Jump to content
Genetry Solar Forums

Off grid in Idaho


Recommended Posts

Not completely off grid, but working on it.  Main house 48v battery bank charged by (48)  320w solar panels.  Water heater 24v bank powered by another (8) 320w panels (so far - more coming this summer).  Auxilliary 12v battery bank in the shed powered by (12) assorted sizes solar panels.

Not enough wind for windmills here.  No water flow for hydro power, so solar is all I've got at the moment.  Wintertime here means short days and long nights, so grid is still necessary for us at present.  Depending on weather, from sometime in April to sometime in October we can be completely off grid.  That's about 2/3 to 3/4 of the year off grid so not unhappy, but would like to figure out a way to kick the grid habit full time.

I have several powerjack and upower inverters.  3 are in use daily.  2 of my oldest 15kw powerjacks are in need of new control boards.  The original v3.6 control boards are just too old to repair again.  Tried some v8 control boards  from Sean without success.  Awaiting the new Genetry control boards with the hope they will do the job.

Link to comment
Share on other sites

Yet another thing I need to write a manual for... 😉

Yes, the GS upgrade is relatively straightforward...that is, if you're comfortable with repinning JST XH connectors, cutting and splicing fan wires, adding a few wires, pretty much tearing your inverter completely apart, soldering SMD zener diodes on every MOS board, maybe even shortening the ridiculously long ribbon cable to the MOS boards (and not breaking or reversing the connector), removing all the big boards...and putting it back together again with what looks like "nothing" in comparison.  Took me about 3 hours IIRC. 

Please note that while Power Jack loves to use dual mainboards, I do NOT recommend that on a Genetry Solar control board.  GS control board Rev. A.1 (which we have in stock as of right now) does have dual outputs, but the Rev. B (coming in the 2nd shipment of GS inverters, and future upgrade kits) only has a single output.  Heck, a properly driven MOS board with 6 FETs should easily be able to sustain 600A without blowing up.  Now if they're poorly driven...they'll blow up pretty easy, or literally randomly.  As of yet, we have not had a single set of FETs burn out with a GS control board (providing the zener diodes were installed on the FET boards!)

Part of the reason that a 15kw inverter has warm/hot FETs at no load is because the LF driver simply is not able to cleanly switch the FETs on and off with a SINGLE mainboard.  Double the capacitive load, and the switching waveform becomes atrocious.  Perhaps as a result of this, it is very easy to determine if an inverter has a PJ or GS mainboard in it (assuming a literal double-blind test): not only is the transformer practically silent in a GS inverter at no load, the inline choke on the transformer primary is going to be cold in a PJ, and hot (quite hot!) in a GS.  Difference?  Firmly and cleanly driving the FETs.

Link to comment
Share on other sites

I do have one powerjack that came with dual mainboards, but removed the 'extra' mainboard almost immediately.  (seemed like a bad idea right from the start!)  It runs better now.  That one is not the model I'm needing to put control boards in. 

I have (3) of the oldest style 15kw powerjacks in the aluminum clamshell cases.  Two of these need control boards. One still powers the house like a champ!  V3.6 control board and all!  Tried v8 pj control boards from Sean in the other two back in Oct (I think? maybe sept).  Failed miserably. I suspect v8 control boards don't mix well with 230v transformers. 30v difference after all.  To my knowledge, v8 boards shipped in powerjacks with 260v transformers only.  Maybe I'm wrong...

Anyway, my pj's have 32v/230v AS2 transformers (3 each in parallel).  I need a control board that can accept that voltage range.  As Sean promised me two of the Genetry control boards when they become available, I'm practicing my Zen patience.  I know you guys are busy.  A lot on your plates!  (Congrats on your new place Sean!)

I'm hoping to get these two inverters up and running before the third one dies.  I will definitely need your instruction manual for the conversion, especially if I have to modify mosboards.  The inverters were already converted to v8 controls if that helps.  If not, I'll change to whatever works.  Pic below shows running pj on top and one of the non-working ones underneath.



Link to comment
Share on other sites

Yeah, extra mainboard was a PJ thing to try to increase the power rating...a bit like adding a 2nd hog to the creek to clear the water up.  Problem is the completely inadequate and imbalanced LF driver.

So I guess I've seen your posts over on another forum with that super long beast...sorry it's been a pain, but we should be able to get this cleared up.  Providing the transformers themselves are not shorted, faulty, miswired, or any other issue like that.  Obviously, you need a set of (working) MOS boards as well.

PJ control boards (at least v4.0 through 10.4) could care less what the transformer voltage is.  All of them regulate the SPWM "throttle" based on the output voltage.  Unless the transformers are miswired, or they are expecting 110v instead of 220...there should not be any issues.  Same with a GS control board...now, if the transformer ratio is a little too low for the battery voltage range, you'll get more or less of a square wave output as the board tries every trick in the bag to try to maintain the output voltage.

No conversion for the MOS boards is required, just the addition of a small zener diode on each one.  Or if your Zen patience is almost limitless, maybe you could wait until I order the first batch of GS replacement MOS boards with all the required parts already included ;-).


When you say the v8 PJ control boards "failed miserably", what happened?  Smoke?  Alarm?  Nothing?

Link to comment
Share on other sites

I have seen pictures of Power Jack main boards that didn't have the removable little mosfet boards is this the kind you have my friend

If you're pretty comfortable with the electricity you can do a test on your Transformers

you can use a old style filament light bulb limit the current

You get a 120 cord off of a old Appliance the light bulb is hooked up in series not parallel on only one wire of the cord

then you put a light switch across the light bulb so that when the light switch is on off position the power has to go through bulb

that one side of your cord has to go through the light bulb to limit your current if switch is on then the light bulb will NOT  limit the current 

You then connect the cord up to the output of your Transformer only one 120 volt leg 

After you get this completed correctly making sure the low side nor the other leg of 120 of your Transformer can get shorted out on anything 

Make sure your switch is in off position so the power goes through light bulb only 

you can then plug your cord into a wall outlet 

After plugged in and if every thing is ok you flip the light switch so the power goes through the switch 

Then you test the voltage of the low side that normally is hooked to inverter board  this will give you the exact low side voltage 

I will add a pic

please only do this if you have a good understanding of electric

  you could DIE if not done right

  I am not response if you or anyone else dies because of lack of understanding 

Very dangerous!!!!!

make sure to hook up all wires in a well insulated way and all unused wire insulated  and light switch in off position before plugging into out let

if light is dim not fully lit you then flip light switch to on position

the voltage on you meter will be the low side voltage of your tranformer  

Without bulb it will flip your breaker the light bulb limits the current so it doesn't flip breaker and it limits current if transformer is bad 

Dangerous do at your own risk!!!

Ask questions if not fully understood!!!!  DEATH!!!!16134620288521225449716398741430.thumb.jpg.4fa3cb6d2cb6f05849238959d4eac4fa.jpg ELECTROCUTION!!!!

Link to comment
Share on other sites

Hi Ben and Sid,

Thanks for the replies and good info.

Ben, I have been planning on fully testing the transformers in much the way you pictured, except I was going to use a variac powered on the low side and read voltage on the high side.  Maybe I'll do both tests.  I'm reasonably sure the xfmrs are good, but need to be sure before risking more parts. Thanks.  Safety warnings noted!

Sid, Since I could only guess, it's good to hear that transformer voltage range includes my 230v model transformers.  As I said, I think mine are ok but will be fully testing them to be absolutely sure.  Here's how the v8 thing went down;

Both inverters were set up side by side with new mainboards/mosboards from Sean.  A new v8 control board and v8 controls installed in each.  Call them inverter A and inverter B.  Turned A on.  Lights lit up properly. Seemed fine.  Output voltage was 60/120v instead of 120/240v.  ??  Turned it off.  Turned on B.  Lights pulsed similar to powersave except faster, then in about ten seconds audible alarm and red lights.  Turned it off not wanting to blow fets.  No, I didn't accidently switch to powersave position.  No, the switch wasn't wired backwards.  Tried it twice more to be sure.  Same results.

Pretty much everybody else said  "You did something wrong." or "It's in powersave mode" 

This was not my first powerjack rebuild.  I have blown up two other inverters in the past because of MY stupid mistakes.  I'm not infallible and perfectly willing to be wrong, so;

1.  Checked carefully.  Not in powersave mode on unit B.

2.  Removed and then swapped the v8 control boards, leaving controls and everything else the same.  This time A pulsed and alarmed, and B ran perfectly fine at half voltage.  Only this time, on the third try of A pulsing, the fets blew.  At that point I was frustrated and done with v8 boards, having spent over $600 trying to get parts that worked.  Conclusion was that the symptoms followed the control boards.  Neither control board worked properly and each had unique results regardless of which inverter it was installed in.

The results from switching of control boards told me that the wiring, controls, and transformers were likely just fine.  Yes, I still need to do a definitive test of transformers, and I will.  (I shouldn't have put it off this long.)

All that said, I have high hopes for your new control boards and look forward to finally getting my two inverters running.  Thanks for listening.




Link to comment
Share on other sites

OK, so I do see one thing right off with the one v8 control board....

29 minutes ago, dochubert said:

A new v8 control board and v8 controls installed in each.  Call them inverter A and inverter B.  Turned A on.  Lights lit up properly. Seemed fine.  Output voltage was 60/120v instead of 120/240v.

I'm assuming this 15kw beast is a "split phase" inverter.  Power Jack has a (very frustrating for me!) habit of changing the CPU program as an absolute last resort--they will change anything on the PCB to make CPU code written during the Egyptian Dynasty still function in 2021.  Regardless, it sounds like you are running into one of those grand issues.  They actually change 2 SMD resistors on the PCB to set the voltage feedback to the CPU...and it sounds like the v8 control boards were from "110v" inverters, and your transformer connections are actually in the "220v" mode. 

On the v8 control board, these resistors are on the CPU carrier board (not the CPU board, but the board it plugs into).  Sean and I figured this one out some time back, but it wasn't worth the headache to replace the SMD resistors.

The quick solution would be to miswire the transformer outputs to the control board in 110v mode--which PJ does quite often, actually.  (That's how they get 110v charge with a split-phase transformer.)  Instead of L1-L2 going to the control board (and N directly to the front panel), you would wire L1-N to the control board, and L2 would go to the front panel.  Yes, the CPU can regulate 110v feedback just fine.


23 minutes ago, dochubert said:

Ben, I have been planning on fully testing the transformers in much the way you pictured, except I was going to use a variac powered on the low side and read voltage on the high side.  Maybe I'll do both tests.  I'm reasonably sure the xfmrs are good, but need to be sure before risking more parts. Thanks.  Safety warnings noted!

Glad to hear that you have a variac.  Those things are so invaluable that I don't have one...wait, how's that go again?  For figuring out tranny problems, there's nothing better.  You can do the test either way (from the high side, or from the low side); what you're looking for is an excessive current.  PJ has been known to get a couple of strands of wire one turn short (or crossed between windings), this would show up as an excessive no-load current (even if voltage does come out the other side of the transformer.)

Please note that toroidal transformers can have a severe inrush current--it's been several times that I've flipped the variac switch on (variac set to full voltage), and instantly tripped the breaker.  With a perfectly good transformer, mind you.  (This is partly why the inverters have soft start...the other being that the CPU doesn't know what kind of transformer is connected, so a ramp-up from zero is the safest way.)

As those are AS2 transformers, no-load current on the (high) AC side should be under 20W (I would expect 10W) for each of them.  If you're connecting the variac to the low side, as you slowly turn the input voltage up, the high side of the tranny should swing right up--if it hesitates or doesn't freely move, you're dealing with a bad tranny.


If swapping control boards moved the problem from the first inverter to the second, then you almost certainly have a faulty control board and/or something miswired/misconfigured on it.

  • The one board working at half voltage is functioning properly.  Just set for 110v feedback, not 220v.
  • The "pulsing" one sounds a bit like one of the resistor grid AC sense boards is not properly connected/missing, or there's a bad connection between the "AC board" and the "control board."  Basically, it's running to 100% throttle over and again, trying to "find" the transformer, but it's not getting any AC feedback.  Probably sounded pretty nasty as it did this.

Blowing FETs with that...sheesh, I roll my eyes at the (frankly) terrible FET driving.  Sean can testify that I had a code bug in a GS inverter that he was testing...a register overflow that wasn't properly handled.  At any rate, it resulted in an absolutely horrendous AC SPWM waveform--and he heard the GS inverter angrily buzzing (at several KW load, too!) from 2 floors up--and he has a house full of noisy kids.  I'm so glad he had the presence of mind to take a photo of the waveform on his oscilloscope before shutting the inverter off--the photo helped me identify the code bug instantly.  But the FETs are completely unharmed/undamaged.  Like I said, to date, we haven't lost a single set of FETs on a properly set up GS inverter.  (Blew 3-4 sets of FETs before I figured out what the issue was, though!)

Link to comment
Share on other sites

Is there a "110v" sticker on the one control board that was running half voltage?  That would be a dead giveaway.

Oh, and for the laughs, here's a comparison between the PJ boards for a 9kw (v9) inverter, and the GS replacement boards for the EXACT inverter:


This Rev. A.1 GS board can safely handle 60-70 amps (85 amp terminals, 120A relay, 100A current sensor).  Rev. B is even smaller than this one, though it's only 50A (relay limitation), but that's still 12kw @ 240v.

Link to comment
Share on other sites

Before I forget...Sean kinda has his hands tied on the GS boards currently in stock.  We discovered a few weeks ago that PJ forgot to include the screws for the terminals--kinda hard to connect to the boards without the screws for the terminals!  I asked them to express ship the screws to Sean, and (assuming they didn't forget again), hopefully those will show up this week.


BTW If the inverters ran perfectly fine at half voltage with the "good control board" in either of them, then we can pretty well assume that the transformers are not the problem.

Link to comment
Share on other sites


Thanks for the explanations!  I really appreciate that!

I could rewire for 110v control, but would rather not.  I prefer to wait for your 240v control boards (waited this long).  I'm pretty much done with v8 boards.

I do like the new simpler v10.3 boards powerjack is using on their upower models.  I would try one of those on my older pjs but have no idea how to set the 10 switch pack (SW1) and no one will publish a setting chart for them.  Guessing is not a great idea....  Also they seem to only exist in 110v config.  No idea if they can take 220v.

Oh, many thanks for confirming what I have been saying; The fact that the issues moved with the control boards and that the half voltage unit did so with both inverters tells me neither set of transformers are faulty (probably).  I will still test them.

For your amusement, sending a pic of the v2.2 control board that came with both my oldest powerjacks.  They seem to be about the size of your new control board.  Yes, they are 230v control boards with L1 and L2 connected and neutral/center tap only going straight to output connector. 


Link to comment
Share on other sites

The "10.3" control boards they manufacture now are literally identical to the V8 boards you have tried, just without the modular approach.  Exact same circuit.  And they're based on exactly the same design principle as the v2.2 board you have there.  Had to laugh when I saw photos of the control board on a 15kw Yiyen inverter...comically, it uses the EXACT same design principles as the PJ boards.  Goodness, is there ANY variety in the Chinese inverter market?!

Have to say that Jack (PJ) is probably looking at the GS control board thinking, "Yup, that's what ours looked like when we started, stupid Americans..."  Well, the funny thing is that he's now seen 3 revisions of the GS control board...and each subsequent revision has been SMALLER than the one prior to it.  I may or may not be trolling them...and since I've hit the mounting screws, I can't make the board any smaller 😆.

We will see if a bag of 400 M5 screws with double washers make it to Michigan this week.  2 mods will need to be made to the GS boards before they're shipped to you (thermal correction, adjust output filter caps), but otherwise, the design is solid.  It is worth noting that the GS boards are almost useless without a WiFi board--almost 100% of the settings/configuration are done through the WiFi board. While you can sorta "share" one WiFi board between 2 GS boards (plugging the CPU comm cable into one or the other), it won't be bench-pretty solution.

30 minutes ago, dochubert said:

I would try one of those on my older pjs but have no idea how to set the 10 switch pack (SW1) and no one will publish a setting chart for them.

That switch is nothing more than an assortment of resistors that you can switch on/off to pull down the output current feedback.  Sets the output current limit and charge current.  It by no means determines system setup--I wish it did, but alas, no.  Battery voltage is set by jumpers all over the board.

Link to comment
Share on other sites

Sid, I've learned more about powerjack inverters this afternoon from a few of your posts than elsewhere in the last year. Thanks so much!

You implied that your GS control boards won't work without a wifi board.  Not knocking the wifi board, but I didn't intend to get one as I've had independent fan controls and output meters that I'm happy with for years so didn't think it would do me any good.  Also, not interested in being connected to internet or remote control of the inverter.  So bottom line; Will the GS control board work for me without a wifi board?  Or is it necessary? 

If you say I need one, guess I'll be buying one.

Thanks again

Link to comment
Share on other sites

I do want to chime in here and say that the 110V sticker is not an indication that the board itself is 110V. Power Jack orders these boards and they always put a 110V sticker on them since they can be used for both 110V and 220V. The only difference is a 220V board gets an output board either as a sorta daughter board or built into the master board. I have seen it both ways.


Power Jack sent me a case of 10.4 boards and all of them say 110V but they can actually be used as both. They started to put output connections on the master board to be used as 110V. So when I need to warranty a 110V inverter I simply discard the daughter board and it works just fine. Also note that the new 10.3+ inverters are split phase in the sense that they have a SP transformer but the L1 half phase is actually connected directly to the front panel and never passes through the daughter board for filtering. That is how they are able to use 110V boards in 220V SP inverters. In V8 and older inverters N went to the front panel from the transformer while L1 and L2 went into the daughter board for filtering and the inverter would refuse to run if either of those were missing.

Bottom line. Stickers dont mean anything.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...