Trying to Design a way to run my Solar Panels with No storage, when Grid is down

[clinquist]

I am an EE and have Tesla panels and inverter, but no Powerwall. I want to start a project that would allow me to use the power from my panels when the grid is "down". Of course, I would never back-feed a dead grid connection. But I have the following questions, someone has to know the answers.:

Does the Inverter determine when the grid is down by not producing any output near zero-crossings of the incoming power, and if the zero-crossings are not present, it assumes the grid is down and shuts off?

If I disconnected the grid "side" of the inverter, and instead drove it with a 60Hz sine wave, would the inverter start? How much power (Watts) would be needed to turn on the inverter? And if I simply made the output voltage of my sine-wave source larger than the maximum output voltage of the inverter, I assume that the inverter would not be able to back-feed my sine-wave source. But what IS the maximum voltage output of the inverter - 250Vm 255V?

I assume that PowerWalls have their own 5KW, 60HZ sine wave inverter. Is that correct?

 

[Silicon Desert]

I don't know anything about the Tesla inverter, however I have done inverter design consulting for SMA and SolarEdge in the past. The short answer is that for those inverters your idea is not going to work and I suspect the same is true for the inverter you have.

Yes, inverters detect the voltage and frequency from the grid.

 

[clinquist]

Of course I know that the inverters detect voltage and frequency from the grid. My question involves HOW they measure it. If the inverter happens to be sending power to the grid, there will normally be voltage on the grid "side" of the inverter even if the grid is down. So the inverter cannot simply measure the voltage on the grid "side" to determine if the grid is energized and disconnect itself. An inverter could "miss" a cycle - say once a second and watch the line during that cycle, but that would create noise and flickering. It would seem that the only way it could determine grid failure is if the Solar Inverter didn't drive the line at all between - say +/- 10V of the incoming sine wave. Since power is proportional to voltage squared, there is very little power in those six degrees around zero-crossing. If the inverter didn't power the system during that time, it could "watch" for the voltage going to 10V (or so) before it started driving anything. it would not only synchronize itself with the line, it would automatically shut down if the grid went down.

Please tell me where I'm wrong, and why my scheme won't work.

 

[wwhitney]

Are you going to build a grid-forming inverter yourself from scratch, or acquire one? How are you going to handle the instantaneous microgrid power balancing? The grid-following PV inverter will by default just want to blast out as much power as it can.

Cheers, Wayne

 

[clinquist]

Are you going to build a grid-forming inverter yourself from scratch, or acquire one? How are you going to handle the instantaneous microgrid power balancing? The grid-following PV inverter will by default just want to blast out as much power as it can.

Cheers, Wayne

I guess I don't know the what a grid-forming inverter is, but my goal isn't to build a solar inverter from scratch - at least not yet.
I have built sine wave inverters before, using PWM to generate a good sine wave. I was thinking that the Tesla inverter would probably run if I fed it 10-15 Watts of a crystal and PLL-controlled 60Hz at a little over 353 Vp-p (250 RMS), and I would like someone to tell me if I'm on the right track.

 

[wwhitney]

I guess I don't know the what a grid-forming inverter is, but my goal isn't to build a solar inverter from scratch - at least not yet.
I have built sine wave inverters before, using PWM to generate a good sine wave. I was thinking that the Tesla inverter would probably run if I fed it 10-15 Watts of a crystal and PLL-controlled 60Hz at a little over 353 Vp-p (250 RMS), and I would like someone to tell me if I'm on the right track.

Suppose it does (not sure), what then? It's going to try to put out some number of kW, depending on how many panels you have, sun angle, etc. Where is that power going to go, at every instant in time? If you want to set up a microgrid, you either need a combination inverter/charger and batteries that can sink any excess power from the PV, or you need a dump load and controller that you can set up to sink any excess power, or a way to tell the PV inverters to adjust their power output at a very very fine grained level.

Cheers, Wayne

 

[clinquist]

Suppose it does (not sure), what then? It's going to try to put out some number of kW, depending on how many panels you have, sun angle, etc. Where is that power going to go, at every instant in time? If you want to set up a microgrid, you either need a combination inverter/charger and batteries that can sink any excess power from the PV, or you need a dump load and controller that you can set up to sink any excess power, or a way to tell the PV inverters to adjust their power output at a very very fine grained level.

Cheers, Wayne

You don't need to "soak up the excess power" with any kind of load. If you have full sun on 8KW of panels, your Powerwall is full and your house is consuming 1KW you have nothing to "soak up" the extra 7KW. It is a mistake to think that just because a system can produce a certain amount of power it always must (or will).

As I mentioned, I don't have a Powerwall. My first goal is to use some relatively inexpensive means of keeping my house powered when the grid is down and the sun is up. My second goal is to add my own battery storage. And it all has to work with my existing Tesla panels and solar inverter. I designed lots of power supplies for the military at the circuit-board level, so I can design/build nearly anything. I could start instrumenting everything on my system and find out for myself, but I really don't want to blow something up simply because I made the wrong assumption.

 

[wwhitney]

You don't need to "soak up the excess power" with any kind of load. If you have full sun on 8KW of panels, your Powerwall is full and your house is consuming 1KW you have nothing to "soak up" the extra 7KW. It is a mistake to think that just because a system can produce a certain amount of power it always must (or will).

That is incorrect. If the grid is down, the PWs are charging from excess PV power, and the PW state of charge approaches ~ 95%, the PW inverters will raise the AC frequency above 60 Hz to get the PV inverters to stop producing energy. The microgrid must have a power balance at all points in time--instantaneous power produced (by any inverters) = instantaneous power used (by loads and any battery chargers).

The PW will do the same thing if you have a mismatch between PV inverter capacity and PW charging capacity--pair one PW with 10 kW of PV inverters and enough PV panels, and off grid on a sunny day when the PV inverters start producing more than about 5-7 kW of power net, that one PW can't "soak up all the excess power" and it will raise the AC frequency to knock the PV inverters off-line.

In order for your plan to work, you are going to have to recreate a Powerwall from scratch.

Cheers, Wayne

 

[CrazyRabbit]

Solar inverters act as a current power supply, not your typical voltage supply inverter. You would need a auto transformer to form the neutral leg, inverters only put out 240v, no neutral. Best bet would be to buy a Powerwall.

This is why enphase created a new inverter that is grid forming, iq8.. also requires their transfer switch.

A solar installer told me his dad use to run his generator with a SMA inverter in parallel.