My DIY 60 kW 220 panel setup

[kundip]

Sounds great, look forward to it. If ur thinking about single pole mounts, would it take too much engineering to make them tracking types?, with maybe a couple of electric actuators or a heavy threaded type rod to make them automatically track the sun, or even a cable/chain arrangement that attaches to each side of the panel array & does a few loops around a motor spindle 45-75 degrees down the pole from panel array for tilting array, or a combination of both methods for true tracking. You would need a fairly beefy type of swivel attachment for full tracking, or perhaps just a decent bush or bearing arrangement for basic left to right tracking Like a see-saw type system. I dont know if you would consider the extra work & engineering worth the gain in ur situation, but it would make for a great review of ur systems capabilities. Look forward to ur updates. Cheers

My thoughts. Unless you have limited space:
If you want to put in tracking systems it is cheaper to put in more panels.
If you want to pay someone to wash the panels it is cheaper to put up more panels.
If you want to put in optimizers after the original install it is cheaper to put in more panels.
If you want to change to more efficient devices it is cheaper to put in more panels....I could go on.
I figure any fiddling with existing, working, systems will cause breakeages or damage to roof etc.
I use Micro Inverters but I put the Inverters in an accessible place where I can change out inverters on the ground.
They are Enphase & Sunny Boy ...so far none have failed.
This won the prize on hackaday 2023 https://hackaday.com/blog/?s=Solar+track Maybe you could make a hybrid of that.

 

[kundip]

RRe your micro i

I actually have an old dual axis STS tracker. It was used to track a 12' satellite dish from back when those were a thing. It is very stout and mounts on a 4" diameter pole. One day I'll see about mounting 6-8 panels to it and figured out how to control the motors in it to track the sun. Should be a fun project.

Since I'm limited by the capacity of the 50 KVA transformer to the grid, another thought I had, was that I could install panels on the hillside facing East and West, kind of like what this guy did:

View attachment 110718

View attachment 110719

Mounting would be real inexpensive using 2x4 treated posts and just doing 1 row in portrait. Uplift would be almost non existent from both sides. Also, I have this mountain of left over 250 watt micro inverters and connecting cables from the conversion to string inverters:

View attachment 110721

The DC specs for those inverters are as follows:

View attachment 110722

Given the max voltage input of 45V, I'm likely limited to 60 cell panels, although I think some 72 cell panels would probably be save.

Doing a quick each, it looks like I can pick up a pallet of 32 used SolarWorld 245W Mono panels for around $100 each. So for $3,200 plus a bit of lumber, I could add 7,800 watts of capacity, with 16 panels facing East and 16 facing West, giving me a nice boost in production in the AM and PM hours and not push my 50 VKA transformer.

Re your Micro Inverters. If you look at this video 4:20 secs in.
This is an alternative Micro Inverter setup that might interest you.

 

[crazydane]

@kundip That's a very interesting setup he has with those 4 inverters wired in parallel like that. So if I was to apply what he did, basically wire the solar panels and inverters in parallel, thus allowing the inverters to be located remotely from the panels. I guess the downside to that is voltage loss. 36 VDC will a much greater loss over, say 100ft, compared to 250 VAC.

Now if you could wire the panels in series and do the same with the micro inverters, that would eliminate the greater loss, but I'm 99% sure that would not end well since the MPPT circuit would likely get really confused and unless the 4 inverters that the exact same resistance all the time, some inverters would see much higher voltage than the others and probably cause some sort of cascading failure that would fry everything.

 

[kundip]

RRe your micro i

Re your Micro Inverters. If you look at this video 4:20 secs in. This is an alternative Micro Inverter setup that mightiinterest you.

It is always a hard calculation

@kundip That's a very interesting setup he has with those 4 inverters wired in parallel like that. So if I was to apply what he did, basically wire the solar panels and inverters in parallel, thus allowing the inverters to be located remotely from the panels. I guess the downside to that is voltage loss. 36 VDC will a much greater loss over, say 100ft, compared to 250 VAC.

Now if you could wire the panels in series and do the same with the micro inverters, that would eliminate the greater loss, but I'm 99% sure that would not end well since the MPPT circuit would likely get really confused and unless the 4 inverters that the exact same resistance all the time, some inverters would see much higher voltage than the others and probably cause some sort of cascading failure that would fry everything.

Over a certain voltage the MPPT switches off on all MPPT devices as I understand it.
The operating range is much higher & wider than the MPPT range.

 

[kundip]

It is always a hard calculation

Over a certain voltage the MPPT switches off on all MPPT devices as I understand it. The operating range is much higher than the MPPT range.

Re panels facing East & West.
My brother and niece live 1 km apart.
His roof faces north and she can only put panels east and west.
July's production for Jim's was 688KW and my nieces was 390KW.
Both systems are exactly the same spec and age - other than the way the panel's face.
The percentage difference is enormous.
June July are our worst months in Western Australia.
Both systems were installed this year.

 

[kundip]

Re panels facing East & West.
My brother and niece live 1 km apart. His roof faces north and she can only put panels east and west. July's production for Jim's was 688KW and my nieces was 390KW. Both systems are exactly the same spec and age - other than the way the panel's face. The percentage difference is enormous.
June July are our worst months. Both systems were installed this year.

One thing for certain there's nothing easy in solar.
It's really hard work, considering all the ins & outs, trying to save electricity costs.

 

[Evil1]

2kw PV on ur tracking mount would be a cool project, there is some good info to be found on YouTube for different tracking setups, with the space u have you could have a larger array on a rear pivot & rail or concrete track for the front guide wheels/rollers. Im sure there is a lot of simple & cheap electronics easily found nowadays to make sun tracking a simple task. Love ur obsession mate, keep it up. Cheers from Australia

 

[Evil1]

You made some good points about the micro inverter cons too, i had wondered how reliable they are given where they are generally placed, so it would make sense for you (or anyone using them) to use them on a ground mounted system or where you have better access for maintenance & repairs. Cheers

 

[stienman]

I have 3.6kw facing east/west (1.8kw each direction, out of a total 7kw system) and while there is a cost in overall efficiency, it does help spread the power out over the day so I have reduced reliance on the grid and storage (batteries). I suspect that many would benefit from such a setup given the cost of batteries and the elimination of net metering most are facing.

I'd suggest adding trees or bushes behind the array closest to the house for visual purposes. As nice as the arrays are, the view would be improved if they didn't take up so much of it.

I long ago decided not to have panels on my roof, but I'm on 4+ acres. I think you're in a similar situation. The pain and additional cost of roof solar just isn't worth it when a ground mount array is so cheap and easy. Plus, clearing them of snow is so much easier if you get the inclination to do so.

I recall seeing your mining setup on a youtuber's channel some time ago - can't recall if it was voskcoin or another, but I enjoyed seeing what you've accomplished, and am really happy to see the details behind the solar, which were largely glossed over in the mining video.

I'm scratching my head, though - with the frequency and length of your outages I'm surprised you haven't added battery storage before now. Given your mining consumption I can see that 30kwh of battery isn't going to last more than an hour or two in the dark, but aside from that your proposed EG 6500 setup doesn't appear to be fully operable in the sunlight - do you have provision to tell the non EG6500 inverters to do zero export so they can continue to work when the power is out, with the EG6500s providing the micro grid, or do you have some other method to make the full array work in power out situations?

 

[Evil1]

My thoughts. Unless up oubhave limited space. If you want to put in tracking systems it is cheaper to put in more panels. If you want to pay someone to wash the panels it is cheaper to put up more panels. If you want to put in optimizers it is cheaper to put in more panels. I could go on.
I figure any fiddling with existing, working, systems will cause breakeages or damage to roof etc.
I use Micro Inverters but I put the Inverters in an accessible place where I can change out inverters on the ground. They are Enphase & Sunny Boy ...so far none have failed.
This won the prize on hackaday 2023 https://hackaday.com/blog/?s=Solar+track Maybe you could make a hybrid of that. Snapshot of some old Sunny Boys I am adding to my system attached.

@kundip That's a very interesting setup he has with those 4 inverters wired in parallel like that. So if I was to apply what he did, basically wire the solar panels and inverters in parallel, thus allowing the inverters to be located remotely from the panels. I guess the downside to that is voltage loss. 36 VDC will a much greater loss over, say 100ft, compared to 250 VAC.

Now if you could wire the panels in series and do the same with the micro inverters, that would eliminate the greater loss, but I'm 99% sure that would not end well since the MPPT circuit would likely get really confused and unless the 4 inverters that the exact same resistance all the time, some inverters would see much higher voltage than the others and probably cause some sort of cascading failure that would fry everything.

Micro inverters on each panel are only really good where there is a higher incidence of shading aren’t they? For ur large arrays wouldn’t u be better to have 1 or 2 large HV MPPT inverters in series for maximum allowable VOC at each array site and send the AC back to main panel, DC has much greater losses over distance than AC. 3 phase inverters would have to be a good consideration also for these larger arrays, more efficient, higher volage, thinner wire runs, more power Production.

 

[Evil1]

Re panels facing East & West.
My brother and niece live 1 km apart. His roof faces north and she can only put panels east and west. July's production for Jim's was 688KW and my nieces was 390KW. Both systems are exactly the same spec and age - other than the way the panel's face. The percentage difference is enormous.
June July are our worst months. Both systems were installed this year.

Wow, that is a big difference

 

[crazydane]

Re panels facing East & West.
My brother and niece live 1 km apart. His roof faces north and she can only put panels east and west. July's production for Jim's was 688KW and my nieces was 390KW. Both systems are exactly the same spec and age - other than the way the panel's face. The percentage difference is enormous.
June July are our worst months. Both systems were installed this year.

I find that very surprising. I can see how North facing panels do ok in the summer months, but would be terrible in the winter month where they may not even get any direct sunlight, depending on the pitch of the roof they are on. But I would have thought East and West panels would produce way more than North facing panels, when looking at a whole year's worth of production.

I know for my South facing panels, I see a significant change based on season:





I'd be very curious to see what similar graphs looks like for someone with panels facing only East and West.

 

[crazydane]

Micro inverters on each panel are only really good where there is a higher incidence of shading aren’t they? For ur large arrays wouldn’t u be better to have 1 or 2 large HV MPPT inverters in series for maximum allowable VOC at each array site and send the AC back to main panel, DC has much greater losses over distance than AC. 3 phase inverters would have to be a good consideration also for these larger arrays, more efficient, higher volage, thinner wire runs, more power Production.

I wish I had 3 phase power on my street, but my little side road is just 14.4K single phase on the pole. I would have loved to have been able to just buy a single honking 60 kW inverter! I did not know that DC had a greater loss over distance than AC, for a given gauge wire.

The only reason I'm looking at East and West facing panels if I add more, is that the peak would occur earlier and later in the day compared to all my south facing arrays. My transformer on the street is only rated for 50 KVA, and I would have to melt it. If I took all my miners offline, I would be pushing 55 KVA through it (assuming a power factor of 1). I'm technically not supposed to have more than 20 kW in a residential system where I live, so I don't want to rock the boat.

Another issue with having 55 kW of power coming in all at the same time, is that I see quite a bit of voltage rise around noon, and my POCO likes to run it hot to begin with, around 246:



Zooming on on 9/1:



I had to get a grid guard code for my SMA inverters are raise the "floor" from 120 to 128 to keep them from cutting out at high noon. I don't know that I would want to go much past 260 VAC.

 

[Shimmy]

Awesome system! Did you use Ubiquiti equipment before? Given how they stick with some things I was really surprised they started SunMax. Looks like they are as reliable as my PTZ Protect camera.

 

[crazydane]

Yeah, I started out with 80 SunMax micro inverters, then talked Ubiquiti into sending me 56 spares for free to have on hand given my high failure rates. I ended using 48 of them on 2 new 24 panel arrays, leaving me 8 spares to swap out as needed and RMA'ing those as they were died.

I finally got fed up with having to swap out failed inverters and got rid of all of them last month and went 100% string inverters.

In their defense, I had issues with the inverters cutting out back in the fall of 2016 when I initially deployed them and worked directly with their design/engineering team overseas to tweak the firmware for the US market. They send me about $20k worth of test equipment to capture the waveforms coming off the grid to troubleshoot the issues I was experiencing. After about 3 iterations of firmware, the cutting out issues was all but eliminated.

The main flaw with their inverters was that during the manufacturing process, some of the surface caps were under too much tension during the ware soldering step, and caused them to separate, resulting in this:



So back on the topic of adding some East and West facing panels. I was thinking about the voltage loss of AC vs. DC, and I found this that seems to favor DC or AC for transmission:

1. Skin effect experienced with AC. There is no skin effect with DC.
   
   
   
2. Higher voltage allowed with DC for the same transmission lines. The lines have to withstand the peak voltage. With AC, that is 1.4 times higher than the RMS. With DC, the RMS and peak voltages are the same. However, the power transmitted is the current times the RMS, not peak, voltage.
   
   
   
3. No radiation loss with DC. Long transmission lines act as antennas and do radiate some power. That can only happen with AC.
   
   
   
4. No induction losses. The changing magnetic field around a wire carrying AC current causes induced voltage and current in nearby conductors. In effect, the transmission line is the primary of a transformer, and conductors near it are secondaries. With DC current, the magnetic field doesn't change and therefore doesn't transfer power.

And then I thought about the fact that I'm only using 2 of the 3 MPPT inputs on all but 1 of my SMA inverters, So I have 7 unused MPPT inputs that I could use for East and West facing panels. Here's my array layout and MPPT inputs used:



The Sunny Boy 7.0's on the shop roof are barely "oversubscribed" with a max of 7.15 kW currently. Ditto on ground array #1 with 6.13 kW on a Sunny Boy 6.0, so those 3 MPPT spare inputs would probably be the first ones I'd use if I went that route.

In hindsight, I guess I should have just gotten all 7.7s, lol.

 

[crazydane]

I recall seeing your mining setup on a youtuber's channel some time ago - can't recall if it was voskcoin or another, but I enjoyed seeing what you've accomplished, and am really happy to see the details behind the solar, which were largely glossed over in the mining video.

I'm scratching my head, though - with the frequency and length of your outages I'm surprised you haven't added battery storage before now. Given your mining consumption I can see that 30kwh of battery isn't going to last more than an hour or two in the dark, but aside from that your proposed EG 6500 setup doesn't appear to be fully operable in the sunlight - do you have provision to tell the non EG6500 inverters to do zero export so they can continue to work when the power is out, with the EG6500s providing the micro grid, or do you have some other method to make the full array work in power out situations?

Yep, voskcoin came out a couple of times shooting videos for his channel. The 2nd one went into a bit more detail about the solar.

AFAIK, my 8 Sunny Boy inverters can't operate at 100% without grid power. They do all have a 2000 watt 120V emergency power output however, that do function during a grid failure, but since the inverters are all located in the my shop building 200' away from the house, I never bothered connecting anything to them or researched if it is possible to connect the emergency outputs from several Sunny Boys together for 240V and or higher wattage.

I'm not sure why the 6500EGs wouldn't be fully operationable in the sunlight. The AC input would be connected to the load side of the 400A transfer switch and the AC output would be connected to the dual 200A sub-panels for all my house circuits. There would then be solar panels connected directly to the MPPT inputs of the 6500s and then of course the battery bank would be connected as well.

So during the day, the house would run on batteries that would be charged by the sun and at night the house would continue to run on batteries that would maintain charge from the grid power coming in via the AC input.

So this way, the 100 kWh consumed daily by the house, would be mostly supplied by the batteries / solar panels connected to the 6500s, and only supplemented by the grid when needed.

As I mentioned previously, during the winter I bring the miners up into the house to provide the heat I need, so I'm thinking that when/if I install the 6500s, I would add another sub-panel at the house, that should be feed directly from the grid and not from the load side of the 6500s. That way I can balance the miner load at the house, and only only connect the number of miners to the load side of the 6500s, that they can reasonably handle. The rest would be connected to the sub-panel connected directly to the grid.

Additionally, during an extended power outage, my diesel generator can kick in when the batteries start getting load, and that would all be automatic since the transfer switch is automatic. If the 6500 doesn't support a generator contact closure when the batteries get below a certain threshold, I'm sure I can use Solar Assistant or Home Assistant to do it for me.

 

[crazydane]

So here's a copy of my latest electric bill:



Imagine what it would have been without any solar! September of last year is when I started expanding my mining operation again and the days of double digit monthly kWh bills was over.

So my total kWh from Sep 2021 through Aug 2022 was 37,877 kWh, for a monthly average of 3,156 kWh.

So I got a good bit to go before I'll be able to eliminate my power bill completely. I added mining rigs during the fall of last year and into spring this year. But I also added 15 more panels and converted to all inverters, so that might be close to a wash as far as the consumption vs. production ratio.

Got an email into santansolar to see how much a pallet or two of his used Jinko 310W 72 cell poly panels would cost to shop from GA to VA. He has them listed for $96.00 each, which is $0.31 per watt, which seems like a decent deal. If he can put 32 on a pallet and I got 2 pallets, that would be just shy of 20 kW for a bit over $6k. I think that would get me close to closing the gap on my electric bill.

 

[stienman]

It's the hotdog/bun problem. When you have more solar, you're going to get more rigs, and then start all over again. ?

 

[Hedges]

I don't think EG inverters support AC coupling.
How about using a battery inverter that does (preferably Sunny Island), so it can interact with your GT PV Sunny Boys?
For a split-phase system that backfeeds grid, the max system is 4x SI with 24kW of SB. Once disconnected from grid, more SB could be switched onto it for total of 48kW.

 

[Browse]

As for the wife and the views from the house, you can plant some 6 FT Leyland Cypress trees to grow into a privacy wall over the next 5 years. Plant a line of them just before the first ground mount array from the house. The Leyland I planted about 5 years ago is already 25 ft tall and about 10 ft wide.