UTILIZING GRID TIE INVERTER OFF GRID - $100 REWARD FOR SOLUTION

[Hedges]

I watch them ramp up slowly. That is the MPPT drawing PV voltage down.
But a load dump happens instantly. Question is what island grid voltage does, and how GT PV responds. I think of GT PV inverter as SMPS for current, with current being a sine wave proportional to voltage. Or maybe proportional to a software PLL following grid voltage.

I'll try the motor-generator, set to run no-load at a frequency which will decrease GT PV output to zero watts but not disconnect. SB off-grid default setting, 62 Hz < no-load frequency < 64 Hz. For a

UL-1741-SA inverter, not clear that it is supposed to remain connected indefinitely when above 61 Hz, the 0W point.

Playing with TriPower, I ended up in a mode where frequency-watts ramps down from 100% at 60 Hz to 0% at 60.5 Hz.

You would want PV not battery to keep motor-generator turning no-load, but don't want to overcharge either. I think this system would be most optimum for hydro plant, with water flow adjusted to regulate frequency. Typical micro hydro would be Pelton wheel with constant flow, DC output into charge controller. It will either go to 2x RPM no-load, or else need a dump load.

 

[zanydroid]

UL-1741-SA inverter, not clear that it is supposed to remain connected indefinitely when above 61 Hz, the 0W point.

1741SA adds ride-through requirements. So it might try to stay up a little bit longer than a 1741 inverter.

Or maybe proportional to a software PLL following grid voltage.

It is proportional, with the ability to change phase shift and amplitude according to how much apparent power it wants to send out, and whether it is injecting leading / lagging reactive power

You would want PV not battery to keep motor-generator turning no-load, but don't want to overcharge either. I think this system would be most optimum for hydro plant, with water flow adjusted to regulate frequency. Typical micro hydro would be Pelton wheel with constant flow, DC output into charge controller. It will either go to 2x RPM no-load, or else need a dump load.

What would be the advantage of this vs using an AC coupled battery inverter's frequency shift control loop? I mean I think it would be cool especially if you can rig it up to be stable with purely analog and mechanical controls, very steam punk. But I mean besides that.

Also I'll note that OP seems to plan to just plug everything together rather than start with a theoretically stable point.

 

[Hedges]

1741SA adds ride-through requirements. So it might try to stay up a little bit longer than a 1741 inverter.

The frequency-watts is "with utility approval". Given that power ramped down to zero, it could sit there indefinitely, ready to ramp up again. Or disconnect for 5 minutes.

Compare to non frequency-watts with full power during ride-through, then disconnect.

It is proportional, with the ability to change phase shift and amplitude according to how much apparent power it wants to send out, and whether it is injecting leading / lagging reactive power

Proportional to actual voltage seen on grid, with all distortions? Or proprtional to a perfect sine wave synchronized to grid?

What would be the advantage of this vs using an AC coupled battery inverter's frequency shift control loop? I mean I think it would be cool especially if you can rig it up to be stable with purely analog and mechanical controls, very steam punk. But I mean besides that.

Steam Punk wouldn't fry transistors. It might fry windings.
I try to be a Luddite as much as possible.

Battery inverter supporting phase shift for 6kW GT PV inverter will run $3000 to $10000.
If a motor-generator just 10% the size of GT PV inverter could make a working system, that could be interesting. But getting it to also keep battery optimally charged would be more difficult. AC powered battery charger?

So I think GT PV add-on to full-time hydro might be an attractive niche application.

Also I'll note that OP seems to plan to just plug everything together rather than start with a theoretically stable point.

Without frequency-watts, I don't expect much. Only good for grid that is bottomless. Or extra capacity on a SolArk, which has greater amount of DC coupled PV to ramp up/down while switching OP's inverter on/off.

 

[zanydroid]

Proportional to actual voltage seen on grid, with all distortions? Or proprtional to a perfect sine wave synchronized to grid?

This, I'm not sure of. I suspect the PLL will lock onto the fundamental (frequency and phase), some kind of volt-meter will take the RMS voltage. And the PWM controller will start from there.

Battery inverter supporting phase shift for 6kW GT PV inverter will run $3000 to $10000.
If a motor-generator just 10% the size of GT PV inverter could make a working system, that could be interesting. But getting it to also keep battery optimally charged would be more difficult. AC powered battery charger?

So I think GT PV add-on to full-time hydro might be an attractive niche application.

Ah that makes sense. What guidelines do you follow for stable AC coupled battery inverter sizing?

Where does the energy/power go in the motor generator and hydro case? I assume the coils will need to be able to eat the VAs that will be going through them. Also does it matter that the GT may not like driving a highly inductive load?

 

[Hedges]

Ah that makes sense. What guidelines do you follow for stable AC coupled battery inverter sizing?

Many brands say PV inverter no larger than battery inverter. SMA says PV up to 2x battery, but wind/hydro only 1x. This may utilize the nearly 2x surge rating and a couple seconds to ramp down. Maybe battery charge will also surge.

I think it is an issue of load add/dump step size, not total watts. What if PV was 10x battery? If insolation doesn't change to rapidly, and load add/dump step does not exceed battery inverter capacity, might keep working.

Where does the energy/power go in the motor generator and hydro case? I assume the coils will need to be able to eat the VAs that will be going through them. Also does it matter that the GT may not like driving a highly inductive load?

Without PV, of course some regulation mechanism.

With PV, my idea is no-load RPM increases until PV curtailed. Does applied AC from PV spin up the motor faster? Wires can take a surge current much higher than nominal. For transformers, short-circuit current might be 20x, 30x operating current.

Should work with gradual load change. Load dump of a resistance heater switching off, leaving only motor-generator connected, might cause backfed island grid voltage to violate limits GT PV expects. When SB is set for offgrid, it has wider voltage/frequency limits. Not sure about rate of change.

My SB will remain connected with 2HP VFD operating offgrid (SI), but if waking up and they look out and see that same waveform, they think they have some internal fault and quit.

 

[zanydroid]

I think it is an issue of load add/dump step size, not total watts. What if PV was 10x battery? If insolation doesn't change to rapidly, and load add/dump step does not exceed battery inverter capacity, might keep working.

Yes, one thing I'd like to understand more are the assumptions / risk analysis tree they use in coming up with these recommendations.

With PV, my idea is no-load RPM increases until PV curtailed.

Yeah, this makes sense. I'd love to see how well/not well this works. Do you think most cheap non-inverter generators (just mentioning the non-invertor part so that others reading this later will be aware that you need direct coupling of the generator output to the generator) would have the right properties?

 

[Hedges]

Maybe, and that is more likely to have wattage similar to inverter.
You might have to adjust it to get RPM you want.

Page 9 shows generator frequency droop with load:

https://files.sma.de/downloads/SIGEN-11FD1212.pdf

Depending on response, it might work with GT PV directly connected. This paper is about systems with SI in between.

Inverter generators, I imagine tapping into DC bus for PV. Same for VFD.

 

[shadowmaker]

Woah ok
"I'm not sure, but for pure off gird setup a cheaper off grid inverter should do the same"
Meaning, my setup might work
My setup: High voltage input -> PVi6000 (grid tie inverter) =>> (2 AC outputs, one to cheap 2nd inverter, and one to Panel box powering tiny home)
Am i understanding correctly?

I think the off-grid needs to be at least as "big" to rule over the on-grid one.

 

[joesmith123]

I just got a chance to reread the new posts, I will be responding to them once I get the chance

Just want to say this update: 12VDC-240VAC inverter that was ordered, was cancelled yet again

New idea: Create the mini grid using TWO separate 12v-120VAC inverters

Those I can get easily and readily

The setup: 220VDC (prius) -> PVI6000 high voltage input -> L1 going to first 120VAC inverter and L2 going to 2nd 120VAC inverter

Predict what you think will happen, I want to set this up asap and get the show moving forward

 

[joesmith123]

I think the off-grid needs to be at least as "big" to rule over the on-grid one.

Yes, I will get huge inverter (6000 watts) once I am sure the setup will work, first I want to lower cost options just to see if it'll even function

 

[joesmith123]

Playing with TriPower, I ended up in a mode where frequency-watts ramps down from 100% at 60 Hz to 0% at 60.5 Hz.

It sounds like the Grid tie inverter shut down once the hertz went up to 60.5, meaning i think it worked

 

[Hedges]

Predict what you think will happen, I want to set this up asap and get the show moving forward

The two 12V inverters will have unsynchronized output, so L1 to L2 will vary from 0V to 240Vrms

It sounds like the Grid tie inverter shut down once the hertz went up to 60.5, meaning i think it worked

Yes, but my point was Rule-21 settings are 100% up to 60.5 Hz, ramping down to 0% at 61 Hz.
Offgrid settings are 100% up to 61 Hz, ramping down to 0% at 62 Hz, and don't disconnect up to 64 Hz.

This one did 100% up to 60 Hz, ramping down to 0% at 60.5 Hz. I've found a new operating mode (inadvertently flipping different setting in GUI)
It would also hover around 60.25 Hz depending on load.

 

[zanydroid]

This is still so overly complicated / not well founded in theory...

There are Meanwell AC-DC and DC-DC power supplies that can take your Prius pack voltage as input. We've been talking about that on a couple of threads (ChargeRectifier and DC DC car charging). If you only need like 1-2 kW of capacity you might as well just go that path - step down from pack voltage to standard 24V or 48V and tack on an inverter. No AC coupling.

For perspective on the theory... they pay a lot of plant operators good money to AC AC synchronize all the generators on the grid, they don't just connect them together and pray. They think through how to do the volt and frequency synchronization, etc. before starting parallel operation. I don't think Westinghouse YOLO'd into scaling out AC system back in the late 1800s...

 

[DIYrich]

New idea: Create the mini grid using TWO separate 12v-120VAC inverters

One inverter to an auto transformer to convert 120v to split phase 240v. Solar edge is the cheapest I found.

Or buck boost 12v to 48v, and use a 48v dc to 240v ac inverter.

 

[joesmith123]

One inverter to an auto transformer to convert 120v to split phase 240v. Solar edge is the cheapest I found.

Or buck boost 12v to 48v, and use a 48v dc to 240v ac inverter.

Yes, I will look into this (the 120vac->240vac using auto transformer)

 

[joesmith123]

here's the test

 

[cs1234]

Where's the fire and brimstone? You must not have done something correctly for there to be so little destruction.

 

[400bird]

Damn, it didn't even make it to the point where the PV inverter started to do anything. The PV inverter has a 5 minute timer before it will output. Does that mean the little battery inverter couldn't even full the capacitors in the PV inverter?

Considering the test length, I'm not surprised it didn't blow up. This isn't the point I expected the test to fail.

 

[timselectric]

It's dark outside.
Waiting for the actual test.
That's when it will burn.

 

[joesmith123]

What I think happened is as soon as the big inverter started to produce some power into the small inverter. Instantly the small inverter was overwhelmed, and went into the protection mode.