diy solar

diy solar

Finally, the start of my 25kw Ground Mount grid-tie system

I've been thinking about the panel angle. The less angle, obviously the more power I will get in the summer and less in the winter. Now, I've read that for all around production your angle matching your latitude is roughly best. That would be 47*. The problem with that is that really becomes a "wind sail" and I don't think any of the configurators online even allow you to do that much. It seems 35* is the most.

Here is 47*
January2.122,404163
February3.223,212218
March4.454,894333
April5.335,585380
May6.026,339431
June5.815,780393
July6.987,011477
August6.916,933471
September5.945,982407
October4.094,487305
November2.392,568175
December1.972,222151
Annual4.6057,417$ 3,904

Here you can see 30* (what I've been thinking of doing).

January1.842,086142
February2.922,910198
March4.294,730322
April5.475,748391
May6.476,817464
June6.436,401435
July7.587,627519
August7.137,174488
September5.765,817396
October3.764,129281
November2.112,260154
December1.661,880128
Annual4.6257,579$ 3,918

Any thoughts/opinions?
 
I think that after a while, those fasteners loosen up due to the teeth on the clamps settling into place on the soft aluminium.
I wonder if an initial period of thermal expansion and contraction of the metals in the rails and PV frames also works to loosens up clamps?

I'll get up on the garage to check my own small off-grid array before Summer storm season arrives. I had the original installers of my grid-tied system come out to do a check earlier in the year, about 2.5 years after installation.
 
I've been thinking about the panel angle. The less angle, obviously the more power I will get in the summer and less in the winter. Now, I've read that for all around production your angle matching your latitude is roughly best. That would be 47*. The problem with that is that really becomes a "wind sail" and I don't think any of the configurators online even allow you to do that much. It seems 35* is the most.

Here is 47*
January2.122,404163
February3.223,212218
March4.454,894333
April5.335,585380
May6.026,339431
June5.815,780393
July6.987,011477
August6.916,933471
September5.945,982407
October4.094,487305
November2.392,568175
December1.972,222151
Annual4.6057,417$ 3,904

Here you can see 30* (what I've been thinking of doing).

January1.842,086142
February2.922,910198
March4.294,730322
April5.475,748391
May6.476,817464
June6.436,401435
July7.587,627519
August7.137,174488
September5.765,817396
October3.764,129281
November2.112,260154
December1.661,880128
Annual4.6257,579$ 3,918

Any thoughts/opinions?

The angle you set your panels at will (should) depend on what you're doing with the energy as its being produced.

For example, I have a net metering contract, I pump energy into the grid and the power company has to give it back for free whenever I want it.. The power company is acting like a battery all year long, so with this in mind, I used PV Watts to obtain maximum energy production for the year.. I don't care when its produced, I want the most kilowatts per year I can get from the array.

However, lets look at another scenario where a person is using a battery system.. Since the battery is a finite storage medium, and since summer days are much longer and you get more radiation, you would increase the angle of the array to more benefit the winter, since that's when the solar horror show begins.

You'll have to figure out the terms of your net metering.. While I'm grandfathered in, the new net metering laws in Michigan are now set up where the POCO sells you energy at retail rates (17c/kWh) and buys it back at wholesale rates (around 4c/kWh). That would be a world of suck.. and I haven't done the math on that to figure out how best to set things up. Lucky for me, I won't have to worry about it until 2027, by which time, lithium cells should be far more available.
 
You'll have to figure out the terms of your net metering.. While I'm grandfathered in, the new net metering laws in Michigan are now set up where the POCO sells you energy at retail rates (17c/kWh) and buys it back at wholesale rates (around 4c/kWh). That would be a world of suck.. and I haven't done the math on that to figure out how best to set things up. Lucky for me, I won't have to worry about it until 2027, by which time, lithium cells should be far more available.

Just for "fun", figure out the cost per kWh for batteries and battery inverters.
Then compare that to the cost of putting in more PV, so you deliver 4.25kWh to earn a credit that buys one kWh from POCO.

My situation is $0.15/kWh off-peak, $0.50/kWh on-peak, so similar but not quite as bad a ratio. Since the numbers are higher, PV is competitive even if I give them 3.3kWh per 1.0 kWh received. Of course to the extent I can, I avoid drawing power during peak times.
 
Just for "fun", figure out the cost per kWh for batteries and battery inverters.
Then compare that to the cost of putting in more PV, so you deliver 4.25kWh to earn a credit that buys one kWh from POCO.

My situation is $0.15/kWh off-peak, $0.50/kWh on-peak, so similar but not quite as bad a ratio. Since the numbers are higher, PV is competitive even if I give them 3.3kWh per 1.0 kWh received. Of course to the extent I can, I avoid drawing power during peak times.
I've been interested in the flywheel systems.. Super Capacitors were on my list but the price for them is just silly.. but the fly wheel thing holds promise.
Given the increase in the number of EV's hitting the road, when there's enough of them, they'll start hitting each other, hitting trees, etc.. If the lithium batteries can come down into the $40/kWh range, that would be ideal.. and we're almost there. My Chevy BOLT EV cells only ran me $80/kWh shipped to my door..
 
the new net metering laws in Michigan are now set up where the POCO sells you energy at retail rates (17c/kWh) and buys it back at wholesale rates (around 4c/kWh). That would be a world of suck.
That's pretty much the case across Australia.

The reason is the all the distributor and network charges (as well as the retailer's weighted wholesale value of energy and margin) are included in the import tariffs, while the export tariff has none of the extra charges applicable, so you are only paid for the energy value. Here the network/distribution costs make up ~2/3rds of the tariff. So an export to import tariff ratio of ~1:3 is normal.

But grid tied solar is a no brainer here financially because solar PV is way cheaper to install here than in the USA.

Grid tied batteries however, financially speaking they suck dog's ball.
 
My Chevy BOLT EV cells only ran me $80/kWh shipped to my door.
I'll start thinking about batteries at that price. One issue I have is cold temps in the winter. What is the performance of these batteries when they get cold? I would think if they can work in a car they must be able to stand freezing temps. Your previous post looks like your batteries are in a shop, I would have to install them in an unconditioned barn, or I could put them in a tack room in the barn that we keep just above freezing in the winter.
 
I'll start thinking about batteries at that price. One issue I have is cold temps in the winter. What is the performance of these batteries when they get cold? I would think if they can work in a car they must be able to stand freezing temps. Your previous post looks like your batteries are in a shop, I would have to install them in an unconditioned barn, or I could put them in a tack room in the barn that we keep just above freezing in the winter.
My batteries are in my home's basement in a steel enclosure.

As soon as I find another deal on good cells (dreaming!), I plan to expand capacity to 100kWh's and move the entire thing out to a purpose build shed. Being in Michigan, that means I'll have to install some temperature controls.. small air conditioner in the winter, good insulation and a small heater for the bitter winter cold.

No charging below the freezing point of water.

Its not the cold weather that is all that harmful, its being stored in high heat that does most of the damage.. since our battery is for emergency backup during the long power outages here in the rural countryside, the cells spent most of their life doing nothing at 3.6 volts.
 
My SunXtender AGM cost $250/kWh capacity delivered.
I operate down to 70% DoD, and supposed to deliver 650 cycles to that depth.
That makes cost $0.55/kWh of power cycled. If used until wear-out, which I don't expect. Just backup so should die of old age instead and probably cost me > $2.50/kWh.

At this price per kWh it doesn't make any sense for peak shifting (except if there are demand charges for highest draw, which my rate tariff doesn't have.) I use it to keep a few things running at night during outages, and for grid-forming during the day so AC coupled PV can power loads. Maybe 20% of my power goes through the batteries, the rest is use-it-or-lose-it direct from PV.

If you can get cells for $80/kWh capacity, assuming 80% DoD cycling 3650 times over 10 years, that's $0.027/kWh of power cycled. That is similar to PV + GT inverters over 10 years (although those should last 20 years for lower power cost.) With BMS and battery inverter added, could still be competitive with grid.

Seems to me at $80/kWh the price is already low enough for application which cycle fully every day, e.g. peak shaving or offsetting part of your utility consumption. If battery bank has to be oversized to make up for days without sun, that costs extra (but might give more years of service.) Similar with PV; on-grid you capture 100% of production, while for off-grid you have to buy extra capacity to make up for times with less sunlight.
 
Seems to me at $80/kWh the price is already low enough for application which cycle fully every day, e.g. peak shaving or offsetting part of your utility consumption. If battery bank has to be oversized to make up for days without sun, that costs extra (but might give more years of service.) Similar with PV; on-grid you capture 100% of production, while for off-grid you have to buy extra capacity to make up for times with less sunlight.
Yes, offsetting my utility consumption would be just the thing at this price. I have a grid tied Fronius inverter backfeeding a subpanel. I need to read up on this, what device do I need that senses when the inverter is not sufficient for the house load and starts using the batteries?
 
If you want battery to feed loads only, zero backfeed, then current transformers on the main utility feed could be read by a suitable inverter.
If backfeed is OK, then you might just use a timer to enable inverting from battery during peak use rates.

AC batteries like Tesla PowerWall and SMA Sunny Boy Storage would have this as a standard feature (possibly still requiring multiple devices to implement it.)

I use Sunny Island, which serves as a UPS and senses grid current through it to downstream loads (or from PV inverters.) I don't think it has a full set of grid interaction functions, but it can do zero export (when connected to a generator it prevents backfeed by disconnecting), and it can limit current draw from grid to a programmed current (e.g. 15A) by inverting from battery. But that wouldn't support allowing more current from grid when battery depleted.

For you, it would depend on your battery inverter. Does it have a way to use current transformers? Its it just a grid-connected AC battery, or does it provide UPS operation?
 
My SunXtender AGM cost $250/kWh capacity delivered.
I operate down to 70% DoD, and supposed to deliver 650 cycles to that depth.
That makes cost $0.55/kWh of power cycled. If used until wear-out, which I don't expect. Just backup so should die of old age instead and probably cost me > $2.50/kWh.

At this price per kWh it doesn't make any sense for peak shifting (except if there are demand charges for highest draw, which my rate tariff doesn't have.) I use it to keep a few things running at night during outages, and for grid-forming during the day so AC coupled PV can power loads. Maybe 20% of my power goes through the batteries, the rest is use-it-or-lose-it direct from PV.

If you can get cells for $80/kWh capacity, assuming 80% DoD cycling 3650 times over 10 years, that's $0.027/kWh of power cycled. That is similar to PV + GT inverters over 10 years (although those should last 20 years for lower power cost.) With BMS and battery inverter added, could still be competitive with grid.

Seems to me at $80/kWh the price is already low enough for application which cycle fully every day, e.g. peak shaving or offsetting part of your utility consumption. If battery bank has to be oversized to make up for days without sun, that costs extra (but might give more years of service.) Similar with PV; on-grid you capture 100% of production, while for off-grid you have to buy extra capacity to make up for times with less sunlight.

A few important observations when comparing to grid costs..
1) So far as I can tell after almost a year, that $80/kWh I bought the batteries for, was a one time thing. This is because 1) Covid has reduced electric car sales 2) Chip shortages 3) The battery cells I purchased (LG Chem N2.1 - 120aH) did not have nice screw terminals any monkey could hook up to.. Instead, it was a delicate aluminum tinfoil tab at one side and a delicate plated copper tinfoil tab at the other, and to further complicate the matter, since the cells had been salvaged, the tabs were welded and then cut, some as short as 3/8 of an inch... So because these things were so difficult to fabricate into a whole battery system, the demand for them was low, allowing me to get in at the low low price of $80/kWh. Then word got out about how to work with them and the supply dried up.
I kick myself over this.. I ordered 56 cells for myself and 14 for a friend, plus 2 spares.. Knowing what I know now, instead of 70, I should have purchased 300 because now I have a 48v golf cart as well..

2) The LG Chem N2.1, uses an NCM-622 cell chemistry. A full cycle (3.0 to 4.2 volts) will give the battery about 300 to 500 charges.. Pulling in the top and bottom ends to 3.4 to 4.05, should result in about 1500 to 2000 cycles before reaching that 80% mark. In other words, 80% DOD for 3650 cycles is way optimistic for a lithium ion NCM-622 chemistry.. and my total capacity after pulling in the ends results in a 20kWh pack.

In short, I am pretty obsessive with purchasing only the absolute best deals on the highest quality stuff I can find. I'm not going to spend $120+/kWh on a battery just because I feel I need one now.. I am patient when spending my money. Additionally, I am fully capable of building my own pack, even if it means having to machine special fixtures to make it happen (which I had to do with the BOLT EV cells).. And after all that, I think its safe to say that the math still doesn't beat grid pricing.... at least not at the 17c per kWh we are charged.

Nope, the off grid system provides peace of mind, but it is not suitable as our main power source if we look at economics.. I suspect that statement might not be true in another 5 to 10 years and I look forward to that time..
 
A few important observations when comparing to grid costs..
1) So far as I can tell after almost a year, that $80/kWh I bought the batteries for, was a one time thing. This is because 1) Covid has reduced electric car sales 2) Chip shortages 3) The battery cells I purchased (LG Chem N2.1 - 120aH) did not have nice screw terminals any monkey could hook up to.. Instead, it was a delicate aluminum tinfoil tab at one side and a delicate plated copper tinfoil tab at the other, and to further complicate the matter, since the cells had been salvaged, the tabs were welded and then cut, some as short as 3/8 of an inch... So because these things were so difficult to fabricate into a whole battery system, the demand for them was low, allowing me to get in at the low low price of $80/kWh. Then word got out about how to work with them and the supply dried up.
I kick myself over this.. I ordered 56 cells for myself and 14 for a friend, plus 2 spares.. Knowing what I know now, instead of 70, I should have purchased 300 because now I have a 48v golf cart as well..

2) The LG Chem N2.1, uses an NCM-622 cell chemistry. A full cycle (3.0 to 4.2 volts) will give the battery about 300 to 500 charges.. Pulling in the top and bottom ends to 3.4 to 4.05, should result in about 1500 to 2000 cycles before reaching that 80% mark. In other words, 80% DOD for 3650 cycles is way optimistic for a lithium ion NCM-622 chemistry.. and my total capacity after pulling in the ends results in a 20kWh pack.

In short, I am pretty obsessive with purchasing only the absolute best deals on the highest quality stuff I can find. I'm not going to spend $120+/kWh on a battery just because I feel I need one now.. I am patient when spending my money. Additionally, I am fully capable of building my own pack, even if it means having to machine special fixtures to make it happen (which I had to do with the BOLT EV cells).. And after all that, I think its safe to say that the math still doesn't beat grid pricing.... at least not at the 17c per kWh we are charged.

Nope, the off grid system provides peace of mind, but it is not suitable as our main power source if we look at economics.. I suspect that statement might not be true in another 5 to 10 years and I look forward to that time..
Its not the fact that the grid costs less. Its that we will have power when others dont when the grid goes down.
I will pay a higher price for that peace of mind. This is especially crucial for folks who require medical devices or refrigerated medicine to always be available.
My neighbor was dying of cancer during hurricane Sandy, and I was glad I was able to run an extention cord over to her house to keep her oxygen machine running for the entire 8 days we had no power, along with her fridge so she had food.
 
Its not the fact that the grid costs less. Its that we will have power when others dont when the grid goes down.
I will pay a higher price for that peace of mind. This is especially crucial for folks who require medical devices or refrigerated medicine to always be available.
My neighbor was dying of cancer during hurricane Sandy, and I was glad I was able to run an extention cord over to her house to keep her oxygen machine running for the entire 8 days we had no power, along with her fridge so she had food.
Yuuupp.. I'm a prepper.. and that's why I built the off grid part of our grid-tied solar system. It's not about the economics for me, its about being able to flip a couple switches and not having to listen to a noisy generator and wait in long lines to spend money on fuel for it.

When we first moved out here, we lost power 4 times in one year.. each time, it was 3 to 5 days with the longest being 8 days. I went through 12gal a day in gas at $3.90 a gallon at the time .. Almost $50 per day if we include the time and gas to go get the gas.

That got old really fast so I converted the generator in a tri-fuel unit for $200.. Now it can run on gas, propane, or natural gas. As it so happens, we have natural gas here, which is odd because everyone else is on propane. Running the generator for 5 days on natural gas only cost me about $5 -$7 per day.. much better, but I still had to change the oil every 24 hours.

Now we have the off-grid solar.. we still lose power three to five times a year but the time frames have dropped to 2 to 3 day outages... Much easier to just flip a couple switches and continue living as if nothing was wrong.. we don't even need to conserve power like we did with the generator.
And if the off-grid system ever fails, our SMA inverters can give us 2000 watts on each, so 4kW total while the sun is shining.
 
A couple images from the site overview I'm working on.

I'm sure I will be tweaking them. I need to work on the "one line drawing" next. Never done one, but I'm sure I can find some examples online. Once I finish that then I think I'm ready to submit the application to the utility company.

We discussed this earlier in the thread, but I put the arrays 3x the height away from each other. They will be a little less than 10ft tall, so I put them about 30ft apart so that they don't shade each other. I was originally going to do 3 different arrays, but this works out better, I think because it gives me more room from the property line and driveway. I also DO NOT want to risk them shading each other either. Peace of mind with this distance.
 

Attachments

  • solar_w_sat.PNG
    solar_w_sat.PNG
    1.2 MB · Views: 12
  • solar_no_sat_white.PNG
    solar_no_sat_white.PNG
    47.4 KB · Views: 12
Running the generator for 5 days on natural gas only cost me about $5 -$7 per day.. much better, but I still had to change the oil every 24 hours.
Why did the oil need to be changed so frequently? Most standby generators on natural gas go 200 hours between oil changes.
 
For you, it would depend on your battery inverter. Does it have a way to use current transformers? Its it just a grid-connected AC battery, or does it provide UPS operation?
I have a Fronius Primo. I think I would have to replace it with a hybrid inverter to be able to add batteries for peak shaving. My problem is very focused, I'm going on hourly pricing which is great for all our applications except in the late summer afternoons. Hourly pricing gets really high from about 5-7pm, that's the time when my solar is winding down but the air conditioning load is going up. So one thing I can do is cool the heck out of the house during the early afternoon then turn off the air until about 8pm when pricing drops back down. We also have a demand from livestock fans, they have to stay on in the late afternoon and they probably draw 3 or 4 kWh during those hot afternoons.
 
My peak rate is 4:00 to 9:00 PM. I try chilling the house extra in advance, but sun heats it quickly. If too hot I do run A/C, typically before 6:00 PM. I am a net producer at true-up, so consuming 3x credits for 1x power used during peak times occasionally isn't costing me anything.

Can't tell if Fronius Primo does frequency-watts for AC coupling to off-grid inverters. Doesn't even mention UL-1741 (SA or otherwise), so not for US market.


To avoid late afternoon consumption, an AC battery could be added without touching PV system. Grid would still supply motor starting loads; the battery would just charge during off-peak and discharge on-peak. There may be utility rules that it can only charge from PV (at least if battery might export to grid?). Commercial systems include Tesla PowerWall and SMA Sunny Boy Storage, which have expensive specific models of batteries.

Any battery inverter that can be persuaded to feed the grid might accomplish the task, feeding all loads.

A UPS type system (e.g. Sunny Island) could charge its battery while grid is connected, and would supply downstream loads from battery when grid is disconnected by means of a timer. But you have to calculate cost per kWh cycled from the battery to see if it saves money or not.


I envision chilling a tank of water off-peak and using that on one side of A/C system to cool house during peak times. But that's a different sort of engineering project. Some commercial systems here build ice at night and use antifreeze loop chiller for facility air during the day.
 
I need a little assistance on some parts.

From 1 array, I will have 3 sets of DC lines coming from the 3 different series. They will not be combined, but I will have to carry them underground to the other array where they will then be run back to the inverters.

Is there a specific type of "junction" box for this or will really anything work? The other thing is what is the best way to join the high voltage DC wires together inside said junction boxes?

Will I need any kind of DC breakers and if so, where?
 
I just built my own. Alibaba breakers/fuses and home depot electrical boxes.
 

Attachments

  • 20210324_095316.jpg
    20210324_095316.jpg
    127.2 KB · Views: 15
  • 20210324_095304.jpg
    20210324_095304.jpg
    232.5 KB · Views: 15
  • 20210324_095310.jpg
    20210324_095310.jpg
    175.7 KB · Views: 14
Back
Top