I did a thing! Sunny Island, growatt min ac coupled system

[1201]

What wattage and where did you get them?

400w , signature solar

 

[1201]

For clarity

I bought the Talesun 400w bifacial panels at .27/watt with free shipping. The $1250 cost is my total cost after the 30% tax credit, which I qualify for.

Payback should be less than 3 years

 

[Johneb]

Hmmmmm. Looking toward my dear wife who has that oh no look.

 

[kommando]

Another option which is on the EU version and hopefully in the US version too but this needs checking

Found out that the Solax X1 Boost G3, which is a cheap and freely available new UK G98 inverter, can be made to control its output via the connected Grid frequency.

https://community.victronenergy.com/que ... erter.html

The two parameters that need adjusting are

FreqSetPoint (Htz)

and

FreqDropRate (%)

So for a Sunny Island these would be set to 51Htz and 10% so that by 52Htz the output is zero.

You also have to set the Grid country to one that supports these parameters, eg EN50438_NL, VDE4105 or User Defined

and also play around with parameters to improve stability

Fac Upper and Fac Upper slow

These parameters are also on the G4 version but hidden a bit further down within Grid Services.

 

[1201]

Another option which is on the EU version and hopefully in the US version too but this needs checking

Found out that the Solax X1 Boost G3, which is a cheap and freely available new UK G98 inverter, can be made to control its output via the connected Grid frequency.

https://community.victronenergy.com/que ... erter.html

The two parameters that need adjusting are

FreqSetPoint (Htz)

and

FreqDropRate (%)

So for a Sunny Island these would be set to 51Htz and 10% so that by 52Htz the output is zero.

You also have to set the Grid country to one that supports these parameters, eg EN50438_NL, VDE4105 or User Defined

and also play around with parameters to improve stability

Fac Upper and Fac Upper slow

These parameters are also on the G4 version but hidden a bit further down within Grid Services.

This is available on the USA sunny island. It's called frequency shifting power control.
. The settings are not adjustable. It just keeps raising the frequency until it gets the power level it's looking for. I tweaked the growatt inverter to set my zero output frequency at 60.5hz

 

[kommando]

This is available on the USA sunny island.

I should have been more explicit, I was referring to the EU Solax having the ability to be output controlled by a SI using frequency so maybe a US Solax could also have the same feature. This would then add the Solax to be another inverter that the SI can control to the SMA SB's and the Growatt.

Both US and EU SI's do frequency control, US is based on 60htz, EU is 50htz (well my EU also has the 60 htz option on initial setup but only at 220V for some reason)

 

[1201]

I should have been more explicit, I was referring to the EU Solax having the ability to be output controlled by a SI using frequency so maybe a US Solax could also have the same feature. This would then add the Solax to be another inverter that the SI can control to the SMA SB's and the Growatt.

Both US and EU SI's do frequency control, US is based on 60htz, EU is 50htz (well my EU also has the 60 htz option on initial setup but only at 220V for some reason)

oh. that's great to know about another pv inverter that works well AC coupled to a sunny island. Thanks.

 

[1201]

Brought another 5.6kw of dirt cheap solar online.

Unirac rm10 mounts. Paid $400 for 25 of these. Drove 4 hours to pick them up. I didn't want to spend $2500 to install them on the roof. That would kill my ROI. With $1750 in the panels($1250 after tax credit) and $450 in the mounts( got to count the gas money) if these make the 7-8mwh a year I'm anticipating, they will pay themselves off in 2-4 years, depending on how much I self consume and how much I export to the poco




I'm experiencing quite a bit of clipping during the middle of the day but I can't justify adding another string inverter right now.

 

[Hedges]

I'm experiencing quite a bit of clipping during the middle of the day but I can't justify adding another string inverter right now.

Consider making some panels tilted steeper than those, half oriented SE and half SW.
That could reduce clipping, widen hours of production.

 

[1201]

Consider making some panels tilted steeper than those, half oriented SE and half SW.
That could reduce clipping, widen hours of production.

That's not a bad idea. I could do two strings instead of one.

Do you think over a year span that would result in greater overall output? I don't know a good way to model that

 

[kommando]

My last purchase of panels is for an east/west array to smooth out the curve, panels were £35 each, inverter an SB3600TL-21 was £75. I did the simulation of the output using this tool.

JRC Photovoltaic Geographical Information System (PVGIS) - European Commission

Photovoltaic Geographical Information System (PVGIS)

re.jrc.ec.europa.eu

There must be a US alternative, this site showed me I would get a flatter curve and add two more potential grid free months per year. All my other panels are south facing.

 

[MajicDiver]

Brought another 5.6kw of dirt cheap solar online.

Unirac rm10 mounts. Paid $400 for 25 of these. Drove 4 hours to pick them up. I didn't want to spend $2500 to install them on the roof. That would kill my ROI. With $1750 in the panels($1250 after tax credit) and $450 in the mounts( got to count the gas money) if these make the 7-8mwh a year I'm anticipating, they will pay themselves off in 2-4 years, depending on how much I self consume and how much I export to the poco

View attachment 200528


I'm experiencing quite a bit of clipping during the middle of the day but I can't justify adding another string inverter right now.

View attachment 200529

Another way to avoid the clipping would be to add a battery, that inverter would charge another 10kW at the same time - kill your ROI though…….

 

[1201]

My last purchase of panels is for an east/west array to smooth out the curve, panels were £35 each, inverter an SB3600TL-21 was £75. I did the simulation of the output using this tool.

JRC Photovoltaic Geographical Information System (PVGIS) - European Commission

Photovoltaic Geographical Information System (PVGIS)

re.jrc.ec.europa.eu

There must be a US alternative, this site showed me I would get a flatter curve and add two more potential grid free months per year. All my other panels are south facing.

I've used that tool a lot. I'll send you a pm to learn how to model different facing arrays at once.

Thanks!

 

[1201]

Another way to avoid the clipping would be to add a battery, that inverter would charge another 10kW at the same time - kill your ROI though…….

Lol. Yes that would kill my ROI. But, I'm not quite understanding -How would it avoid clipping?

 

[timselectric]

It looks like you could definitely reorient some panels for evening production. And get better overall daily production. This also keeps you from dipping into the battery until later in the day.
It looks like morning is probably as good as it can be.

 

[MajicDiver]

How would it avoid clipping?

The inverter can use up to 11.4 + 10 = 21.4kW of PV power, 11.4 in AC power, another 10 to battery, if there is enough PV. I’ve never had that much PV and load at same time though, and batteries are already charged by then.

 

[1201]

It looks like you could definitely reorient some panels for evening production. And get better overall daily production. This also keeps you from dipping into the battery until later in the day.
It looks like morning is probably as good as it can be.

That's definitely something I need to think about. Thanks!

The inverter can use up to 11.4 + 10 = 21.4kW of PV power, 11.4 in AC power, another 10 to battery, if there is enough PV. I’ve never had that much PV and load at same time though, and batteries are already charged by then.

Oh ! I had no idea. That's cool!

 

[Hedges]

That's not a bad idea. I could do two strings instead of one.

Do you think over a year span that would result in greater overall output? I don't know a good way to model that

Estimate amount of clipping, how much total power would be available if within SCC limits.
Subtract 2% from total as penalty for strings of different angles not being exactly at maximum power point.

Manufacturers always said to have all panels on a single MPPT same orientation. Installers wanted to do otherwise due to multiple roof orientations. SMA conducted a test with strings of same length but different orientation paralleled, reported total kWh about 2% less than with each orientation on its own MPPT.

The W/V curves have peak at slightly different voltage, but with strings paralleled they operate at same voltage, slightly reduced power.

If two parallel strings have equal number of panels shaded, their voltage will track. If one has a panel shaded but other doesn't, it is OK for similar number of unshaded panels, e.g. 9 in one string and 10 in the other, again because slightly off maximum power point voltage they still produce similar power.

If one string is 50% shaded and the other has no shade, e.g. 10s unshaded || 5s unshaded + 5s shaded, you'll only get the power of 10 panels. if separate MPPT would have been 15 panels.

When one string of different orientation gets no sun due to angle, it of course produces no power, but it doesn't pull down the other. Voc in the shade is about same as Vmp in direct sun, so it neither sources nor sinks (if SCC is pulling max power.)


Another good thing about East and West (SE and SW) orientation is battery is kept charged later, starts recharging sooner. Lower depth of discharge at night. Avoids missing out on PV power if you are both charging and operating loads during the day.

 

[kommando]

I've used that tool a lot. I'll send you a pm to learn how to model different facing arrays at once.

Thanks!

Not found a way to do it on the website, I instead do each string individually and export the monthly values as a CSV file and pull them into a spreadsheet. Then I add a total row at the bottom so I can see the total for all the strings. Then a row of the monthly KWH used and see which months exceed this plus 10%. This only works if you have batteries to exploit all the PV output.

You can use the hourly irradiation figures to get a sample days output per string and do the spreadsheet again to see the E/W array complementing the south arrays.

I also have the E/W nearly vertically mounted so the early/late sun is optimised.

 

[Johneb]

What tool are you using?