Off-the-grid with grid-tie inverter? - Page 2

Xringer · 03-16-10, 11:28 AM

Yeah, I guess it could be simpler..

The cheapest SunnyBoy GTI I've seen is over $1,000
SMA Sunny Boy 700U Grid-Tie Inverter, 700 Watt, 150, 200, 250 Volt
Putting out 700 watts (if you have 700 watts of PV panels).

For that price, you could get ten of the cheap Chinese 300 Watt GTIs (3,000 watts)..
If you had 10 panels, you could reap the benefits of power-point-tracking on ten panels..
And, all those panels could be of different brands & power levels.

I'll bet that SunnyBoy has to have the whole array coming in on one pair.
Meaning that a shadow one panel brings down the whole system power..

NiHaoMike · 03-16-10, 08:44 PM

Put a high efficiency DC/DC converter on each panel. It would actually be more efficient that way.

Xringer · 03-16-10, 09:56 PM

Quote:

Originally Posted by NiHaoMike

Put a high efficiency DC/DC converter on each panel. It would actually be more efficient that way.

Not if you wanted AC...

Ryland · 03-16-10, 11:14 PM

you are going to have issues if your PV systems output is higher then your demand as it's not going to have any load, the grid is acting as a load, back feeding a small inverter as a load is going to do damage, not sure how fast it will let the magic smoke out of the small inverter but it will happen, or fuses will blow, either way it will shut down.

Xringer · 03-17-10, 12:10 AM

If you have a few Enphase GTI AC outputs tied together in parallel,
what keeps them from destroying each other?

I do see what you mean with the excessive PV power.
The GTI will want to dump that PV power to the grid (or your fake grid), in the largest quantity it can..

Just thinking about it, it seems you would need your load,
be large enough to use all the power from the PV array
and a little bit of power out of the battery..?.

I think the Enphase monitors the AC line in small time slices and then
puts it's own replica AC on the line during it's output cycle.
That cloned AC from the Enphase is slightly boosted in amplitude,
to push power onto the grid.

The nice thing about resistive loads and some motor loads is,
when the voltage does up, the load draws more current (uses more power).
I=E/R

I guess my oddball configuration working is going to be very dependent
on the functionality of the GTI.. Those cheap Chinese GTIs might not
would with this kinda of scheme.

Ryland · 03-17-10, 09:49 AM

The voltage will stay the same and the amps will stay the same as well, that is where either the grid tie inverter will shut down or the sign wave inverter will burn up, the grid tie inverter is designed for unlimited demand, it's not designed to be limited.

NiHaoMike · 03-22-10, 12:04 AM

Quote:

Originally Posted by Xringer

Not if you wanted AC...

That reminds me of a story my ECEN 441 instructor told. A contractor from NASA asked him about what power system to use onboard the space station. His response was DC. One company had tried to build a 20kHz power system and they could not get it to work properly. Then, based on the note the instructor wrote, another company built it on DC. It worked and DC is what they use today!

The answer seemed almost too obvious. Batteries are DC, as are solar panels. Electronics also run from DC. So why waste energy converting to AC only to convert it back to DC? Motors often run on AC, so use a point-of-load inverter that supplies exactly the right frequency and voltage needed at that time.

Xringer · 03-22-10, 12:45 AM

The main reason that utilities use AC is because very high voltage can be used.
High voltage and low amperage can deliver the power, to transformers.
(Without the use of monster cables).

Those transformers are about 99% efficient and can step-down the voltage so
we can use it in our homes.

DC melts transformers..

NiHaoMike · 03-22-10, 11:49 AM

AC makes sense for long distances, but not for short distances. A few hundred or even few thousand feet is not considered long distance.

Xringer · 03-22-10, 02:35 PM

At the lower voltage ranges (that you might need in a house), DC voltage needs
a Large amount of current..

So, if you have a few thousand feet of wire, it has to be Monster cable..
Otherwise, it will have significant resistance..

Notice that voltage isn't a factor in this formula, P=R x I squared..
In other words, if you have some Resistance, the Power loss will go up with the square of the current..

That's why we no longer use any long low voltage DC lines..

When I was a young man working in Boston, one of the old Phone co. buildings had 110VDC outlets in the offices.
Once in a while, someone would plug in an AC device and watch it smoke..

03-16-10, 08:44 PM	#12
NiHaoMike Supreme EcoRenovator Join Date: Oct 2008 Location: Austin, TX Posts: 1,154 Thanks: 14 Thanked 257 Times in 241 Posts	Put a high efficiency DC/DC converter on each panel. It would actually be more efficient that way. __________________ To my surprise, shortly after Naomi Wu gave me a bit of fame for making good use of solar power, Allie Moore got really jealous of her...

03-22-10, 11:49 AM	#19
NiHaoMike Supreme EcoRenovator Join Date: Oct 2008 Location: Austin, TX Posts: 1,154 Thanks: 14 Thanked 257 Times in 241 Posts	AC makes sense for long distances, but not for short distances. A few hundred or even few thousand feet is not considered long distance. __________________ To my surprise, shortly after Naomi Wu gave me a bit of fame for making good use of solar power, Allie Moore got really jealous of her...

03-16-10, 11:28 AM	#11
Xringer Lex Parsimoniae Join Date: Feb 2009 Location: Woburn, MA Posts: 4,918 Thanks: 114 Thanked 250 Times in 230 Posts	Yeah, I guess it could be simpler.. The cheapest SunnyBoy GTI I've seen is over $1,000 SMA Sunny Boy 700U Grid-Tie Inverter, 700 Watt, 150, 200, 250 Volt Putting out 700 watts (if you have 700 watts of PV panels). For that price, you could get ten of the cheap Chinese 300 Watt GTIs (3,000 watts).. If you had 10 panels, you could reap the benefits of power-point-tracking on ten panels.. And, all those panels could be of different brands & power levels. I'll bet that SunnyBoy has to have the whole array coming in on one pair. Meaning that a shadow one panel brings down the whole system power..

03-16-10, 11:14 PM	#14
Ryland Master EcoRenovator Join Date: Mar 2009 Location: Western Wisconsin. Posts: 913 Thanks: 127 Thanked 82 Times in 71 Posts	you are going to have issues if your PV systems output is higher then your demand as it's not going to have any load, the grid is acting as a load, back feeding a small inverter as a load is going to do damage, not sure how fast it will let the magic smoke out of the small inverter but it will happen, or fuses will blow, either way it will shut down.

03-17-10, 12:10 AM	#15
Xringer Lex Parsimoniae Join Date: Feb 2009 Location: Woburn, MA Posts: 4,918 Thanks: 114 Thanked 250 Times in 230 Posts	If you have a few Enphase GTI AC outputs tied together in parallel, what keeps them from destroying each other? I do see what you mean with the excessive PV power. The GTI will want to dump that PV power to the grid (or your fake grid), in the largest quantity it can.. Just thinking about it, it seems you would need your load, be large enough to use all the power from the PV array and a little bit of power out of the battery..?. I think the Enphase monitors the AC line in small time slices and then puts it's own replica AC on the line during it's output cycle. That cloned AC from the Enphase is slightly boosted in amplitude, to push power onto the grid. The nice thing about resistive loads and some motor loads is, when the voltage does up, the load draws more current (uses more power). I=E/R I guess my oddball configuration working is going to be very dependent on the functionality of the GTI.. Those cheap Chinese GTIs might not would with this kinda of scheme.

03-17-10, 09:49 AM	#16
Ryland Master EcoRenovator Join Date: Mar 2009 Location: Western Wisconsin. Posts: 913 Thanks: 127 Thanked 82 Times in 71 Posts	The voltage will stay the same and the amps will stay the same as well, that is where either the grid tie inverter will shut down or the sign wave inverter will burn up, the grid tie inverter is designed for unlimited demand, it's not designed to be limited.

03-22-10, 12:45 AM	#18
Xringer Lex Parsimoniae Join Date: Feb 2009 Location: Woburn, MA Posts: 4,918 Thanks: 114 Thanked 250 Times in 230 Posts	The main reason that utilities use AC is because very high voltage can be used. High voltage and low amperage can deliver the power, to transformers. (Without the use of monster cables). Those transformers are about 99% efficient and can step-down the voltage so we can use it in our homes. DC melts transformers..

03-22-10, 02:35 PM	#20
Xringer Lex Parsimoniae Join Date: Feb 2009 Location: Woburn, MA Posts: 4,918 Thanks: 114 Thanked 250 Times in 230 Posts	At the lower voltage ranges (that you might need in a house), DC voltage needs a Large amount of current.. So, if you have a few thousand feet of wire, it has to be Monster cable.. Otherwise, it will have significant resistance.. Notice that voltage isn't a factor in this formula, P=R x I squared.. In other words, if you have some Resistance, the Power loss will go up with the square of the current.. That's why we no longer use any long low voltage DC lines.. When I was a young man working in Boston, one of the old Phone co. buildings had 110VDC outlets in the offices. Once in a while, someone would plug in an AC device and watch it smoke..