Power Jack PSWGT-1200 manual/repair guide

Umm...unfortunately I have never had one of these on my bench, so it's really difficult to provide any assistance.  Best bet would probably be to see if you could get a replacement board.

I will note that these are Chinese products: there is no "repair manual" or anything of that sort.  Not like this is a carefully engineered and fully documented U.S. product (though those are largely a thing of the past anyhow...easier to rebrand Chinese products!)

Hi Sid, thanks for the reply. Appreciate it. Powerjack888 channel on youtube has guides on this, and lots of channels out there that had a manual or a guide of repair that even came with the product. I know its like trying to find a needle in a haystack, Im hoping someone out there may have a schematic I can reference to aid in my diagnosis. I got this a while ago, and I thought just changing the blown IGBTs would do the job. It worked then stopped again. There is 1 smd zener diode between the gate and source of each of the 4 IGBTs, and Im trying to figure out what the diodes should test to. The IGBTs are FGL60N100BNTD, if thats any help, but I think Im on a wild goose chase at the moment. I have new IGBTs to replace but Im not gonna try firing it up yet until Im sure all is well. There doesnt seem to be any problems on the DC side though.

Thanks a million,
Cheers.

That zener diode likely serves to prevent spikes from blowing out the IGBT gates; it should register ~0.7v with meter positive to the emitter of the IGBT, and likely somewhere under 18v with meter negative to the emitter of the IGBT.  Somewhere a little less than the max gate-emitter voltage.

Don't think the zener would be at fault. 

Thanks Sid,
I have replaced the 2 blown IGBTs and needed to replace a 1002 smd resistor. All I think looks ok, but to be on the safe side, I am putting a 2A quick blow on the AC side to save my IGBTs as they are not cheap, and are from a reputable electronics company and not ebay. It seemed to work ok before up to a point where it was working out the max power point it think, as the boogie leds went from slow to quick to faster and faster, etc, the blew the fuse and the 2 IGBTs. As I have said its a PSWGT-1200, without the lcd display. I have seen users run these units off of 24v batteries, I have 2 x 12v 80Ah Yuasa sealed units, could this be what is causing my problems? Is it because these batteries can deliver a higher wattage than say a standard solar panel and its going into output runaway. I suppose I either limit the current somehow, or run this unit from the output of a charge controller that current limits to the inverter.

Any advice always greatly appreciated.

Kindest regards
Col.

<a href="/monthly_2021_03/20210329_151728.jpg.a9f319cd937adb419f467dc7d6d39867.jpg" class="ipsAttachLink ipsAttachLink_image"><img data-fileid="210" src="//forums.genetrysolar.com/applications/core/interface/js/spacer.png" data-src="/monthly_2021_03/20210329_151728.thumb.jpg.62ef062faa79cf07cf2f9721f3139918.jpg" data-ratio="133.45" width="562" class="ipsImage ipsImage_thumbnailed" alt="20210329_151728.jpg">

Colin,

I'll donate my 2 cents worth here.  Hope you don't mind...

I have two of these I played with a couple of years ago with disappointing results.  One is a 1200w and the other is an 1800w. 

First, powerjack said in their instructions for these not to connect them to batteries.  Probably supports your theory that the batteries are causing it to 'run away' until it self-destructs.  Specifically, you are supposed to size your solar panel input to the wattage limits of the inverter.  This is from memory, as I can't find any paperwork/manuals that came with mine, so accuracy is not guaranteed.

Second, even connected to correct size panels, mine seemed to work sometimes and do nothing useful most of the time.  Disappointing. Eventually I got frustrated, put them on a shelf in the shed, and mostly forgot them.  In my opinion, they just aren't worth the time and trouble spent messing with them.  Powerjack sadly is not known for building quality products and the grid tied units are a good example of that.

Probably more productive use of funds would be getting a regular powerjack or genetry inverter that will appreciate your batteries and give you a solid output.

Sorry I can't offer more positive information.

Hi there,

Thank you so much for the very useful information you have supplied me its certainly positive in my book and the information is certainly worth more than 2 cents in value. Thanks for confirming the statement about using batteries. Do you think its worth connecting this up to a charge controller that can output 10A max?.

All your comments are well received, and great.
Shame I cant find any paperwork anywhere on this unit.

What size batteries were these "other users" running the units off of?  It's worth noting that older and smaller batteries (particularly lead-acid) will have a much higher resistance that this could sorta get away with.

Like @dochubert said, grid-tie inverters (across the board) are designed to connect directly to solar panels, no battery anywhere in the circuit.

While I've never used a grid-tie unit before, it is worth noting that the principle of operation is for the processor to find the "maximum power point" of the solar panels.  They do this by progressively loading the panels down and running a V*I calculation to find the peak wattage; at some point, the voltage will start to fall, and the processor will back off to the "peak power" setting.  It is worth noting that solar panels' "maximum power amps" is practically next door to "short circuit amps" (i.e. 7.07A max power amps, and 7.47A short-circuit on a 330W Boviet).

With a battery, however, there is no "peak power" setting until you're at a dead short--current just keeps going up as the voltage goes down.  Worst part of that is if the processor's "current sense" range maxes out just above the rated maximum power (for peak resolution in the mathematics), that means the processor will be absolutely clueless as to what's going on as the current spirals out of control and blows your IGBTs.

Not saying this is happening, nor that there isn't some fault with the grid-tie inverters.  At this point, it's really hard to say exactly what's going on.

Thanks for being so nice about my mostly negative comments.

Maybe you could try running the inverter off of just solar panels to see if it works correctly, then progress to some kind of current-limited battery input.  Just a thought....

Hi guys,

Thanks you all for such amazing input and detailed explanation as to what could be and what is going on. No worries on the comments, all greatly received
I revisited the youtube channel, doh, Im such a doughnut.

What the guy was saying was that there are panels tied to a charge controller going to the battery bank, and a change over switch to switch from the charger controller solar input, over to the input of the inverter. The batteries never actually powered the inverter, although the commentary was a bit vague.

So it looks like this thing is solar panel only, and the mppt is from the processor trying to work out what it can get before power drop and hitting max current draw, solar good, batteries bad idea. As you will never get to a point where there is a power drop, not on these 2 beasts I have as a battery bank.
All makes sense.

Guys you are all super stars, sorry for wasting your time, and I feel that I have got to the bottom of things now.

I will keep you updated when I have hooked it up to an appropriate panel.

Thanks a million to you all,
Regards

Col.

Hi all,

Please can someone help me, I now have a Power Jack PSWGT-1200 with LCD screen so its a bit more recent than my previous unit.
I have been donated this unit, however it appear to not have the current sensor inside, which is attached to the hot wire/live from the grid, which should be connected to the LCD board. Please can someone tell me where I can get one to install on this unit.
As this unit has current sensing, I take it that this is only for monitoring and calculating values displayed on the LCD. I was hoping that it may manage current output, and as such this unit possibly could be run off of batteries?

Any help or suggestions greatly appreciated,
Kind regards.
Col.

Well, could you take the current sensor from your other unit??

That might give you a clue.

Am not familiar with the PJ grid-tie inverters.

The pswgt1200 is one of the units I have.  See pic I previously posted above. 

It is still likely it will not run off of batteries, and will self sestruct if you try it. 

The current sensor for the lcd screen has nothing  to do with the unit's output current sensing and/or current limiting abilities.  The unit's control circuits  operate independently of the lcd screen.  You can unplug the lcd screen  completely and the unit will still run the same.

Hi,

Thank you for the reply, that makes sense.

Could you tell me how I can test my lcd screen on a bench without powering up the inverter at all please, what voltage does it need to power it on?
 Also, do you know where I can get a compatible CT clamp so that I can connect it to the lcd board.

Many thanks.

6 hours ago, Colin G said:

Could you tell me how I can test my lcd screen on a bench

That is something I have never done. Sorry.  For that I would have to refer you back to Sid.  I can say that the lcd screen for that inverter is no different from those used in other powerjack products, so a current transformer (ct) from almost any older model powerjack inverter would probably work.  I say older model because powerjack did change the lcd screens slightly at some point, and lcd's from newer models now have different connectors/pinouts, so they could also use a different ct.