Temperature Based Solar Disconnect

[pgoldston]

I tried searching and maybe I am not using the right terms.
I am currently designing a solar system to go off grid but still able to use about 66kw a day with battery backup. I live in south Georgia and during the lowest solar days in Jan numbers show an average of about 3 hrs a day. This brought me to sizing a 30 KW system and leaning towards 6 inverters in parallel. Reason for 6 instead of larger inverter is redundancy and if 1 inverter goes down or string issues I can balance the loads and still have a large chunk of power available. This is to power a house and large shop and trying to future proof as best as possible. Currently during the summer with 3 AC systems running and machines/house I am at about a 52KW average on the highest load days.

When sizing the inverters and running initial math for my panels/strings I am showing that I should only run 7 (710W) panels per MPPT for when it gets cold to prevent over voltage. This is using all time low temp (3 f) that was set in 1985, its south Georgia, we sweat on christmas day.... That being said with the average temp during the summer my numbers show I can run up to 8 Panels or 9 panels with a surface solar temp of 145ish. This means from winter protection to summer heat I am potentially leaving 4.2 KW - 8.5 KW per hour on the table during the summer when I need the most production.

So my question concerning this is, is there a surface temp based disconnect/bypass relay for solar panels. This does not seem like it would be a very hard build with the current relays and ICs in the market. I would think the approach would not be to disconnect when the temps got too low, rather connect when the surface temp got hot, or even just monitoring the voltage point at the last 1/2 panels in the string and connecting/bypassing them from the string as needed.....

Or am I over thinking this?

 

[sunshine_eggo]

Or am I over thinking this?

Yep. Just design it right. There's a reason this product doesn't already exist.

Optimize your panels for both Voc and PV current limits. You may have a hard on for the 710W panels (these will be a nightmare to mount), but they may not be the best choice for your inverters. You may find ~400W panels with 2 parallel strings on each inverter might be a better combo.

 

[DIYrich]

Double pole, double throw relay.
Controlled by a thermistor.
If needed, use a silicone controlled rectifier to amplify the current from the thermistor.

 

[hwy17]

When I first started thinking about my designs I wanted to push to string size too. I got over it.

Your kw and kwh's are mixed up but I think we can read through them. You mean 66kwh a day.

You should share more of the other system design thoughts. I'd rather have 2 good inverters than 6 pieces of crap. It's hard to escape entire system faults, and more inverters does not help you do so.

 

[Hedges]

I am currently designing a solar system to go off grid but still able to use about 66kw a day with battery backup. I live in south Georgia and during the lowest solar days in Jan numbers show an average of about 3 hrs a day. This brought me to sizing a 30 KW system and leaning towards 6 inverters in parallel. Reason for 6 instead of larger inverter is redundancy and if 1 inverter goes down or string issues I can balance the loads and still have a large chunk of power available.

You may find ~400W panels with 2 parallel strings on each inverter might be a better combo.

With two strings per MPPT, they can have two orientations for more hours production.
Then maybe 4 inverters in parallel, 3 standing if one fails, would be enough.

If your system was DC coupled, Midnight Classic tolerates over-voltage, by detecting and not operating. 150, 200, and 250V models (250V model costs more for less wattage.) Their newer Barcelona is higher voltage, may or may not have this feature.

There are 600V string inverters, and 1000V string inverters. Those could be AC coupled to grid-forming battery inverter.

When I first started thinking about my designs I wanted to push to string size too. I got over it.

With most inverters, you can push amperage over limits. Voltage is the one to religiously observe.
With over-paneling, it will just clip on good days, make full power on not so good days.

 

[hwy17]

make full power on not so good days

That's what I dream of now, a double paneled CC that flips full output on like a light switch in the morning. But then I look at my wallet and go back to working on my 8 panel array.

 

[Hedges]

Maybe if you scrimp and save long enough, you'll have $360 to spare and can buy another 8 panels.

Blemished Canadian 250W Solar Panel | SanTan Solar

These Canadian 250W Solar Panels have water intrusion, which has caused discoloration on the vinyl and busbars. Please see pictures for more details.

www.santansolar.com

($45.00 each)

 

[hwy17]

$360 to spare and can buy another 8 panels.

With the time it takes to install ground mount racking, I can't afford to use anything less than 480w.

 

[SamDeleted]

I tried searching and maybe I am not using the right terms.
I am currently designing a solar system to go off grid but still able to use about 66kw a day with battery backup. I live in south Georgia and during the lowest solar days in Jan numbers show an average of about 3 hrs a day. This brought me to sizing a 30 KW system and leaning towards 6 inverters in parallel. Reason for 6 instead of larger inverter is redundancy and if 1 inverter goes down or string issues I can balance the loads and still have a large chunk of power available. This is to power a house and large shop and trying to future proof as best as possible. Currently during the summer with 3 AC systems running and machines/house I am at about a 52KW average on the highest load days.

When sizing the inverters and running initial math for my panels/strings I am showing that I should only run 7 (710W) panels per MPPT for when it gets cold to prevent over voltage. This is using all time low temp (3 f) that was set in 1985, its south Georgia, we sweat on christmas day.... That being said with the average temp during the summer my numbers show I can run up to 8 Panels or 9 panels with a surface solar temp of 145ish. This means from winter protection to summer heat I am potentially leaving 4.2 KW - 8.5 KW per hour on the table during the summer when I need the most production.

So my question concerning this is, is there a surface temp based disconnect/bypass relay for solar panels. This does not seem like it would be a very hard build with the current relays and ICs in the market. I would think the approach would not be to disconnect when the temps got too low, rather connect when the surface temp got hot, or even just monitoring the voltage point at the last 1/2 panels in the string and connecting/bypassing them from the string as needed.....

Or am I over thinking this?

You could just get a 2 pole DC transfer switch and manually switch between 6 panel string & 8 panel strings as and when you choose
Maybe seasonally?

Or only when it gets cold out ...

Just don't forget lmao !!

 

[Darkirhorse]

I recently bought 26 blemished panels from San Tan, Savannah Ga for the clearance price. They're closing their doors. I would have bought more but my trailer just wouldn't hold the weight. I have 2000 sqft of house with 3 ton ac, and 2 well pumps. not interested in putting panels on my roof so I want to use solar tracking ground mounts. 6 panels per mount. concrete footers 18 inches deep?\ish. I'm good against shading and rarely see a day below 30F here in sunny Florida. I have a vague idea how this all works together. but get a little overwhelmed by the details of; transfer from grid to off grid, size of hybrid inverters, calculating my demand for current, and just how many days of battery reserve I will possibly need. I have almost settled the question of battery sizing in favour of go small and expand as needed. I'm using a baseline consumption 5.5 Kwh per day. I have not verified what that load is. Very soon though I will have to put real number in the equation. Anyone out there who has done this and reached a successful conclusion I would love to hear how it went and what you did. naturally, I would like to know what aspects of your journey did not work out so well. I like the Idea of temperature sensitive cutout control.

 

[Hedges]

I don't think tracking is worthwhile due to cost and reliability. It was marginal 20 years ago, but with panel prices 1/10th or 1/20th as much today, absolutely worthless IMHO.

Instead, make a string oriented for AM sun, another for PM sun, and wire in parallel. Using about 1.5 to 2x as many panels as you would have put on tracking mounts. Maybe more aimed at noon sun in the summer if you have heavy load then.

If you bought 7.5kW of panels and need 5.5 kWh/day, it is sufficiently over-paneled that almost any orientation should work.

Whatever inverter, look up its no-load consumption and include that in budget.

Battery, I sized for just one night's draw. My system is grid-backup. If you are entirely off-grid you might want 3 days consumption.

Read your utility bill, or look up consumer energy labels for your appliances.
Motors draw 5x their label current when starting, so inverter has to be oversize to handle that.

"temperature sensitive cutout control"
No, for PV array Voc, KISS. Using record coldest temperature, calculate adjusted Voc and do not connect enough panels in series to inverter PV input voltage spec.
For lithium battery, if temperature gets below freezing, have a BMS which prevents charging.