diy solar

diy solar

Funny story about the east-west panel spread-out idea

Summer Solstice surprised the shit outta me as the total for the day is a little higher:

View attachment 150326
Sun is never very far south in the longer Summer days, so the non-S "penalty" the E-W array would suffer isn't nearly so big as shorter days when the sun is pretty far to the south at high noon. Similarly, the sun spends a pretty significant amount of time in the west and east, since it rises in the ENE, and sets in WNW, so it "swings through" both due east and due west. Contrast to the winter months, it rises in ESE and sets in WSW, sot it is never fully due east or west. If you don't know this site, it is worth a bookmark: https://www.suncalc.org. You can move the cursor to your property to dial it in more precisely, and it will show the sun's arc over the sky.
 
I have east west, both pointed 60 degrees above the horizon, and it works wonderfully well. It starts up at least an hour before my main (and much larger) array even wakes up.
In late afternoon it keeps on going after my main array is fast asleep on the job.

A lot depends on your latitude, and time of year, but here, the sun is actually BEHIND the main array at sunrise and sunset for several months.

I only have a very small battery right now, and power right immediately at sunrise, is power that means I am getting off night time discharge sooner, and into charging at least an hour sooner.
Likewise at sunset I am still running on solar for about at least an extra hour later. So hours of darkness (and battery discharge) are much less than they would otherwise be. That will all change when I get my much larger battery of course. I will then be able to store massive peaks of energy, and coast through gloomy days without a worry in the world. But right now that is not the case.

Another advantage, not often appreciated is that in total cloud cover, those evil grey days when you cannot actually see where the sun is, it makes very little difference which way the panels face, the light is totally diffused and very weak anyway.
The more panels you have out there, the more you get, whichever way they face, but it will still be pitifully low power.

Armchair simulations measuring total average daily power, do not tell the whole story. There are some situations where a bit extra at the most useful times can have much more value than a lot extra when the battery is full, and you cannot really use it.

I have had this east west system successfully running for almost six years, and can thoroughly recommend this, ESPECIALLY if you are battery capacity challenged...

This is what it looked like in 2018, six 200w panels due east, six panels due west
.
It works so well I have since extended it to eight panels east, and eight west. Could not be happier with it. The frame is an equilateral triangle on a flat garage roof, very strong, the wind cannot get underneath the panels, and it has survived many severe storms without any damage at all.

This may not be something everyone needs, or can take advantage of. But a few extra east west panels can really help a marginally sized battery installation.
 

Attachments

  • 2018-01-24_0001.jpg
    2018-01-24_0001.jpg
    835.9 KB · Views: 10
Last edited:
I think these things will be serviceable for 20+ years if treated right. There are some folks in here that have been running them what? 11-13 years and report 95%+ capacities. I’m SWAGing that a bit as I can’t be arsed to go a trip through forum search results.
I first installed 12 100 watt panels in 2000, I "retired" them in 2019 because of higher density panels were available.
I installed 6 310 watt panels in their stead

Those 12 panels were in my bone pile in the backyard until yesterday when I sold them to a friend for $10 each
we tested them when he picked them up and were generating some 60+ watts
siemens panels - still producing for the next guy
 
My response should not be interpreted as being anti E/W placement. It is intended to demonstrate the claims made in the OP are at best inaccurate and at worst very misleading. while summer performance may yield comparable total daily production with more favorable morning and evening power delivery, the overall performance, especially in colder months, is notably worse.

When all is said and done, I have about 10kW of SanTan panels in addition to my main S 6kW array (3kW mounted). I fully intend to deploy a portion of the SanTan panels in both E and W arrays to overpanel my main S array for more favorable morning/evening production.
 
1.4 cents per kW production per day.
The calculations up to that point don't include time, so there's no assuming 1 or 5 hours.

$0.50 per watt original panel cost, divided by 7,000 days is a cost of 7.1 cents per day per kilowatt of production capacity. You can then calculate hours of full sun and estimate a cost per kwh, but I didn't include time until later in the post where I assumed a 4 hour production day vs 8 hour production day.

I'd be very interested in a source of solar panels and mounting hardware that together cost $0.025 per kw per day. $0.175 per watt original cost for panels and mounting hardware is exceptional for 20 year panels. My only solar panel buy years ago was a pallet with about 7.5kw at $0.40/watt including shipping, not including mounting hardware, and that was new old stock 13% efficiency.
 
@sunshine_eggo Do you feel like doing any of your spreadsheet magic for South vs. All West? I have ran the pv numbers for norcal, but I'm a bit limited to my gut feelings about what I'm looking at, rather than better analysis.

I know that All West is much better suited to our summertime A/C load. But the winter losses may be worse than I'm imagining.
 
I'd be very interested in a source of solar panels and mounting hardware that together cost $0.025 per kw per day. $0.175 per watt original cost for panels and mounting hardware is exceptional for 20 year panels.

I meant per kWh, amortized over a presumed life. "kw per day" is not a metric I've ever heard before. Wattage and days presumed lifespan.

$1.00/W of PV panel, mounting hardware, grid-tie inverter works out to $0.025/kWh over 20 years.
If RSD required that raises price a bit.
 
Here's my system. The green (Inverter 2) is the East-West array, Blue (Inverter 1) is South facing. I did my East-West array because that's how my shop roof is oriented. With the frequent cloud cover in the inland PNW, sometimes it's advantageous to catch some sunshine early before clouds roll in and/or the sun climbs above the clouds.
1685041688060.png
(the chart is from early May)
 
@sunshine_eggo Do you feel like doing any of your spreadsheet magic for South vs. All West? I have ran the pv numbers for norcal, but I'm a bit limited to my gut feelings about what I'm looking at, rather than better analysis.

I know that All West is much better suited to our summertime A/C load. But the winter losses may be worse than I'm imagining.

If you generate the hourly csv files, zip them and PM them to me, I'll see what I can do.
 
I have used small systems like this to power my off grid shop for several years.

The reason that I can get away with it is that I have some panels on the racks that allow pointing them at the sun early in the morning and late in the afternoon. When I had the minivan, I had 300 watts on top of it, now I just have 150 watts on the explorer.

In the photo it is just capturing the late afternoon sun.

The ability to capture the sunlight all day long is critical to how it all works with so little battery storage.

It has battery SOC monitoring but I just glance at it - don't capture it for trending.

Very roughly what I see is that if I were using a fixed configuration based on how this setup works, I would face:
~ 25% toward the early morning sun
~ 25% facing roughly up
~ 50% facing the very late in the day sun

Yes I know it is very simplistic but it works.
 

Attachments

  • vertical panel late in the day crop compr.jpg
    vertical panel late in the day crop compr.jpg
    197.5 KB · Views: 17
Bingo !
I think this cuts right to the point.
Not everyone has this undersized battery problem, but for a small system, its something well worth experimenting with.

Yes - it is only 2 kW-hrs (nameplate ) of storage so I can run it for a while after the sun goes down but obviously there are limitations.
 
Bingo !
I think this cuts right to the point.
Not everyone has this undersized battery problem, but for a small system, its something well worth experimenting with.
Might even be worth experimenting with if you have a bigger system with AC coupling like SMA if you are trying to make sure you have possible amps available as much as possible (goes back to the post I started a couple of days ago referencing the ability of Sunny Islands and max amp capacity).

I may break down a put a a string on my shop’s east roof and a couple of strings on the west roof though I was trying to avoid roof mounts for a plethora of reasons.* But I do have a spare Sunny Boy 7.7 that could be put to use. At the very least, I’d boost available energy for battery storage in the winter.

*servicing a pain the a$$
*cleaning a pain in the a$$
*tempermental rapid shutdown if you want to comply with current NEC (I don’t have to, but still)
*metal roof, lots of wildlife, and high voltage DC = high voltage welder if things go wrong
*how long are sealants going to last? Turning a 50-100 year roof into a 20 year roof is not a good idea. Especially when there is spray foam involved.
 
Last edited:
There are metal roof mounts which do not require penetration. Depends on the type of roof.
Mine is ag. I’d been another year building the shop if I tried finding a standing seam guy in these crazy a$$ed times.

Closest I can come are the ones that screw to the side of the ribs rather than the valleys and that has its own concerns (not tied into the purlins).

I have house wrap between the foam and the metal which has pluses and minuses. The minus in this case is that the roof structure is not glued together as well as could be (the plus is that repairs and alterations are a hell of a lot easier).

I probably also overthink things.
 
Last edited:
Here's some real data from my system:
I have only had it since early October last year, so hasn't had a full year cycle yet. But I still have some reasonable data.

I have tried to capture data from some perfect blue bird days throughout the last 8 months. This wasn't easy over the summer as I have an export limit of 5kW, so the other 10kW that are being generated during the middle of the day have to be used, and only a small amount could go into my battery.

Location: About 37.7 degrees South in New Zealand
Roof pitch: 15degrees
Arrays: 6 in total, all equal size 2460W, 2 facing ENE, 2 facing WSW, and 2 facing NNE
Total production of ENE Arrays: 4251kWh
Total production of NNE Arrays: 4365kWh
Total production of WSW Arrays: 3825kWh

Edit: Ignore the cables lying on the roof etc, this photo was taken during installation.
solar_panels (Medium) (1).jpg

Mid-spring. You can see that one of the East facing strings gets going a bit earlier (the other one was shaded). But in the evening the North and west facing arrays are pretty much the same.
1685072869268.png

Mid-summer: This one had some cloud unfortunately, but there was very little difference between the arrays in the morning and evening. The North facing array just has a bit more total production.
1685072897392.png

Early Autumn - Like in spring, the east facing arrays definitely get going quite a bit earlier in the morning. The slightly south facing 'West' arrays have noticeably lower production in the middle of the day.
1685072916885.png


Same as above.
1685072937137.png
Late Autumn: still very similar, I don't have a good explanation for the strange kink in the late afternoon.
1685072953844.png


1685072995460.png
 

Attachments

  • 1685072967838.png
    1685072967838.png
    41.8 KB · Views: 2
  • Solar Strings Production Curves Over Time from Grafana.zip
    382.5 KB · Views: 1
It's not instead of south facing.
It's in addition to south facing.
Panels are the cheapest part of solar.
Yep!
The best Plan in my mind for people who already have their inverters at or near capacity with South facing panels is to just add a separate low cost SCC with two MPPT inputs and load each MPPT with East and then West panels. Your batteries are getting charged earlier in the morning plus that precious evening charge.
 
Last edited:
Might even be worth experimenting with if you have a bigger system with AC coupling like SMA if you are trying to make sure you have possible amps available as much as possible (goes back to the post I started a couple of days ago referencing the ability of Sunny Islands and max amp capacity).

I may break down a put a a string on my shop’s east roof and a couple of strings on the west roof though I was trying to avoid roof mounts for a plethora of reasons.* But I do have a spare Sunny Boy 7.7 that could be put to use. At the very least, I’d boost available energy for battery storage in the winter.

*servicing a pain the a$$
*cleaning a pain in the a$$
*tempermental rapid shutdown if you want to comply with current NEC (I don’t have to, but still)
*metal roof, lots of wildlife, and high voltage DC = high voltage welder if things go wrong
*how long are sealants going to last? Turning a 50-100 year roof into a 20 year roof is not a good idea. Especially when there is spray foam involved.

It isn't the roof that you should be focusing on. Mount them on the west facing vertical wall so they are getting sunshine from 16:00 - to - as late as possible.
 
It isn't the roof that you should be focusing on. Mount them on the west facing vertical wall so they are getting sunshine from 16:00 - to - as late as possible.
Can’t. Mature trees would cast shade there in the mornings and evenings. I’m not taking out the trees (already took out a bunch). The top two thirds of the roof remains unshaded though.
 
Back
Top