[mab]

thanks bob,

I suspect the voltage feedback trans on the 2nd unit, as it's hanging off the pcb by its wires atm, but haven't had time to investigate further.

[mab]

yep that trans is definitely a problem, trouble is the fine wire for the 230v winding has broken off both terminals, and I can't seem to get a good connection to the tiny bit left on one end.

[boB]

Yep, the old hair-wire problem.   That was always a problem because the wire is so small.

40 gauge or so and about 5000 (or 4000) turns.  If you're real careful, you might be able to
fix it.

boB

[mab]

Cheers Bob - really didn't want to rewind it so I very carefully soldered an extension to the shorter wire so I could reconnect to it's pin. I think I might have done it  - the 230v coil is showing ~2.7Kohm, and when connected to 250v (grid) and secondary is showing something of the order of a volt (ac voltmeter sensitivity issue) - I did think that was low until I traced the circuit to the AD736 and read the datasheet showing 200mV input so it may be OK.

resoldered onto the board, but I think that that was a secondary issue from the unit being dropped at some point. The primary fault seems to be one bank of FETs has blown (shorted drain-gate and a leaky open circuit D-S) so i'm going to sub  2 or 3 of them and test again.

the main trans work fine when disconnected from the pcb and used in reverse on grid power.

[mab]

Success!    I removed the dud rfp50n06 fets and sub'd in a couple of irfz46n's  - they're only 55v rather than 60v but I've got 6 of them and they're recovered from an old pcb so didn't cost anything - I'm hoping they'll work OK.

it's running OK - only tested to ~20w load so far (running of a bench PSU).

Now to start look at the other one.

[mab]

Feeling very smug now - both inverters are working again. 

other one didn't have blown fets - there was something odd going on in the voltage feedback - eventually removed the ad736 true rms device and tested it on a breadboard - worked fine (good thing too - they're expensive). soldered it back in to PCB - inverter now working fine. guess it was a bad solder joint.

Idly wondering now if I should sell them, keep one for emergencies, or try & convert one to 24v (my system is 24v).

[DamonHD]

Well done!  Smugness definitely allowed!

Rgds

Damon