Genetry Solar Updates 3/7/2020 Live

Thanks for the update,  though I posted a question to Sid on the GS Other list, about the thing of the DC charging and over loading the mosfets.  Makes me concern on over useing the main GS for chargeing and powering my house, I wonder if useing the slave as the DC charger would be better off and can be done to save the mosfets from over working.  I think a Simple GS DC charger should be made.

12 minutes ago, The Blind Wolf said:

I think a Simple GS DC charger should be made.

You can run a GS inverter in "charge only" mode, where it'll only ever run in charge mode to take generator power to charge the batteries.

4 minutes ago, Sid Genetry Solar said:

You can run a GS inverter in "charge only" mode, where it'll only ever run in charge mode to take generator power to charge the batteries.

But, that would be one expensive charger. . . . and I think that if I got master and slave GS inverts, that the slave can't charge if the input is bieng used to power the breaker box. . .

1 minute ago, The Blind Wolf said:

But, that would be one expensive charger. . . . and I think that if I got master and slave GS inverts, that the slave can't charge if the input is bieng used to power the breaker box. . .

Unless the slave can take over the load while the master switch to charging only then back to master mode after the charging is done..

Cause if I got the 6k master maxed out while the slave is buffing the remainder load and the batts needing to be charged, that would push the mosfets, and the way sean was saying I would be basicly cooking that poor things. . .

We will have the cable interference problem dealt with in all outgoing inverters.  Sean ordered some ferrites via Amazon Prime, they should arrive Tuesday hopefully.  Expect to see a 12kw load test video after they arrive...well, for you, hear a load test video 😉

FETs should be good to >30kw continuous @ 48v (15kw @ 24v), so that should not be a problem for you.

That starts to get quite complicated...technically possible though requiring the inverters to work together to switch modes, etc.  Maybe a later project--or something a central system controller could be set up to do.

EDIT: I take that back.  Slave taking over the load while the master switches to charging.

Definitely not going to be the easiest to throw together.  But it MIGHT be somewhat feasible, as if the master inverter switches to charge mode, the generator output will pass straight through the inverter to the output.  The slave "daisy chained" inverter will of course immediately drop out when the transfer takes place, then pop back on again afterwards...while the "master" inverter charge current will be somewhat limited, it might possibly work.

Though the master inverter can't throttle the slave inverter down, as it'll be in "charge" mode, and frequency is controlled by the generator....

...if the voltages don't line up, it could turn into quite the can of worms.

Hmm. but, your saying that it shouldn't be a issue with the modsfets if I'm charging with the master while running the house then?  Just don't want to run into a issue with over loading them like sean did with the 12k, though that was due to the rainbow cable mainly.  I think on the PJ they have their own charging board, and on the GS it don't?

1 minute ago, The Blind Wolf said:

I think on the PJ they have their own charging board, and on the GS it don't?

The GS inverter is based on the PJ design concept, though rethought and completely redesigned from the ground up.  PJ just has so much clutter on their design that they overflowed it into a separate "charge" board.  The only thing the GS inverter design is missing (compared to the PJ design) is an AC input choke and filter cap.  But the charge principle is very similar.

2 minutes ago, The Blind Wolf said:

Just don't want to run into a issue with over loading them like sean did with the 12k, though that was due to the rainbow cable mainly

With the bigger tranny in the 12k (maybe the lower primary resistance??), the FET problem is severely exacerbated.  But we have found a solution, which will be implemented in all GS inverters going out.

Basically, the L_MOS FETs take turns holding one transformer primary wire to battery negative, while the other pair runs an SPWM cycle.  Somehow due to EMI, a considerable amount of noise is showing up on the H_MOS that is supposed to be in off-state for the whole half-wave.  Hundreds of not-so-little spikes induced by the switching activity on the other side are actually partially switching the H_MOS on, causing it to sink a considerable amount of heat by shorting the battery out.  (That the spikes are in the resistive range of the FET is the reason the inverter doesn't instantly blow out.)  The higher the load on the inverter, the stronger these spikes--and consequently the more heat the FETs generate, until they blow up due to overheat.

Fortunately by adding ferrite chokes to the rainbow cable, we can squelch all of this off-state nonsense--but Sean had to order some ferrite chokes to test.  If they work, he will have to order more ferrite chokes in order to get the full order of inverters out.  Future inverters will have this fix implemented from the factory.

Okay, well as long as mine is done, since I will be useing two 6k dasy chain and one be tied into the gen to charge the batts when needed, I just don't want to over load the master.  Of course I will ahve more solar panels up and charging to keep the batts charged during the day, but there are weeks where we have such a cloudy sky and rain for days on end that thats not going to help charge the batts.

You better believe all GS inverters going out will have this fix implemented.  Have to say that's one of the good things that has come out of the shipping delays...finding potentially serious problems, and fixing them before they ship.

NOTE: I posted the update in the wrong subforum, so moved it to the appropriate section....