Overlapping PV panels to shed rain

[bob prohaska]

[Not to be confused with "shingled solar panels"]

It looks relatively straightforward to mount solar panels so they
can shed rain when pitched for lattitude. This doesn't seem to be
common practice and I'm wondering why. Some additional flashing
will be needed to finish the edges, but it does not look extensive.
Overlapping the panels by the width of the frame will let them shed
water with no loss in exposure, use of bifacial panels would add some
output if the area beneath is relatively open. The panels will be
canted slightly relative to the mounting plane, so spacers to fill the
gaps will be desirable if not required.

The missing piece of the puzzle is hardware to mount the panels
with a frame-width overlap. All that I've found is for side-by-side
mounting.

The prospective motivation is a shade structure over my back patio.
A big shade tree has been declining for years, when it finally has
to go something to replace it will be urgently wanted. A PV patio
cover might be an entertaining, if expensive, experiment. Water
shedding would make the space underneath much more useful.

Thanks for reading, and any thoughts.

bob prohaska

[Jim Wilkins]

"bob prohaska" wrote in message news:udnokn$15rch$1...@dont-email.me...

-------------------
If you find the custom mounting brackets difficult to fabricate, what would
you do for the vertical seams between them, and the ridge? Sheet aluminum
mounting brackets are short enough to bend with hand tools but the others
would require a long bending brake, such as a Van Mark or the Tapco like I
used to cover my window frames. You may need some clearance between the rows
to install and tighten the fasteners since one long strip of connected
panels would be too clumsy and fragile to install as a unit.

This is a press-in thread insert to put a strong threaded blind hole in
aluminum, like the solar panel frames, where you can't reach behind with a
nut.
https://www.amazon.com/uxcell-Clinching-0-11-Inch-Stainless-Mounting/dp/B08XYLF4RH/ref=sr_1_5?c=ts&keywords=Self-Clinching%2BNuts&qid=1654275593&s=industrial&sr=1-5&ts_id=317399011&th=1

Unfortunately they don't give the max metal thickness and the shorter ones
for thinner metal may push out unless firmly staked, for which I made a
portable tool on the lathe.

Each of individual solar cell produces the panel's full rated current at
0.6V, so a small amount of shading from the edge of the panel above can
cause a large reduction in output. If you angle the roof half way between
the winter and summer angles you may lose power to shading in mid summer,
and early and late in the day from the vertical cover strips, or have too
little overlap to prevent splash from passing through the gap.
https://gml.noaa.gov/grad/solcalc/azel.html

For an outdoor roof that doesn't have to be as absolutely storm-proof as a
house you might put up corrugated panels with gaps between them to attach
the panel brackets from below. At the panel overlaps the lower panel could
connect to the roof purlin with a flat aluminum bracket and the upper one
could have its bent bracket pre-attached, and screw to the purlin on or
beside (less critical) the lower one's. The solar panels could then abut
horizontally over the centers of the corrugated ones, without any top cover.
If the patio roof is a single pitch you don't need a ridge cap.

Don't forget that you need to lift the panels onto the roof.

[bob prohaska]

Jim Wilkins <murat...@gmail.com> wrote:
>
> -------------------
> If you find the custom mounting brackets difficult to fabricate, what would
> you do for the vertical seams between them, and the ridge?

An upper panel frame would overlap the one below, with the panels at 45
degrees to the frame, runoff water will flow down from one to the next.
There have to be flashings at the top, sides and bottom, however.

Using old magazines on a table will let one construct the pattern.
It's not unlike shingles, except that the panels pile up 3 high in
the high spots. Imagine the far edge of the table as the ridge and
the near edge as the eave. With the panels at 45 degrees to square
each drains from one to the next downslope.

> Sheet aluminum
> mounting brackets are short enough to bend with hand tools but the others
> would require a long bending brake, such as a Van Mark or the Tapco like I
> used to cover my window frames. You may need some clearance between the rows
> to install and tighten the fasteners since one long strip of connected
> panels would be too clumsy and fragile to install as a unit.
>
> This is a press-in thread insert to put a strong threaded blind hole in
> aluminum, like the solar panel frames, where you can't reach behind with a
> nut.
> https://www.amazon.com/uxcell-Clinching-0-11-Inch-Stainless-Mounting/dp/
> B08XYLF4RH/ref=sr_1_5?c=ts&keywords=Self-Clinching%2BNuts&qid=1654275593&s
> =industrial&sr=1-5&ts_id=317399011&th=1
>
> Unfortunately they don't give the max metal thickness and the shorter ones
> for thinner metal may push out unless firmly staked, for which I made a
> portable tool on the lathe.
>
> Each of individual solar cell produces the panel's full rated current at
> 0.6V, so a small amount of shading from the edge of the panel above can
> cause a large reduction in output. If you angle the roof half way between
> the winter and summer angles you may lose power to shading in mid summer,
> and early and late in the day from the vertical cover strips, or have too
> little overlap to prevent splash from passing through the gap.
> https://gml.noaa.gov/grad/solcalc/azel.html

Early and late shading from panel overlap was something I hadn't considered.
It'd probably have to be accepted as a design compromise, with some hope that
bifacial panels would offer some mitigation.

>
> For an outdoor roof that doesn't have to be as absolutely storm-proof as a
> house you might put up corrugated panels with gaps between them to attach
> the panel brackets from below. At the panel overlaps the lower panel could
> connect to the roof purlin with a flat aluminum bracket and the upper one
> could have its bent bracket pre-attached, and screw to the purlin on or
> beside (less critical) the lower one's. The solar panels could then abut
> horizontally over the centers of the corrugated ones, without any top cover.
> If the patio roof is a single pitch you don't need a ridge cap.
>

> Don't forget that you need to lift the panels onto the roof.

The scheme would require the panels go in (and, if needed, out) one
at a time. With access from below (possible for a patio cover) that
might be a little easier than otherwise.

This idea is still very much a pipedream. The idea of bifacial panels,
either mounted at elevated angles on an otherwise-unsuitable roof or
flush in a shade structure, is a new twist to my knowledge.

Thanks for writing,

bob prohaska

[Bob F]

Are any of the "solar shingles" we have seen previously talked about
actually being sold?

[Jim Wilkins]

"Bob F" wrote in message news:udon7k$1dhje$1...@dont-email.me...

Are any of the "solar shingles" we have seen previously talked about
actually being sold?

---------------------------

My sister's neighbor in Maine has them. She doesn't know the details.

[Jim Wilkins]

"bob prohaska" wrote in message news:udoi9r$1d0rt$1...@dont-email.me...

Jim Wilkins <murat...@gmail.com> wrote:
> If you find the custom mounting brackets difficult to fabricate, what
> would
> you do for the vertical seams between them, and the ridge?

An upper panel frame would overlap the one below, with the panels at 45
degrees to the frame, runoff water will flow down from one to the next.
There have to be flashings at the top, sides and bottom, however.

Using old magazines on a table will let one construct the pattern.
It's not unlike shingles, except that the panels pile up 3 high in
the high spots. Imagine the far edge of the table as the ridge and
the near edge as the eave. With the panels at 45 degrees to square
each drains from one to the next downslope.

[[[[[[[[[[[[[[[[[
Do you mean to mount them diagonally, so the upper panels overlap on both
one side and the bottom? I think that would create more bracket and flashing
problems than it solves. The more I consider it the more I like straight
rows of panels that overlap at top/bottom with corrugated plastic panels
under the side gaps, perhaps white to match the underside of the panels and
split narrower to give more clearance to lift the panels from below and cost
less. Then you can install them and connect the wiring from a stepladder.
They might not completely block a wind-driven rain but would you be using
the patio then?

You could model the installation with 2x4s leaning on a sawhorse for the
roof and two panels tied to them with string, to test shading and drip
resistance. If you don't have better test equipment solar panels can be
shorted through an ammeter to measure output, compared to their rated Isc
(short circuit current).

I use this Volt-Amp-Watt meter to measure their power into a variable load:
https://www.amazon.com/PEACEFAIR-PZEM-031-6-5-100V-Electric-Industry/dp/B07RQTXNRN
For the load a 5 Ohm 250 Watt rotary rheostat is ideal for a 100W panel, a
cheaper tubular resistor with a sliding contact becomes too hot to adjust.
]]]]]]]]]]]]]]]]

Early and late shading from panel overlap was something I hadn't considered.
It'd probably have to be accepted as a design compromise, with some hope
that
bifacial panels would offer some mitigation.

The scheme would require the panels go in (and, if needed, out) one
at a time. With access from below (possible for a patio cover) that
might be a little easier than otherwise.

This idea is still very much a pipedream. The idea of bifacial panels,
either mounted at elevated angles on an otherwise-unsuitable roof or
flush in a shade structure, is a new twist to my knowledge.

[[[[[[[[[[[[[[[[
Some of my panels are on the lawn where I can manually track the sun. They
give very little output when facing the sky in the wrong direction. I made
folding supports for them from 1/2" EMT cut in half.
]]]]]]]]]]]]]]]]

[bob prohaska]

Jim Wilkins <murat...@gmail.com> wrote:
>
> [[[[[[[[[[[[[[[[[
> Do you mean to mount them diagonally, so the upper panels overlap on both
> one side and the bottom? I think that would create more bracket and
> flashing problems than it solves.

Exactly. At 45 degrees to vertical there is no "bottom", just two upper edges
overlapped by panels above and two lower edges overlapping the panels below.
Edges will of course require flashing.

This does result in a stacking effect, with panels three high at one corner.
The other corners will elevate by varying amounts depending on the panel shape.
But, the elevation above the frame is regular. Zero, one or two spacers seem
like enough.

> The more I consider it the more I like straight
> rows of panels that overlap at top/bottom with corrugated plastic panels
> under the side gaps,

Flat mounting is certainly the norm, and it's much simpler geometrically.
Maybe there's a simple flashing scheme that'll be rain-shedding.

> perhaps white to match the underside of the panels and
> split narrower to give more clearance to lift the panels from below and cost
> less. Then you can install them and connect the wiring from a stepladder.
> They might not completely block a wind-driven rain but would you be using
> the patio then?

To some extent, yes. It's nice to have outdoor space that's dry.
Not mandatory, but desirable.

> You could model the installation with 2x4s leaning on a sawhorse for the
> roof and two panels tied to them with string, to test shading and drip
> resistance.

I agree that a scale model might be the next step. My "National Geographic"
model on the dining table is helpful, but the magazines are too thin and
too flexible to reveal the stacking problems. It looks like about a dozen
panels are needed to see how stacking effects accumulate. Small boards with
shape and thickness proportional to panels would be helpful.

> If you don't have better test equipment solar panels can be
> shorted through an ammeter to measure output, compared to their rated Isc
> (short circuit current).
>
> I use this Volt-Amp-Watt meter to measure their power into a variable load:
> https://www.amazon.com/PEACEFAIR-PZEM-031-6-5-100V-Electric-Industry/dp/B07RQTXNRN
> For the load a 5 Ohm 250 Watt rotary rheostat is ideal for a 100W panel, a
> cheaper tubular resistor with a sliding contact becomes too hot to adjust.
> ]]]]]]]]]]]]]]]]
>

I'd think panel manufacturers would offer output-vs-angle figures, I'll look.

> Early and late shading from panel overlap was something I hadn't considered.
> It'd probably have to be accepted as a design compromise, with some hope
> that
> bifacial panels would offer some mitigation.
>
> The scheme would require the panels go in (and, if needed, out) one
> at a time. With access from below (possible for a patio cover) that
> might be a little easier than otherwise.
>
> This idea is still very much a pipedream. The idea of bifacial panels,
> either mounted at elevated angles on an otherwise-unsuitable roof or
> flush in a shade structure, is a new twist to my knowledge.
>
> [[[[[[[[[[[[[[[[
> Some of my panels are on the lawn where I can manually track the sun. They
> give very little output when facing the sky in the wrong direction. I made
> folding supports for them from 1/2" EMT cut in half.
> ]]]]]]]]]]]]]]]]

I can't really do meaningful experiments on the ground; too many trees.
Rooftop exposure is much better. For now it's just a thought experiment.

Thanks for writing!

bob prohaska

[Jim Wilkins]

"bob prohaska" wrote in message news:udq3gp$1kqea$1...@dont-email.me...

Flat mounting is certainly the norm, and it's much simpler geometrically.
Maybe there's a simple flashing scheme that'll be rain-shedding.

------------

Strips cut from corrugated steel roofing may be stiff enough to hold their
shape close to the panel surface. This will cut it lengthwise without
distortion if you are careful:
https://www.harborfreight.com/air-tools-compressors/air-tools/air-cutters-shears-nibblers/pistol-grip-air-shears-98580.html

If bent or wrinkled the corrugations can be hammered back into shape over
1-1/2" pipe, but it may not remain straight.