• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Let's get my system built this weekend, on paper!!

skyking1

Solar Enthusiast
Joined
Mar 28, 2024
Messages
219
Location
Washington State
Here is our playground:
PXL_20250115_004433544.jpg

The house is skewed around to 200 degrees or so, which is ideal for morning clouds/afternoon sun which is so common in any part of Western Washington.
The roofs are assymetrical, with the southern facing roof being 264" (22') long on both the upper and the West Wing roofs. This was purpose built for solar.
Upper roof is not obstructed by trees above the horizon created by the hill to the south. I do not see any shading happening that will darken a string up there, no matter how it is arranged.
The roof pitch is not ideal for solar, at 3:12. It is good for my personal survival. I will not tilt and the roof is therefore at 14 degrees.
It will have standing seam metal on it starting Thursday.
My plan is to use S5! clips and put rails to that. I hear iron mountain talked about, is that the ticket?
I have Oatey roof jacks for conduit. I plan on two at the far (east ) end of that upper roof, and make an 18" clear path top and bottom and a 36" clear path up the near (west) end for NFPA regs.
Available space after removing those paths is 228" tall by 396" wide.
This will accommodate 3 rows of shorter panels in portrait. The big 80" panels would go to landscape.
Our goal is to fill that space first.
It will be somewhere near 8K of panels in two strings.
The West wing roof near us is the stepchild roof. It gets morning shade from the 2nd story.
I want panels on that to a target of 11~12K total.
That roof is 228" tall by 180" wide after removing the 36" path from along that wall, the logical place due to shade.
We can leave more than that out and just use the bottom edge and then up the west edge to minimize that shade effect.
 
I am leaning hard at the Midnite Solar AIO 15K.
It has 3 MPPT's. Two for that upper roof and one that will track the stepchild with shade roof.


So let's talk panels!!

This thread, for an example.
EBAY panels

Those could phsically go on 3x8 in portrait. Two strings of 12 each. 5040 W and 456 open circuit volts per string. This seems to be in the wheelhouse of that AIO.
The west wing could do 3x4 in portrait for a little too many watts for the AIO, so we could drop a panel from each string and get to 13860 watts.
it would bring the VOC down to 418.
 
Last edited:
I personally don’t think bi facial mounted on a roof is much of a benefit typically they need at least 3 ft behind them with a light reflection surrounding also the panels are heavier than standard panels
 
  • Like
Reactions: Zwy
Come on Lonestar! Help me buy racking and a pallet of panels!
rickmoranisdarkhelmet-645x370.jpg


Perhaps you need more details.
OK.
The grid is coming to my house sometime in 2025. I want to get this up and running in advance of that so we can enjoy a little lighting etc.
The home gets DC lighting right off that battery, so that does not need backed up.
The output of 10Kw is plenty for critical loads going forward. This AIW checks my boxes.
I need to get 6 energy credits for a new home build in Washington, and this PV installation gets the maximum of 3 credits easily.
I was going with 1.5 credits from my Ground Source Heat Pump, but have run afoul in some details there so I am pivoting to PV now rather than later.
 
Last edited:
OK, looks like S5-N Clamps and IronRidge XR10 rails.
After some study I see that cheap bifacials are least expensive per watt, totally ignoring the potential for backside power. Is there any reason not to just run them on the primary side only?
 
looks like S5-N Clamps and IronRidge XR10 rails.
What's your wind zone ?

After some study I see that cheap bifacials are least expensive per watt, totally ignoring the potential for backside power. Is there any reason not to just run them on the primary side only?
Buy the cheapest per front watt, but make sure they are rated strong enough for your climate.
 
What's your wind zone ?


Buy the cheapest per front watt, but make sure they are rated strong enough for your climate.
110 MPH wind exposure C snow load is the basic 25 PSF
Roof is built to standard 15 PSF dead 25 PSF 110% snow load.
EDIT:
Looking at the XR10 rail, it is rated for 4' OC at 30 PSF and 160 MPH. The roof panels are 16" so that works fine.
(54) S5-N clips on that upper roof on 6 rails.
 
Last edited:
Never dealt with them personally, but they've only been on eBay since May. Also, 99.5% and above is pretty good feedback. It shows the seller cares about customers and tries to fix problems to an extent. They're waaaay below that. Granted, only 1 neg with so little feedback, but its a gamble.

Similar CS panels pop up on eBay from time to time (I know because we bought a couple truckloads in 2022/2023 and sold on eBay). When we had them, $3500-4000/pallet for 30x 400w was about the going rate. I picked up my pallets of 540w CS for my own system on eBay as well for ~$5500/pallet of 30. Be patient, and buy from sellers with a solid history. An expensive freight purchase is the worst kind to have an issue with on eBay
 
I went to those sites and made some observervations.
At SanTan, they have quite a few used and blemished SSG and Yingli panels. Returns with some sort of issue and significant power degradation.
I was interested in Yingli, but that does not bode well.
What brands/models are better quality?
I'm willing to pay more to do it once.
 
the playground is getting real-er!
PXL_20250131_235719775.jpg

It just needs barge and roof cap, and done. It has two of these jacks in place for the solar.

PXL_20250131_202344929.jpg


I have one left to put on the lower roof. I figured each string gets a conduit.
Cuts down on longer runs.
 
Those could phsically go on 3x8 in portrait. Two strings of 12 each. 5040 W and 456 open circuit volts per string. This seems to be in the wheelhouse of that AIO.
The west wing could do 3x4 in portrait for a little too many watts for the AIO, so we could drop a panel from each string and get to 13860 watts.
it would bring the VOC down to 418.
You'll very rarely achieve full nameplate power, usually you'd be looking at maybe 70% of what you'd think could achieve.

Main thing is never exceed max voltage. The current is drawn by the mppt, so you can't overload them with panels (within reason).

Therefore overpanelling, or connecting more nominal PV watts can help actually deliver closer to your aios full charge capability. An overpaneling ratio of 1.2-1.3 is a frequent rule of thumb.
 
This brings me back to the larger question of putting too many panels on the AIO and not being able to sell it back to the grid. I'd want to take advantage of the summertime peaks and spend that meter as hard as I can backwards.
 
This brings me back to the larger question of putting too many panels on the AIO and not being able to sell it back to the grid. I'd want to take advantage of the summertime peaks and spend that meter as hard as I can backwards.
The conditions where you would achieve 100% nameplate is prob less than 1% of time on average. Whereas having additional panels will produce additional output for the rest of the year, where you won't be hitting peak nameplate.
 
The conditions where you would achieve 100% nameplate is prob less than 1% of time on average. Whereas having additional panels will produce additional output for the rest of the year, where you won't be hitting peak nameplate.
the unit is rated at 11,400 Watts.
So they have a maximum of 15Kw. Is that what you are talking about 1.2~1.3 , the ratio between 11.4 and 15, or are you suggesting 1.3x the 15KW max on the inverter?
My understanding is respect the maximum voltage by multiplying panels x VOC.
 
the unit is rated at 11,400 Watts.
So they have a maximum of 15Kw. Is that what you are talking about 1.2~1.3 , the ratio between 11.4 and 15, or are you suggesting 1.3x the 15KW max on the inverter?
My understanding is respect the maximum voltage by multiplying panels x VOC.
No, that with a max of 15k of PV charging, you could put say 20kW of panels on there to get your average production up. You were hesitant to exceed by a few hundred watts so wanted to clarify that. Yes, just the max voltage to be wary off, taking into account record cold temps.
 
Thanks I understand that issue. If I use VOC to 550 volts on a string, is that a sweet spot to get best efficency and still leave a bit of headroom?
I noticed that 550V is the upper limit of MPPT operating voltage range.
This is pretty exciting. Essentially I can cover my south facing roofs and send it all to the AIO, within reason.
The upper roof will have consistent sun so it can go on two MPPTs as I was thinking, and the lower roof with shade issues can go on another.
I am now grasping that panels with lower VOC are the way to go.
 
Thanks I understand that issue. If I use VOC to 550 volts on a string, is that a sweet spot to get best efficency and still leave a bit of headroom?
I noticed that 550V is the upper limit of MPPT operating voltage range.
This is pretty exciting. Essentially I can cover my south facing roofs and send it all to the AIO, within reason.
The upper roof will have consistent sun so it can go on two MPPTs as I was thinking, and the lower roof with shade issues can go on another.
I am now grasping that panels with lower VOC are the way to go.
I would target 500V per string to give you cold weather leeway. 30A/22A/22A limits on the trackers, so you could do parallel strings if needed on tracker 1.
 
I would target 500V per string to give you cold weather leeway. 30A/22A/22A limits on the trackers, so you could do parallel strings if needed on tracker 1.
OK so now we look for sub-15 Amps too.
It opens up possibilities.
I may have to do ala carte rather than pallets of panels to use my spaces to the maximum effect.
 
I would target 500V per string to give you cold weather leeway. 30A/22A/22A limits on the trackers, so you could do parallel strings if needed on tracker 1.
I had read that keeping the voltage similar is best for efficiency.
Using Silfab SIL 410 HC ( 11.15 Isc 45.59Voc )
top roof a 3x9 in portrait
MPPT1
2 strings of 9
MPPT2
1 string of 9
total Voc 410V
Vpmax of 350V
Go to lower roof
MPPT3
I can physically put 12 down there. Respecting the 500V suggestion, that limits it to 11.
total Voc 501V
Vpmax 429V

So that is a total of 38 panels with a nameplate output of 15.6 Kw

Would it be best to just drop two panels off the lower roof, and have all the voltages the same?
It is not much difference in nameplate, and it might pay off in system performance.

I feel like I am closing in on a decision here, which is great.
 
I had read that keeping the voltage similar is best for efficiency.
Using Silfab SIL 410 HC ( 11.15 Isc 45.59Voc )
top roof a 3x9 in portrait
MPPT1
2 strings of 9
MPPT2
1 string of 9
total Voc 410V
Vpmax of 350V
Go to lower roof
MPPT3
I can physically put 12 down there. Respecting the 500V suggestion, that limits it to 11.
total Voc 501V
Vpmax 429V

So that is a total of 38 panels with a nameplate output of 15.6 Kw

Would it be best to just drop two panels off the lower roof, and have all the voltages the same?
It is not much difference in nameplate, and it might pay off in system performance.

I feel like I am closing in on a decision here, which is great.
Each mppt tracker is independent. The voltage of one doesn't affect the others.
 
Nice place but i don't see any heavy equipment to go along with that huge pile of topsoil? (i recognize your avatar)

The San tan site is sort of revealing about what kind of issues one might come up against with panel defects. green busbars, snail tracks, browning etc.
 

diy solar

diy solar
Back
Top