Solar Panel Mounting Sanity Check

[Danjo]

In regard to fasteners

I used a lot of nylock, lock washer and double-nut configurations on everything that got attached to the bus, but for my panels I’m relying only on locknuts and judiciously applied torque. So far so good. but at some point I came across the Top Lock Flange Nut and wish that I had found them before doing the panels. The nut has a couple indents in the top of the nut that add enough friction that they will never freely spin loose and there’s no nylon bushing to rot out in the sun

https://www.clipsandfasteners.com/M-...iABEgJB4PD_BwE

[Salty Walty]

It seems like the aluminum frame is often the failure point for when people lose panels (bolts tear through the aluminum as they’re not built to be especially robust in weight bearing, especially when connected to one another without additional support). On another skoolie forum, I had suggested two length wise struts, with perpendicular struts spanning the distance (roughly six feet) and attaching the panels to the perpendicular struts. I was told that’s too far of a distance for a strut to bear weight.

I found the spec sheet for P1000 from unistrut and, with an assumed weight of 200lbs per 100W panel (someone else’s math based on speeds of 90mph), it does seem a bit risky to not include a center strut. https://www.unistrut.us/product-details/P1000
That said, I'm trying to find the sweet spot between safety, being conscientious about weight (struts are heavy), and allowing for a modular design that can be updated. This is my dilemma.

[Danjo]

Originally Posted by Salty Walty

It seems like the aluminum frame is often the failure point for when people lose panels (bolts tear through the aluminum as they’re not built to be especially robust in weight bearing, especially when connected to one another without additional support). On another skoolie forum, I had suggested two length wise struts, with perpendicular struts spanning the distance (roughly six feet) and attaching the panels to the perpendicular struts. I was told that’s too far of a distance for a strut to bear weight.

I found the spec sheet for P1000 from unistrut and, with an assumed weight of 200lbs per 100W panel (someone else’s math based on speeds of 90mph), it does seem a bit risky to not include a center strut. https://www.unistrut.us/product-details/P1000
That said, I'm trying to find the sweet spot between safety, being conscientious about weight (struts are heavy), and allowing for a modular design that can be updated. This is my dilemma.

If the supposed failure point is on the panel frame, assuming that all connection points are utilized, how can adding any amount of strut change that? I suppose if the panels are set high enough off the roof so they become wings or kites and the panel frame is the weakest link then that’s where it’s going to fail. I think that in that scenario it’s an installation design failure.

Keep the panels as low profile as possible. I’ve seen some installations with frames designed for a flat installation on a crowned roof. It doesn’t look right and there’s a lot of surface area exposed to the wind. I see the reason for setups like this. Either it’s the idea that a flat panel is the least suboptimal configuration or the chosen panels are too big for the purpose. In the case of trying to have an array of panels that are all facing the same way, in this case all flat, the assumption is that the panels will all catch sun. Well, OK. The problem is that this assumption does not include the position of the bus in relation to the angle of incidence and if you’re lucky it’s going to be right once a day. It’s a better solution in the case of a vehicle to over-panel by 25% and keep the panels as close to the roof as possible.

We are all very quick to run to the home centers for parts. I know I did when I got the strut. I got the low profile strut. It’s 17 pounds per stick. If you look around you’ll find aluminum low profile strut that’s half the weight. I found it at Grainger that I purchased for another installation. It’s a little pricy. I think it was $10/ft.

[Salty Walty]

The failure point is often the panel frame, but especially when the frames are connected to one another without bolstering (there would still be four contact points per panel with perpendicular struts). I definitely get the point about low profile and also the fixed positioning being less than optimal. Again, sort of why I’m obsessing over a sound base design that won’t require me to open up the ceiling in the future. I could build something low and adjustable for 100W panels, but then I’m stuck with that.

[Danjo]

I haven’t seen a failed frame connection or Z clip. The failures I have seen have been sheet metal screws torn from the sheet metal roof. Never from a through-bolted connection. I’d love to see failed installations if you have some photos. That would make a great thread.

[HamSkoolie]

Our design, and a number of others, purposefully mount panels well above the roof and allow air flow under the panels. This is done to reduce the heat load applied to the roof by utilizing the panels (as well as a deck or other covering) as a sun shade thus eliminating or significantly reducing solar heating of the roof and keeping it near ambient.

Anyone who has worked on things outside in the desert knows, the ambient temp is nothing compared to the temp of things in direct or reflected solar heating. We would routinely fry eggs(and occasionally bacon) on the wings of the F4's and later the F18's when stationed at Yuma or 29 Palms to demonstrate to the new troops just why they were under orders to wear gloves on the flight line.
Such designs, many using unistrut, others using tubing and/or angle, are working just fine with no failures I'm aware of.

[Simplicity]

I have two thoughts/concerns:

1) Per the OPs picture of the bracket attached to the roof with the channel hanging off the L bracket, I'd move the channel to the other side of the L bracket to provide more suppor over the L brackets base.

2) Per Ross's suggestion for adhering/sealing the brackets, if it's bare, or maybe even scuffed paint, that's cool. Yet, if you've coated your roof with something like and elasticseal product, you may want to consider removing that from under the brackets.

[DoubleO7]

What was the overall area of panels, or just the width of panels from one side of bus to the other?
One P1000 supported every six feet is good for 560 lbs uniform load.
The lengthwise-of-bus-struts would be installed in from panel edges maybe 12" or so, the perpendicular struts being out to full width.
A low profile strut down the high point of roof could also be utilized adding 50% more holes in the roof.

[LargeMargeInBaja]

Originally Posted by TJones

A rivnut would be much stronger. A self tapping screw or bolt will not hold much on thin sheet metal.

Ted

.
Ted is correct! [ applause ]
.
A bolt into a nut has >three or four or five+ turns to snug.
A screw into sheet-metal has part of one turn... the thickness of the sheet-metal (16-gauge?).
If your roof was 1/8-inch steel plate, your screw might 'bite' a turn, turn-and-a-half.
Most roof material is less than that, much less.
.
A bolt-n-screw combination is engineered to slightly stretch under the load of proper torque, smushing the threads into an almost-welded condition.
A screw into sheet-metal will pull out before any adequate torque is reached.
.
A bolt-n-screw combo is engineered for a load parallel to the bolt shaft.
With a screw into sheet-metal, you are the engineer (aka 'test-pilot').
.
.
The way I see my purpose on this forum:
* I am here to suggest alternatives to costly learning experiences... injuries to your wallet or to your delicate sensitive favorite flesh.
Me puncturing holes in your carefully-crafted plans doesn't bother me a month-old Twinkie©.
Listen, do not listen... either way, I am good.
.
I may come across as a cranky old know-it-all... only because our time together is limited.
That, and this whole 'bus fiddling with' is supposed to be fun... and chasing treasured chunks of equipment after it leaves its home at high-velocity -- often with great ceremony and to the delight/dismay of innocent by-standers -- is less fun than it sounds.