Air Gap in Pump Filter with Solar Panels

Jump to latest Follow Reply
By the way, Mark, in case we've sent you running for the hills, everything we're talking about, if it turns out that it was done wrong, is totally correctable. Probably in the few hundred dollar range. Most everything you bought is probably OK, just needs to be reconfigured a little.

Theoretically, the installers should come out and do most or all of it for free, if they can be convinced they did it wrong.

First step is to get a hold of the panels' owner manual, if there is one. That's free! ;)

That should spec preferred installation configuration and optimal flow rate.
 
Dirx, that link you posted earlier had a different piping setup than what I have. . I do have a vacuum breaker. I’ll ask the company about the draining path. There is a check valve (item in grey in pic) upstream of the diverter. There is a way to isolate the return/supply with vent valves on each line to drain for the winter. My brand of solar panels is “Solar Industries”. Actually, the Solar Industries brochure they gave me shows a picture of the inlet/outlet on same side of the roof and the supply going to the lower right and coming out the top right, JUST like I have. I’ll find it or take a picture when I get home.

Poolzzz: I was told by the installers that the panels are tilted – I’ll have to take a pic with my good camera. Like Dirx said, I’m not sure what my flow rate is at 26psi – is there any way to measure?

I appreciate you all helping me out!!
 
To be clear ... the pipes may leave the roof at the same side, but they should enter/exit the panels on opposite corners.
There may be a run of PVC just above the panels that we can't see in the picture from the upper left corner exit over to the right to leave the roof.
 
If you zoom in on the picture there appears to be a pipe running across the top of the panels. Also on the left there is a U shaped pipe where it connects to the panels. At least that is what I see.
 
mas985 said:
If you zoom in on the picture there appears to be a pipe running across the top of the panels. Also on the left there is a U shaped pipe where it connects to the panels. At least that is what I see.
Click to expand...
Gets kind of blurry, but I think I agree that I see the U on the upper left, so likely a parallel run of pipe above the panel header. Meaning they would be plumbed fine.
 
These off the “Solar Industries” website. You can clearly see in the diagram how to plumb the panels. Not so clear is the photo, but if you look closely you'll see the return starts from the opposite corner from the supply line. Is it possible you didn't see that in the photo in the brochure? Is it possible that you can't see from the ground your return pipe going around to the other side of your array, above/behind the array? When I first looked at your roof photo, I thought I could barely make out the telltale set of 90s up in the left corner. It really would be weird for your installers to get that wrong.

The illustration also shows the check valve I was describing. It's optional, so not a big deal it's not in your system. And the ball valves wouldn't be causing your issues. They're just not a quality part.

Can you hear your vacuum breaker "croaking" when the filter pump shuts off? Then that's working as it should, and your system is probably draining as it should. Do you see a large amount of bubbles in the pool when your system first starts up each day?

The website has a chart with the optimal flow rates for their products. Here. Without the model number I can't quote it here, but it's around 4 or 5 GPM per panel, about the same as mine.

schematic.jpg panels.jpg
 
OK, consensus? So if the panels are plumbed correctly, then is it just a matter of using the three-way to bypass some of the flow to optimize the panels? I'm still bugged by 2" going up and 1.5" coming down, but I can't offer any science about that, one way or another. It seems like that is an unnecessary restriction. Maybe that's by design, to offer up a little back pressure? I read about back pressure somewhere, but can't now remember why it's needed or how best to achieve it. It might have something to do with flow. If there's a little back pressure, then the water would not find such an easy way out, and be inclined to fill and flow through all the panels and tubes more evenly?
 
goody222 said:
Poolzzz: I’m not sure what my flow rate is at 26psi – is there any way to measure?
Click to expand...

FlowVis! (Honest, I'm not on commission.)

Isn't there a quickie formula for flow from [something] x pressure gauge reading?

I've heard others can feel the panels and determine if they're getting the right flow (by their temperature). Or the water coming out of the returns (should only be a few degrees above the pool temp). Old school, baby!
 
If the pump is an up rated 1.5 HP TriStar (SP3210x152), based upon the available information in this thread, the flow rate will be around 66 GPM.

For those interested, you too can estimate flow rates in plumbing systems with the tools in my signature.
 
just spoke to the company! The supply on the right side goes ALL the way down to the left side, U-turns, THEN supplies the panels. Also, I count several shingles higher on the left side, meaning it is tilted correctly. Also the design is such that the panels will depressurize, not drain, when the pump turns off. He is talking to the installers about why we use 1 1/2" piping return and if they can come out and play with the cams in the diverter free of charge. I've been worried about this and feel a bit better now.
mas985, I see that my pump (67gpm) is close to the max for 7 panels. Also before the system turns off, my ThermoWorks food thermocouple shows the temperature on the inlet vent is the same as the outlet vent when I crack them open, either meaning there isn't enough temp differential OR flow is going through too fast. Sounds like bypassing will help - we'll see what the company says.
Clearer picture below.
IMG_1304.jpg
 

Enjoying this content?

Support TFP with a donation.

Give Support
goody222 said:
The supply on the right side goes ALL the way down to the left side, U-turns, THEN supplies the panels.
Click to expand...
That does not seem to match what your picture shows.
Also the design is such that the panels will depressurize, not drain, when the pump turns off.
Click to expand...
They should drain (and they will). You do not want them to "depressurize" ... whatever that means. If they do not drain they will super-heat in the sun and pull a vacuum due to the weight of the water column and can damage the panels.

There should be a VRV up on the roof and the panels should drain down through the 3-way (did they drill a hole in the diverter or use a solar valve?) or some will also be drained via siphon up and over to the the return pipes.
 
goody222 said:
just spoke to the company! The supply on the right side goes ALL the way down to the left side, U-turns, THEN supplies the panels. Also, I count several shingles higher on the left side, meaning it is tilted correctly. Also the design is such that the panels will depressurize, not drain, when the pump turns off. He is talking to the installers about why we use 1 1/2" piping return and if they can come out and play with the cams in the diverter free of charge. I've been worried about this and feel a bit better now.
mas985, I see that my pump (67gpm) is close to the max for 7 panels. Also before the system turns off, my ThermoWorks food thermocouple shows the temperature on the inlet vent is the same as the outlet vent when I crack them open, either meaning there isn't enough temp differential OR flow is going through too fast. Sounds like bypassing will help - we'll see what the company says.
Clearer picture below.
View attachment 85308
Click to expand...

Going too fast is fine. The energy picked up is the same just less per pass.

Declaration: I design fiberglass lay ups for filters to British Standard 4994

Act Shirley I strong dislike this set up. It may be standard practice in the States but it places too much backpressure on the filter.

Looking at the filter first. The filter is a pressure vessel. It is designed to flex a little on start up (assuming made correctly and actually engineered not a copy of a copy). Its usually visible as a minor expansion. Changing pressures by sending flow to the roof, changing pump speeds etc can cause the filter to flex and flex and flex. As it ages the overall filter life is reduced.

Personally I prefer to fit a booster pump to the solar on the roof. Especially in a double storey lift situation. It reduces pressure changes on the filter vessel.


Depressurise...really? So I have water sitting on the roof for possibly weeks and weeks without circulation and then one day I turn the solar on and I get old water back to my pool? Do bacteria not grow in solar panels? Or was this a simple explanation to get out of a call?
 
You don't want your GPM to be near the max, you want it to be at the optimum. Use the chart. It's closer to 30ish. There are diminishing returns. Too low or too high and you lose heat exchange efficiency. The upper limit is spec'd so you don't blow out the panels, it's not more is better in this case.

Jason, he thinks he's got the VRV. I asked him if it croaks when the pump stops. That would confirm the VRV and drain-down valve are working. Though not that the panels drain completely. He's confirmed the tilt, so he's probably OK there.
 
goody222 said:
Trying to read the pump curve for a SP3210X152 pump - the following only gives the "Low Speed" specs. Any idea how to figure out the High speed for that pump? https://hayward-pool-assets.com/assets/documents/pools/pdf/literature/TriStar-MAXRate-SpecSheet.pdf?fromCDN=true
Click to expand...

I don't know why they don't show both curves but it is easy enough to figure out the high speed from the low speed curve.

If one has a pump curve at one speed they can always translate to another speed using the general pump laws that flow is proportional to rpm and head is proportional to rpm squared.

On these pumps high speed is double low speed. So for any point on the low speed curve the corresponding high speed curve would have 2x the flow and 4x the head. The reverse is true also, cut the rpm in half and the pump will have half the flow and 1/4 the head.

What are you trying to do? If you are trying to estimate your system flow rate based on 26 psi (60 ft of head) then the corresponding low speed curve is 15 ft of head which gives 40 gpm. Meaning high speed would flow 80 gpm (but see next para).

However your actual flow will be less than 80 gpm because 26 psi is not the total head of your system. It is only the portion from your gauge to the pool return. You have to also include the head from the pool inlet to the pump (suction side head) and the head from the pump to wherever your gauge is. So your total system head is actually more than the 60 ft indicated by your gauge. I'm not sure since i haven't looked, but from mas985 comments it sounds like he has spreadsheets that calculate all this out correctly. He's already given the value as around 66 gpm.

Finally, after all that, I personally don't think optimum flow is a big deal even with solar panels. More important is pressure. Your pressure is to high. Since low speed doesn't have enough head to work given the height of your panels, you have to run high speed and bypass to get a reasonable pressure as per mas985 instructions in the beginning of the thread. Do that and your flow will be in the ball park which is plenty good enough.

...and I meant the u-turn is on the top.
Click to expand...

Ok, but that would then be inconsistent with your comment from the installer that this is the supply feed and the panels are fed from the left.

You also said the panels are tilted and declared the tilt correct but didn't actually say which side is higher.

The picture is fuzzy enough we can talk ourselves into anything. Probably it's OK but still not totally clear. A better picture would be useful since there's a lot being lost in the words.
 
Finally, after all that, I personally don't think optimum flow is a big deal even with solar panels. More important is pressure.
Click to expand...

Respectfully, I don't agree, but I have no science to back that up, other than manufacturers' recommend flow rates. Since I've seen similar recommendations across multiple brands, I'd like to think they've each done their own independent testing and have determined their own best rates for their own products, which happen to corroborate each other. That could certainly be a naive assumption, but I'm assuming that none the less.

I don't believe the amount of flow through a panel is like salt to an SWG. It's not "once you get enough flow, you get all the heat you're going to get." I believe the heat exchange efficiency is on a curve, and that it relates and tracks directly with flow. So there is an optimum point that one should shoot for, if one is shooting for the warmest pool at the lowest cost.

What I don't know for sure is once you pass the optimum point, if you actually lose heat exchange efficiency (because the water is in the panel for too short a time), or if you're just wasting pump energy and losing cost efficiency because of it. I suspect it's both.

That's not to say pressure is to be ignored. Too much pressure could certainly shorten a panel's lifespan, and potentially cause a premature failure. But I'm virtually certain that if you satisfy a panel's optimum flow rate, it'll be at a pressure that is safe for the panel, so pressure is the less important concern. They're designed to handle the pressure of the optimum flow rate. So I'm sticking by my installation and setup technique. I used a FlowVis to establish the manufacturer's suggest flow rate, and continue to use it to monitor and adjust flow rate as things change (filter pressure, for one). I think I'm getting better heat exchange because of that, which means my system reaches my heat set point sooner than it would otherwise, and so I have to run my pump less. Eventually the FlowVis will pay for itself, if it hasn't already...

And, for me, the combination of the Vis and the VS pump is ideal. When I catch my panels' flow off by even 1 GPM, I just go into my automation and tweak the pump's RPM a bit to bring the flow back to spec. It's really easy to run. I'm not fooling with valve cams or sticky ball valves, just a push of a software button.
 

Enjoying this content?

Support TFP with a donation.

Give Support
You must log in or register to reply here.
Thread Status
Hello , This thread has been inactive for over 60 days. New postings here are unlikely to be seen or responded to by other members. For better visibility, consider Starting A New Thread.
Top Bottom
Back
Jump to latest Follow Reply