What makes a grid tie inverter TURN ON?

[Chuck Yerkes]

many panels. bright sun. grid tie. No power.

Understood. Accepted.

What, on the inverter's grid side, happens to make the inverter turn ON
and let power flow from the grid and from the panels into my happen
little house?

 

[Anthony Matonak]

many panels. bright sun. grid tie. No power.

Understood. Accepted.

What, on the inverter's grid side, happens to make the inverter turn ON
and let power flow from the grid and from the panels into my happen
little house?

Grid tied inverters don't control what power flows into your house.
Your house is still plugged into the grid the same way it always was
and when the grid power is restored your house would have power. The
inverter simply looks at the power coming from the grid and makes
sure it's stable before it starts pumping power back into it. In a
way, the inverter is pumping power into the entire grid and it just
happens that your house is also connected to the grid nearby.

Anthony

 

[Chuck Yerkes]

Grid tied inverters don't control what power flows into your house.
Your house is still plugged into the grid the same way it always was
and when the grid power is restored your house would have power. The
inverter simply looks at the power coming from the grid and makes
sure it's stable before it starts pumping power back into it. In a
way, the inverter is pumping power into the entire grid and it just
happens that your house is also connected to the grid nearby.

Okay, so how does the inverter define "stable"? Is there something more
than just a token sine wave at a couple milli-amps?

Where I'm trying to go with this is that for a grid tie system, in an
outage during solar generation time - what is needed to 'trick' the
inverter into allowing the solar power to flow (with grid cutoff via the
kill switch of course).

Now, granted, a big ol cloud can drop your solar output to a couple amps
and that would harm things plugged in. But if I have something on the AC
side of the inverter that can temporarily pick up the holes, then I
should be set.

What that SOMETHING is is what I'm trying to figure out. If I'm down to
5 amps (lights off cause it's daytime, but perhaps just fridge and
minimal stuff) then a 1000watt generator would do it. Or a *small*
battery system powered by just a couple panels an the alternate small
inverter.

I'm just straining because to me I should be able to take a 500-1000
watt battery backed system and use that to kick on a 4-5kw system.

How minimal a system is needed for this? If 2 panels can charge up
batteries that are used ONLY for backup, then you can have many
batteries, they will charge eventually. Once charged, any "extra" can
be used to the grid on house side of the kill switch.

Lose power? Hit the kill switch, get the second inverter to go into
generate mode and 4kw of roof solar flows. In darkness you have 1kw.
In bright sun, you have 4kw + 1kw.


What's wrong with this picture?

 

[Anthony Matonak]

Okay, so how does the inverter define "stable"? Is there something more
than just a token sine wave at a couple milli-amps?

Yes. I believe they typically measure a few different things and one
of them is how much the input from the inverter affects the power on
the grid. Since the grid generators are huge compared to a home sized
inverter their output, and hence the grid waveform, isn't going to be
affected much by the inverter.

Where I'm trying to go with this is that for a grid tie system, in an
outage during solar generation time - what is needed to 'trick' the
inverter into allowing the solar power to flow (with grid cutoff via the
kill switch of course).

Since inverters are designed specifically to not do this then I'd say
you would probably have to work hard to fool it. You would do better
to simply get an inverter + battery combination that does what you
want in the first place. That is, you want an inverter/charger/battery
that will maintain a battery bank and provide your house with power
automatically when the grid fails. Such things exist but they are
a little more expensive than straight grid tied inverters.

Anthony

 

[William P.N. Smith]

many panels. bright sun. grid tie. No power.

No output from your grid-tied inverter, if there's no grid.

Understood. Accepted.

If you want power when the grid is out, get a solar power system
that'll sell power back to the grid when the batteries are full (like
Outback). Tricking a UL-???? inverter into turning on when the grid
is out is _supposed_ to be a hard problem.

 

[William P.N. Smith]

You could (theoretically and practically too I think) take a grid tie
only inverter (sunny boy, ST, PVPowered etc.) and parallel that with a
sine wave battery powered inverter. The battery inverter being the
"grid".

Theoretically, you could, but in the real world there aren't any
four-quadrant battery inverters like that available to the RE market.
There's probably some lab equipment that could do it, but it's not
going to sell for a buck a watt...

Just as a zeroth problem, what will you do when the solar power input
exceeds the house load?

 

[[email protected]]

----- The following addresses had permanent fatal errors -----
<[email protected]>
(reason: 550 User unknown)

Greetings Mr. Guppy,

UL1741 defines a number of parameters , Frequency , Voltage (both single
cycle and multicycle values , which are different tolerances)...

I was about to order a copy until I discovered it was $795, or a mere $335
in pdf form I gather it's 145 pages, and the last rev was in 1/17/01.
Do you have a copy? Would you send me the parameters and limits in
a sentence or two or fax me a page or two listing them?

I'd like to help make a small grid-tie inverter with a 350W 90%-efficient
Veco "filtered MSW" inverter with a 2-year guarantee ($49 at Wal-Mart, who
also sell its $69 700 W bigger brother) and a standard $5 600 W lamp dimmer
and a $5 current limiting choke and a $10 PIC for MPPT (with an analog output
for the dimmer) and anti-islanding and writing a "Small Guerilla PV" story
about it, in an effort to inspire some company to get a small UL-1741
inverter back on the US market, which might be reasonably simple repackaging
of the one in the story. I'm thinking the lamp dimmer filter should meet
the rms distortion requirement (if not, why not, ie why should utilities
be more concerned about source than load distortions, unless they are
trying to give grid-tie a hard time?)

This might be an interesting small system, with a single 12 V battery and
a way to get some 120 V during a power outage...

Thanks for your help, if possible...

Nick

Nicholson L. Pine System design and consulting
Pine Associates, Ltd. (610) 489-1475
821 Collegeville Road Fax: (610) 831-9533
Collegeville, PA 19426 Email: [email protected]

Computer simulation and modeling. High performance solar heating and
cogeneration system design. BSEE, MSEE, Sr. Member, IEEE. Registered
US Patent Agent. Web site: http://www.ece.villanova.edu/~nick

 

[William P.N. Smith]

If there isn't a large enough load for the Sunnyboy (or any battery-less GT
inverter) to put its power into the grid , the voltage will simply rise and
the UL1741 high line will trip the inverter off-line.

And then the battery inverter will stabilize the voltage, and (after
some timeout) the grid-tie will come back on, rinse, lather, repeat.
Sizing the battery inverter, the grid-tie inverter, and the house
loads becomes another non-trivial engineering task. Not that it can't
be done, but IMHO it's another piece of complexity you don't
need/want.

 

[William P.N. Smith]

And at that point, you wouldn't need the extra support of the extra
grid tie inverter until some more AC loads came on.

Sure you would (might), unless the battery inverter can carry the
whole house load, and then you might as well have done the Outback
"full solar but sell the excess to the grid" route.

 

[William P.N. Smith]

I am not suggesting this is the best implementation , rather , you indicated
it is only done in the lab and I was informing the group SMA has guided
actual installations to use this typology

Fair enough, though my brain hurts when I try to imagine a stable
control configuration that would work properly. Do they tie the two
units together in some way so the battery inverter can try to balance
the load, battery charger, and instantaneous solar power input?