diy solar

diy solar

Solar "Hump" for aerodynamics for Class A RV

I would not screw into the roof material unless I knew I would also hit the structural member below it, but that's just me.
due to the size of the panel (80x40) it is almost impossible without building a massive racking system - to only screw into structural members.
Since the RV roof is made of 96x48 pieces on 16 inch centers.

Base Panel simulation side by side:

aU13YJv.gif
0n1uZFV.gif


When you look at them in direct comparison - no only the tail wagging vortex is less - (which I can see and feel while driving)

Also the red pressure zone above the RV is not as tall. Interesting, that is actually what is predicted in the Studies I posted on the first page.
 
due to the size of the panel (80x40) it is almost impossible without building a massive racking system - to only screw into structural members.
Since the RV roof is made of 96x48 pieces on 16 inch centers.

Base Panel simulation side by side:

aU13YJv.gif
0n1uZFV.gif


When you look at them in direct comparison - no only the tail wagging vortex is less - (which I can see and feel while driving)

Also the red pressure zone above the RV is not as tall. Interesting, that is actually what is predicted in the Studies I posted on the first page.
nice!
 
due to the size of the panel (80x40) it is almost impossible without building a massive racking system - to only screw into structural members.
Since the RV roof is made of 96x48 pieces on 16 inch centers.
A 1/8" X 1" aluminum strip fastened to the roof structural members every 16" and then bolted to the panel system would not add much thickness.

In any event, consider using blind rivet nuts if you are only screwing the panels down to sheet metal. Buy the correct depth nuts for your sheet metal thickness, and make sure they are not too long overall for the gap between your roof and ceiling.

You could also use 3M 5200 sealant to seal around the roof holes.

If you wouldn't consider standing over your installation and jerking it upward as hard as you can, it probably isn't nearly good enough to drive around.
 
I like the tape. I think it's easy to work with seals well, don't need to worry about a tube of lap sealant drying up when I don't use it.

Some manufacturers started using it as standard between the front and rear fiberglass caps as a standard installation. So must have some merit.
Hey, when I replace my vents, I do the same thing; it lasts where lap sealant doesn't. Also, when I pulled the J-Rails off the roof to the sidewall joint, I went back with the 4-inch tape folded over the seam and then replaced the J-Rail over it.
 
A 1/8" X 1" aluminum strip fastened to the roof structural members every 16" and then bolted to the panel system would not add much thickness.

In any event, consider using blind rivet nuts if you are only screwing the panels down to sheet metal. Buy the correct depth nuts for your sheet metal thickness, and make sure they are not too long overall for the gap between your roof and ceiling.
how do you propose using that strip? Just flat outside on the roof? The roof is made out of ribs running from outside wall to outside wall.

1642292095754.png

This is a stock picture - the manufacturer my RV was built, is no more, the folded in 2008 or so - so I'm not able get any details about the construction. I know there are Aluminum, or steel box trusses (roughly 1x1 inch) in the roof - because I saw a metal truss through a hole when I removed the antenna.
The roof decking is wood - somewhere between 1/8 and 1/4 thick, under that is Styrofoam. The trusses must be embedded in that foam.
- the trusses are not - perfectly every 16 inches - just on average.
When you got a heavy item on the roof - like the A/C, or a cut out around a vent or skylight - the trusses are moved to adapt.

Another crazy idea is what I got from how homes in Florida are built:
Tie downs and piers. -if I'm able to find solid trusses and get a got anchor point attached to it - I can use some turnbuckles to tie the panel frame against the roof.

WindZone1Anchors.jpg


Practically - piers - are the solar panel mounts - which are holding the panel Up against gravity and also against sideway movement.
While tiedowns - have only have the task providing tension against uplift.
 
1642308034032.png
1642308068407.png

still messing up a lot while learning flowsquare.. this 3d model was generated from two BMP input images*

*(the attached images are not BMP format)

1642308161458.png

1642308175601.png

1642308274735.png
 
regarding accuracy in modeling.. photogrammetry is an option that has been brought up, taking photos from different angles towards the end of generating a 3d model of the exterior geometry ???

metashape (formerly photoscan) photogrammetry program is one i've used for years to generate 3d models from multiple normal everyday 2d images coming from normal cameras, be it phone or DSLR; it uses CPU and/or GPU mode (CPU only, GPU only, CPU+GPU, i usually use GPU only mode.)

if you do a photogrammetry image sequence of the bus and generate 3d model, or even just 3 photos, one from above, one from side, one from front/back. if given a 3d model, i would orthographically render its outline from each of the three origin axes, X, Y , Z, looking towards origin, and input that to flowsquare. i've got spare cpu cycles ?

due to the general convexity of the RV, at a minimum, 3 images, one of each plane should be able to mostly approximate the features?
 
1642312691131.png
using three BMP, one for top view, one for side, one for front

(it's a spaceship in this pic, deal with it! ;D)
 
3d_anim.gif
model generated from 3 BMP images
figuring out 3d mode
no panel config
rough draft

edit; the blocking percentage is like 30-50% depending on axis. tomorrow will try to get decent geometry detail on a 5-10% blockage ratio, even though 0-3% is optimal from what i heard.
 
Last edited:
how do you propose using that strip? Just flat outside on the roof? The roof is made out of ribs running from outside wall to outside wall.
Yes, flat strips running perpendicular to the trusses. Use flat head bolts and blind nuts so that your panels are tied to the frame.

panel_strips.png

Then you can screw your panel brackets to the strips. I would put studs in the strips from the other side before I fastened the strips to the roof.

panel_strips2.png
 
Yes, flat strips running perpendicular to the trusses. Use flat head bolts and blind nuts so that your panels are tied to the frame.

View attachment 79930

Then you can screw your panel brackets to the strips. I would put studs in the strips from the other side before I fastened the strips to the roof.

View attachment 79932
Thanks for the idea.

That would certainly tie the whole thing into the main structure. Yet, 1/8 aluminum strip wouldn't give much resistant against bending up.
So the strips would have to be at least 1/4 thick or - L- Angles, and then I'm running into water drainage issues.

While your solution is distributing the load over a wider surface - that's not the direction the force is going.

But your drawing and straps gave me an idea - the panel frame is basically a strip - When I just connect the panel frame - with trusses in the roof ? What do you think? Connection points marked in red.

panel_strips3.png

I've just verified that my studs are Aluminum, and ordered an aluminum stud finder.
 
Some manufacturers started using it as standard between the front and rear fiberglass caps as a standard installation. So must have some merit.
Merit: to save labor cost on assembly deck.

I’ve done warranty repairs to tape stuff. Design errors, assembly mistakes….. tape doesn’t flow. Sealant is pretty guaranteed.
don't need to worry about a tube of lap sealant drying up when I don't use it.
I considered a tube to be a ‘cost’ on anything roof-wise. If it has any leftovers and it wastes away- it was already paid for in the work.

Lap sealant with or without reinforcing fabric is way, way, way less expensive than a roof leak in an RV. Buy it use it if there’s anything usable left over that’s a freebie
Either polyurethane caulking or lap sealant never came back failed.
 
Merit: to save labor cost on assembly deck.

I’ve done warranty repairs to tape stuff. Design errors, assembly mistakes….. tape doesn’t flow. Sealant is pretty guaranteed.

I considered a tube to be a ‘cost’ on anything roof-wise. If it has any leftovers and it wastes away- it was already paid for in the work.

Lap sealant with or without reinforcing fabric is way, way, way less expensive than a roof leak in an RV. Buy it use it if there’s anything usable left over that’s a freebie
Either polyurethane caulking or lap sealant never came back failed.
I'm not doing much RV roof work. So my use case is - having a piece of tape while being on the road, tubes tend to dry up after a few years - even without being open.

I know Tape is not for every application, roof needs to be clean, pressure needs to be applied everywhere.
Eternabond doesn't like humidity - at all. I know it's not working in the morning when it's damp - literarily the only time I can use it - after a long warm day when the whole RV is warm enough.

So the window of opportunity is pretty small. But - when you get over those shortcomings - its a good product - once applied correctly it usually good forever.
 
Which way do you suppose the force is going?
any direction away from the roof. Up straight- Up right - Up Left - Up front - Up back.

Force direction down I got covered - the 6 regular mounts are solid against gravity.

The solar panels I got - have very substantial frames. They are double extruded - so nice and strong. So I would like to utilize this.

structur idea.png

Not up to scale or dimensions, early morning.... need more coffee. The standard solar panel mounts - sit on one truss for each panel. In the middle of the panel - I'm going to put some L- brackets to tie into the trusses which are underneath the panel. Those have to mainly provide support against up.

What do you think?

I fully agree with your suggestion to tie into the main trusses, Just searching for a way, without interfering with water run of and increasing the weight on the roof.
 

Attachments

  • 1642421074535.png
    1642421074535.png
    1.7 KB · Views: 2
What do you think?
Yes, exactly that. You seemed to suggest early on that you couldn't do that, so I was trying to come up with a way to get it done. I would still suggest using blind rivet nuts to help distribute the force on the roof frame members.

It would probably be good to skirt the sides and the gap between the two with sheet metal to keep large eddy currents from forming.
 
You seemed to suggest early on that you couldn't do that, so I was trying to come up with a way to get it done. I would still suggest using blind rivet nuts to help distribute the force on the roof frame members.
probably didn't get it right away :p my apologies.

I see what you mean with Rivet Nuts now. It didn't make sense to me initially - because most of my solar panel mounts are screwed into wood.
But the trusses are tube aluminum. So something like a rivet nut would certainly increase the pull out strength considerable.

So there is about 1/8 to 1/4 inch wood above the the truss- how do I get the riv-nut to seat correct?

What about just a regular Rivet? https://www.amazon.com/Aluminum-Rivets-100pcs-626-750-Persberg/dp/B088G32BK1/
Something with a large flange? I've worked with rivets in the past - and those things really never give -the get loose sometime - but never had one pull out.
 
probably didn't get it right away :p my apologies.

I see what you mean with Rivet Nuts now. It didn't make sense to me initially - because most of my solar panel mounts are screwed into wood.
But the trusses are tube aluminum. So something like a rivet nut would certainly increase the pull out strength considerable.

So there is about 1/8 to 1/4 inch wood above the the truss- how do I get the riv-nut to seat correct?

What about just a regular Rivet? https://www.amazon.com/Aluminum-Rivets-100pcs-626-750-Persberg/dp/B088G32BK1/
Something with a large flange? I've worked with rivets in the past - and those things really never give -the get loose sometime - but never had one pull out.
I think you can buy rivet nuts with the correct depth range to do your job. I would go with stainless because I think aluminum is too weak. I have had large head aluminum pop rivets shear off (not the head, but between the panel and the truss) with motion from trailer walls. Even though they didn't pull out, they can't take the motion that will eventually shear them.

Stainless rivet nuts for 0.165" to 0.26" depth at McMaster:

Probably worth buying the cheapie tool to install them also:

It might pay to practice installing a few on some scrap before you begin the real deal....
 
I would go with stainless because I think aluminum is too weak. I have had large head aluminum pop rivets shear off (not the head, but between the panel and the truss) with motion from trailer walls. Even though they didn't pull out, they can't take the motion that will eventually shear them.

Stainless it is then good argument. Now I'm going back in forth between Rivet Nuts and Regular Pop Rivets

The Rivet Nuts + Screws are larger and potentially removable. But through the simple handling of regular Pop Rivets - I can just use more per bracket.

1642503922717.png
1642504130060.png

I'm trying to read those data sheets - it seems like a 4.8MM Rivet has about double the tensile strength - even compared a M10 Riv Nut?



I don't find a apples to apples comparison in pull out strength.

The Nuts are fastened to the roof structure first and then a bracket is screwed into it. (Loctite, Locking washer) The Regular Rivets - would fasten the bracket directly to the structure.

I'm leaning towards regular rivets - since this is safety structure and I see less user error and potential for screws coming out down the road (and it will be down the road)

Thanks for your time again.
 

Attachments

  • 1642503874887.png
    1642503874887.png
    37.9 KB · Views: 4
Back
Top