How's my solar heater design? Will it work? Where to put the vacuum breaker?

[Dirk]

Thank you for your honesty. It's definitely better to know sooner rather than later.

Thanks for taking it as I intended it.

Note the framing is just sort of a "proof of concept,"

That is a great approach, and I'm glad to hear you've got things squared away with the neighbor.

I didn't know if my pump would be strong enough to reach the roof,

Keep in mind that the pump only needs to overcome the initial fill. Once the water is flowing to the roof, the water coming down is "pulling" the water up the other side (simplified explanation). So once the panels are filled, the pump has much less to do. Mark is our "pump guy." If he says it'll work, it'll work.

I thought that throwing together some 2x4s was less of a commitment to get my feet wet (so to speak) with this solar experiment. In the back of my mind I also thought that once I get the system working I would consider moving to the roof, either to move the entire array there or to add an additional array.

All good stuff! With ya.

Second, my roof is not really optimally situated for solar.

Bummer, not with those panels anyway. I was projecting some. My roof is like your neighbor's. I filled one side with pool heater and the other side with PV panels. I kind'a lucked out. If it helps, all my pool solar is on the north side. I was told (especially during the summer), the sun angle is much less critical than with PV solar panels. In fact, the sun really arcs over the north face at least as much as the south face. That's not true in the winter, but I don't heat my pool in the winter. I saved the south side for PV. I expect in Arizona, in the pool season, your panels could be on any facet of your roof.

And yes, the array doesn't all need to be together. You can split it up as needed, but you do need to be aware that the connections (the plumbing schematic) are somewhat unintuitive and have to be arranged properly so that water is flowing from sub-array to sub-array most efficiently. That website I gave you explains how to do that.

When installing long solar panels on a roof, can I drape the panels over the ridgeline or use them in a 90 degree turn?

I would guess not. (1) The entire array is going to be expanding and contracting quite a bit, all day, every day. If the tubes are touching the roofing material, especially over the ridge, that dragging and sawing is going to mess up those tubes to the point of failure, in short order. I could go on for pages about what I did to make sure no part of my array, manifold or tubes, touches the roof. Only the mounting brackets and the pipes. That's the ideal.
(2) More importantly: solar panels are supposed to fill from one bottom corner and return from the opposite upper corner. You have that exactly right. But if the panels are folded over the ridge, that is no longer going to be the case. I don't know exactly what the ramifications of that would be, but I would expect it could greatly reduce heating efficiency, or it might not even work at all. Maybe @mas985 would know.

Any other mounting recommendations? And what hardware will I need?

I know that you have to accommodate expansion and contraction (as described above), and you have to account for wind loads. The mounts that came with my system do both, so I didn't have to learn or engineer anything in that regard. I just followed the instructions! Sorry, it's outside of my experience how to deal with those factors on someone else's array. Perhaps if you can identify the manufacturer of the panels, they can assist and advise what mounting system is appropriate for your installation. I'd call them as a first step.

BTW my other thought was to build a pergola with a slanted roof in the area between the pool and the house

I like that idea. I fly a shade sail over my pool, because even while in the pool the sun can be too hot. I expect AZ is even worse on the ol' noggin. I'd be tempted to build that pergola out over the pool a bit. That will reduce the sun on the water, so kinda contradictory to the heating effort, but I'd want a little area of shade that I could hang out in while in the pool. Better than umbrellas, which a lot of people use for this purpose, or shade sails like I do, because it would be permanent and totally stable (wind-proof). And maybe even accommodate an extra panel or two. So you might think in terms of getting a two-fer out of that pergola. I have two, and that's where everybody hangs out, under one or the other. So you'd be getting a "three-fer:" mounting for your panels, shade for your pool, and shade for your yard.

And you can always put some on the pergola and some of the roof.

Keep up the research and the brain-storming, you'll get there.

 

[everthustodeadbeats]

But if the panels are folded over the ridge, that is no longer going to be the case. I don't know exactly what the ramifications of that would be, but I would expect it could greatly reduce heating efficiency, or it might not even work at all. Maybe @mas985 would know.

@mas985 any thoughts on the feasibility of draping a solar panel over a ridge?

There are 2 scenarios. First is the one labeled in red in the figure below, where the feed header and return header are on opposite sides of the roof's slant. I would make sure that the return header is at a higher elevation than the feed header.

In the second scenario (shown in yellow below), both header tubes are on the same roofline -- and I'd angle the panels to make sure that the top of the return header (the exit from the panel) is at a higher elevation than the highest point of the feed header --- but due to the roof layout, there would still be a bump in the middle of the panel at the ridge point, which at its highest point would be a tiny bit higher (maybe 12-18 inches higher) than the exit from the panel header. If it would help it's possible I could add a little loop at the exit sticking straight up to compensate and render it higher than the ridge line.



Thanks!

 

[mas985]

@mas985 any thoughts on the feasibility of draping a solar panel over a ridge?

There are 2 scenarios. First is the one labeled in red in the figure below, where the feed header and return header are on opposite sides of the roof's slant. I would make sure that the return header is at a higher elevation than the feed header.

In the second scenario (shown in yellow below), both header tubes are on the same roofline -- and I'd angle the panels to make sure that the top of the return header (the exit from the panel) is at a higher elevation than the highest point of the feed header --- but due to the roof layout, there would still be a bump in the middle of the panel at the ridge point, which at its highest point would be a tiny bit higher (maybe 12-18 inches higher) than the exit from the panel header. If it would help it's possible I could add a little loop at the exit sticking straight up to compensate and render it higher than the ridge line.

Neither version will self drain which is going to be a PITA when winterizing.

Why not the flat area on the opposite (North) side of the house? Or is it actually tilted?

Or rotate the panels 90s and put them along the gable roof ridge on the left. Will they fit?

Had you gone with 4x8 panels, you might have made them fit on the left side of the gable roof.

 

[Dirk]

It's not that the return manifold has to be higher than the supply manifold, it's that the return manifold has to be higher than the highest part of every tube, and the supply manifold has to be lower than the lowest part of every tube. Every part of the plumbing to the return manifold also has to be lower than the return manifold. It's all gotta drain down to the solar valve when the pump stops.

Then there is this, which I'm not sure would be satisfied. Study the two animations, and their captions, on this page:

Plumbing a Solar Pool Heater | Hot Sun Solar Pool Heating

The basic concepts you must know to correctly plumb a solar pool heater.

www.h2otsun.com

 

[everthustodeadbeats]

Why not the flat area on the opposite (North) side of the house? Or is it actually tilted?

I will definitely put some on the north side. I do have a slight pitch change in the middle of the slope on the north side which might make the plumbing a little more complicated if I cross it. And I don't want the panels to be visible from the front of the house so I don't want to go down all the way to the edge of the roof. But I plan to fill up at least the top half of the north side (before the pitch change). I think I can fit two side-by-side stacks of either 3 or 4 panels per stack (and I want to plumb the stacks in parallel, correct?).
On the other hand, the lower half of the roof has shallower slant (more horizontal) --- is that better for the sun and should I want to put some panels there?


In any case, I asked about the south side because I think I will still have some panels left over after I put as many as I can fit on the north side. Also I hear that south-facing is better so I wanted to know if that's a possibility.

Or rotate the panels 90s and put them along the gable roof ridge on the left. Will they fit?

I'll try that, but it's going to be tight. Maybe I can get 1 panel at the very top of each side (east and west) near the ridgeline. Farther down the slope it will probably won't fit, unless I can rotate one of the manifold tubes 45 degrees on the north side (that is, the north-est part of the south side), or otherwise squish it so that not all of the panel will be totally flat against the roof. But that doesn't sound very secure, does it? I'll see what will fit and if it seems iffy I'll be sure to take some pics and post them here before I finalize.

Had you gone with 4x8 panels, you might have made them fit on the left side of the gable roof.

That definitely would have been my preference but I bought these used, locally on Facebook marketplace (for a price I couldn't pass up). These are a little unwieldly due to the extreme length, it's true. 8 or 10 feet in length would be the sweet spot.

Thanks again for all your help and feedback!

 

[mas985]

I will definitely put some on the north side. I do have a slight pitch change in the middle of the slope on the north side which might make the plumbing a little more complicated if I cross it. And I don't want the panels to be visible from the front of the house so I don't want to go down all the way to the edge of the roof. But I plan to fill up at least the top half of the north side (before the pitch change). I think I can fit two side-by-side stacks of probably 4 panels per stack (and I want to plumb the stacks in parallel, correct?).
On the other hand, the lower half of the roof has shallower slant (more horizontal) --- is that better for the sun and should I want to put some panels there?

Yes, for northside placement, flatter is better. But even on the tilt side, you can make up for the tilt by having more panels. It is just an efficiency trade off.

I'll try that, but it's going to be tight. Maybe I can get 1 panel at the very top of each side (east and west) near the ridgeline. Farther down the slope it will probably won't fit, unless I can rotate one of the manifold tubes 45 degrees on the north side (that is, the north-est part of the south side), or otherwise squish it so that not all of the panel will be totally flat against the roof. But that doesn't sound very secure, does it? I'll see what will fit and if it seems iffy I'll be sure to take some pics and post them here before I finalize.

Any time the panel is above the roof surface, wind can get underneath it and lift the panels which is not a good situation.

 

[everthustodeadbeats]

Any time the panel is above the roof surface, wind can get underneath it and lift the panels which is not a good situation.

That's what I figured.

I just realized something. If I take your advice and put panels on the east- and west-facing sides (assuming they will fit), how will I get the water to the east-facing panels at all? Don't I have to go over the roof ridge with PVC to get to the east side panels in the first place? Even though I'll be feeding the panels from the bottom, before the supply line reaches the panels it will have to travel to a point higher than the panels themselves. Isn't that a problem?

 

[Dirk]

On the other hand, the lower half of the roof has shallower slant (more horizontal) --- is that better for the sun and should I want to put some panels there?

Technically yes, and if it makes no difference to you or the installation, use that flatter area first. But if it's better for you to use the other section, do that, as it won't make any noticeable difference in the heating of your pool.

In any case, I asked about the south side because I think I will still have some panels left over after I put as many as I can fit on the north side. Also I hear that south-facing is better so I wanted to know if that's a possibility.

Yes, south is better, but I'd recommend you do what I did. Save the south-facing areas for PV solar. For PV, north or south makes a huge difference (for pool heating it doesn't). Even if you're sure you won't ever have PV, the next owner might, and "PV-ready" could conceivably be a selling feature. On the other hand, if this is your forever home, I'd be willing to bet you'll someday have PV solar. We all will.

how will I get the water to the east-facing panels at all?

The return plumbing can go over the ridge. It's the supply plumbing that shouldn't. Just picture how the water will gravity-drain to the pool pad. That's the goal. The water will drain from both the return plumbing and the supply plumbing, it just needs to drain completely. You don't need to avoid high spots, you only need to avoid low spots.

So the challenge is getting the supply pipes to the array without going higher with them than the array's supply manifold. You might have to run a pipe underground and then up the wall to get to one side or the other (that's how I did it). Or across the back of the house under the eve.

The supply or return pipes can go underground, lower than the pad (mine do). They'll trap water, but they won't trap water up on the roof (that's the main consideration). If at the end of the season you need to remove that trapped underground water, then you can blow it out. You could plumb in isolation valves and drain valves to facilitate that. Which will all be much easier to deal with than trapping water up on the roof.

If your frost line is like mine (I don't have one), then you can just leave the water in the underground pipes as I do.

 

[everthustodeadbeats]

So the challenge is getting the supply pipes to the array without going higher with them than the array's supply manifold. You might have to run a pipe underground and then up the wall to get to one side or the other (that's how I did it). Or across the back of the house under the eve.

I suspected as much. I could run it around the side of the house under the eaves, but more likely I'll just skip that side and not bother with a panel there.

If your frost line is like mine (I don't have one), then you can just leave the water in the underground pipes as I do

Here in Michigan it freezes pretty hard in the winter. You can't get away with leaving any water in any underground pipes. I blow out all the pipes and pour in antifreeze as part of winterization.

 

[Dirk]

Here in Michigan

Yes, of course. Sorry, got you mixed up with someone in AZ I'm working with. Sounds like you already winterize your pool. If you have to go underground, that'll be just one more pipe to clear.

 

[Dirk]

Here's a site that has some info about sun angles throughout the year, across the U.S. You can check to see if the various sections of your roof are going to make a difference. I just did a quick google search for this, not necessarily endorsing the site's data.

Site Maintenance

This illustration depicts what I was describing, how in swim season the sun is probably going to hit all your roof well enough for pool solar, but why you'd want to save the south side for PV, because you want the best angles all year round to maximize electricity generation.

 

[mas985]

Regarding the setup below, there is another option.

When you have low points that won't self drain, you can plumb in a smaller diameter line at the lowest point and have a separate route back to the solar supply/return that travels on a continuous downward slope. The advantage here is the line can be as small as irrigation tubing and it will still drain the panels, albeit very slowly plus the smaller line can be routed where it is not noticeable.

 

[everthustodeadbeats]

When you have low points that won't self drain, you can plumb in a smaller diameter line at the lowest point and have a separate route back to the solar supply/return that travels on a continuous downward slope. The advantage here is the line can be as small as irrigation tubing and it will still drain the panels, albeit very slowly plus the smaller line can be routed where it is not noticeable.

Interesting. But I think I will start with filling up the north side, then if I need more room go to the southwest, and I'm not going to worry about that southeast side for now. I know my limits and don't need that kind of hassle in my life right now

Going back to the check valve discussion, I really don't have any room between the filter and the 3-way, but like @Dirk suggested, I could make some room between the pump and the filter. I might have to modify the plumbing a bit to make room, so quick question: I know that there's a thing about needing 10-12 inches of a straight run leading into the pump from the supply, but is there a similar issue with the filter? Or can I put a 90 leading immediately into the filter?

 

[mas985]

Going back to the check valve discussion, I really don't have any room between the filter and the 3-way, but like @Dirk suggested, I could make some room between the pump and the filter. I might have to modify the plumbing a bit to make room, so quick question: I know that there's a thing about needing 10-12 inches of a straight run leading into the pump from the supply, but is there a similar issue with the filter? Or can I put a 90 leading immediately into the filter?

That is a bit of myth. Many manufactures will suggest that in their manuals but in reality, pumps perform nearly identically with the 90 far away vs near the pump.

The filter doesn't matter at all.