Author Topic: Power jack inverter as charger from GTI inverter  (Read 4571 times)

0 Members and 1 Guest are viewing this topic.

Offline sunnypower46

  • Jr. Member
  • **
  • Posts: 33
  • Karma: +0/-0
Re: Power jack inverter as charger from GTI inverter
« Reply #15 on: January 26, 2019, 01:19:57 am »
For lighthunter and all, a little more to add, if you please.  I am now dumping excess microinverter wattage via halogen "tubular" bulbs.  I'm using a 4-channel voltage comparator that samples the PCM voltage output of the Morningstar unit with some RC filtering and an emitter follower transistor stage for buffering.  This provides a comparator input voltage of 0-13vdc.  The comparator channels are staggered to activate their respective output relay at inputs of 9.0, 9.5, 10.0 and 10.5 vdc.  The relays forward 12vdc to four separate solid state AC relays that power the bulbs.  I've designed in about 4 vdc of hysteresis for the comparator to slow down the turn on turn off action of the system. 

Each output channel, when activated, triggers its SSR and 240VAC is sent to two series connected 240 VAC 500 watt bulbs.

The bulbs roughly share the voltage, operate at about half their wattage each, but still total 500 watts per channel.  I did it this way so they weren't so damn bright and would potentially have a greater lifespan.  This works very well in practice.  Now if I could just get them to stop giving off heat . . . .

In actual operation, groups of bulbs turn on and off as loads switch on and off, according to the amount of excess energy that needs to be dumped beyond what the Morningstar unit can handle.  Because the bulbs are "soft" resistive loads, the PowerJack powering the local grid can handle it ok and the microinverters stay happy.

A final observation regarding microinverter usage with PJ style inverters:  While DC operating voltage of the inverter and type of battery and cable size and storage amount (AH) are factors, I think it's the dynamic response of the PJ style inverter on the AC side of things that's the controlling factor.  The feedback loop just isn't fast enough to satisfy the needs of the microinverters when it comes to controlling voltage spikes.  Also, consider that up to 14 microinverters on a branch are all trying to stay in sync with each other while our PJ tries its best to keep things stable under changing (especially reactive) house loads.  That's a tough job in the real world.