# How does a roof-mounted solar system affect system head?

#### Titanium

I am trying to decide on the exact upgrade path from an existing 2 HP single-speed motor. My two options so far are:

1. 2 HP two-speed motor, single-phase 230 Volt
2. 2 HP single-speed motor, three-phase 230 Volt AND a adjustable speed motor drive (which will accept a single-phase input and output three-phase power)

Option #1 above, the two-speed motor, will output 2 HP at the high speed and only 1/3 HP at the low speed. My dilemma is that I'm not sure whether, after being primed by the high speed, the second-story roof-mounted solar system will run at a system horsepower of only 1/3 HP. The pump curve shows that the 1/3 HP low speed has a max head of only 20 feet (or about 8.6 psi).

In trying to figure this stuff out, I've come to the conclusion that I'm not exactly sure how a roof-mounted solar system affects the system head. Does a solar system, when the solar valve switches to solar mode, add the height of the solar system (i.e. around 18-20 feet) to the system head (which would seemingly max out the available head in low speed mode)? Or does the height (head) of a solar system not matter after priming because the head used in getting the water up to the solar panels is regained by siphon effect of the water coming back down to ground level?

I have read the excellent pool solar heating website at www.powermat.com, but even this leaves me confused because they say in one spot that:
All the pool pump has to do to activate solar is raise the water to the solar panels. From there the water falls back to the pool by gravity. We don't try to make a closed loop where the water falling back down pulls the feed water up creating a syphon effect meaning the pump doesn't have to do any work. No No No. That would create negative pressure in the solar collectors and a negative pressure (a vacuum) can collapse pvc plumbing when it gets hot and hot it will get.
And yet the Powermat website SEEMS to show solar heating plumbing that looks like a closed system and use a siphon effect?!?

Can anyone help me sort this out?

Thanks!!

Titanium

#### JasonLion

Platinum Supporter
TFP Expert
Most solar systems are installed as semi-closed systems, with a vacuum breaker valve, so that the static lift only counts against you when you are priming the panels. The pump would need to be on high speed when the solar turns on, but could run the system on low speed once it got going.

Also, keep in mind that TDH calculations depend on flow rate. The TDH is much lower at lower flow rates. The static head to push the water up to the roof when priming the panels remains the same, but the dynamic head goes way down.

#### Titanium

JaonLion,

I was hoping that the physics of the solar system worked as you have indicated. So it sounds like I could use the two-speed pump on HIGH speed for solar system priming, and then switch to LOW speed in a couple of minutes once the priming process is complete?

So it seems that perhaps a low-speed horsepower of only 1/4 or 1/3 HP (and only 20 foot max head) may work after all. :-D

But I bet that the normal skimmer and PoolSkim will not be terribly happy at those low flow rates at low speed. That's OK since the normal skimmer doesn't clean the surface of the pool very well in the best of circumstances. And I can merely turn the motor to high speed for the 5 minutes or so necessary to "sweep" debris towards the PoolSkim.

Thanks for your help.

Titanium

#### chem geek

TFP Expert
Titanium,

I have a roof-mounted solar system and everything Jason said was correct. I want to add that solar panel systems are more efficient when the flow rates are higher, though there are diminishing returns. The FAFCO solar panels I have recommend a minimum 3 GPM flow rate (75% efficient), have a recommended 4 GPM flow rate (80% efficient), and a maximum 8 GPM flow rate (90% efficient). Such panels are typically connected in parallel and though the head loss through the panel is relatively low (1 foot at 3 GPM, 2 feet at 4 GPM, 8 feet at 8 GPM), the loss through the rather long length of pipe from the pump to the house up to the roof through all the panels at the other end and then back again is quite long. I have a much higher head loss when using the solar system than without and it has nothing to do with the height to the roof, but the total path length for the water. Also, there is a head loss through the solar panels even when just "passing through" the bottom tube of the panel to get to other panels and this loss is more than a straight pipe since it's really like many tiny "T" intersections.

Because I have 12 solar panels and the recommended flow (at 80% efficiency) is 4 GPM, I run at 48 GPM and in my system this is 3000 RPM and 1530 Watts. If I run at the same 48 GPM without going through the solar panels, then this is 2310 RPM and 835 Watts. I have a variable speed pump and you can see the details here. With the solar off, I generally run at 26 GPM which is 1495 RPM and 275 Watts to save energy and achieve around one turnover in the 10 hours that the pump runs each day. My system is not typical as my solar panels extend a great distance across the length of my roof and are not very close to the pump house, but the principle of high flow rates and power consumption when using the solar vs. when not using it will still apply to your system.

Richard

#### tphaggerty

I have 14 roof mounted solar panels, 9 on a 2nd story and 5 on a single story. Same situation as Richard (my system is very similar to yours, panels all the way across about 55 feet of roof and over 130 feet from the pump), I have an Ikeric VS (variable speed) pump (it is a 3-phase unit). I set the pump to kick up to high when the solar valve opens (I run a bit higher than he does, about 70 gpm or so due to more panels) and let the pump run at slow speed otherwise. I use the 110 relay (but not powered) as a switch for the Ikeric to kick it up high, this works really nicely because the relay doesn't cut on until AFTER the valve is fully turned, so there is less pressure on the 3 way valve when it is turning.

(Correctly designed) Solar systems are closed until they turn off. Then, the pull/weight/negative pressure (pick your favorite) opens the vacume relief valve and air is allowed to enter the system. This does 3 really good things: (1) eliminates suction pressure on your expensive panels, (2) allows all that water weight to come off the roof and (3) prevents possible over-pressurization of your expensive panels when they shut off because you pool is already nice and toasty, but the sun is still shining and heats all of the water to really high temperatures.

I don't think it is worth the extra effort to switch from high speed to low speed after the panels are primed - just switch back to low when you are done with the solar, much easier to handle. Most likely, you will want the flow anyway to maximize the efficiency of the panels.

#### mas985

TFP Expert
Chemgeek/Tphaggerty,

Have either one of you tried to run the pump with solar at it's lowest flow/rpm setting? I was curious to know if vacuum release valve would stay shut at a very low flow rates or even what is the lowest flow rate/filter PSI you can run without releasing the vacuum. I suspect, although not certain, that there is a practical lower limit.

#### Titanium

tphaggerty,

I don't think it is worth the extra effort to switch from high speed to low speed after the panels are primed - just switch back to low when you are done with the solar, much easier to handle. Most likely, you will want the flow anyway to maximize the efficiency of the panels
I am think you are correct that the solar might need to stay in the high speed for solar. This inconvenient fact has me looking much more closely at a variable speed pump. I am convinced that the solar would still need quote a bit less flow than the normal high speed, and yet more than the low speed can provide. Hence the need for a variable speed pump that could be dialed in somewhere above 1750 rpm, but still below 3450 rpm.

chem geek,

Because I have 12 solar panels and the recommended flow (at 80% efficiency) is 4 GPM, I run at 48 GPM and in my system this is 3000 RPM and 1530 Watts. If I run at the same 48 GPM without going through the solar panels, then this is 2310 RPM and 835 Watts. I have a variable speed pump and you can see the details here.
I fear that I must admit that I don't have any idea how many solar panels that I have. The only pictures I have seen of the solar panels is from satellite photos, and they don't show very much detail. My yard is such that I can not see the solar panels from my yard, nor from any of my neighbors yard, because of the trees and terrain. I plan to do a roof-top reconnaisance on the solar panels later this year after I have this pool motor replaced and the AquaLogic system up and running.

I assume that you are in the same \$0.30 per kwh (or higher) third tier of PG&E electrical costs like I am. Have you ever calculated the break-even point of BTU gain from solar versus the differential cost of electricity to pump the water up to the second story and through the solar panels?

And it make me wonder whether my limited roof space is more valuable as pool solar heating or as photovoltaic panels to get out of this darned third tier, or at least reduce the amount of third tier power used. My gut feeling is that some of my limited roof space may have a better financial payback as photovoltaic panels as versus additional pool solar heating. I'm going to have to sit down and do some engineering calculations...

ams985,

Have either one of you tried to run the pump with solar at it's lowest flow/rpm setting? I was curious to know if vacuum release valve would stay shut at a very low flow rates or even what is the lowest flow rate/filter PSI you can run without releasing the vacuum. I suspect, although not certain, that there is a practical lower limit.
I was wondering the same thing. Especially after my experience two months ago where I experienced my solar panel vacuum breaker not properly closing after my ststem pressure was adversely affected by a very ditry cartridge filter. And of course, I may be having a vacuum breaker that is on the ragged edge of going bad. I have no idea how old my vacuum breaker is, but the solar system appears to date back to 1983 or so according to the paperwork that I have, unless there was some upgrading in the interim that my records do not show.

Titanium

#### mas985

TFP Expert
I am think you are correct that the solar might need to stay in the high speed for solar. This inconvenient fact has me looking much more closely at a variable speed pump. I am convinced that the solar would still need quote a bit less flow than the normal high speed, and yet more than the low speed can provide. Hence the need for a variable speed pump that could be dialed in somewhere above 1750 rpm, but still below 3450 rpm.
If you are looking at a 2 HP two speed, then you will likely get somewhere between 40-55 GPM at low speed with solar depending on the plumbing. That should be enough. At 40 GPM, you could support 13 panels so if you have less than that, no worries. However, you are likely to have more flow than the worst case. So my guess is you could support up to at least 15 panels.

However, the Intelliflo will cost a lot less @ 50 GPM than low speed of a 2 HP pump. Not sure about an add on VFD.

Whisperflo 2 HP @ low ~50 GPM is about 60 GPM/kw
Intellifo @ 1500 RPM ~ 50 GPM is about 150 GPM/kw or about 60% less costly to run.

#### chem geek

TFP Expert
mas985 said:
Chemgeek/Tphaggerty,

Have either one of you tried to run the pump with solar at it's lowest flow/rpm setting? I was curious to know if vacuum release valve would stay shut at a very low flow rates or even what is the lowest flow rate/filter PSI you can run without releasing the vacuum. I suspect, although not certain, that there is a practical lower limit.
Nope. Never tried that. However, my guess is that the pressure relief valve will only be open during negative pressure. If the pressure is near neutral, then I suspect it will close or perhaps it closes at slightly positive pressure. Otherwise, during the filling up of water the valve might stay open though at some point the pressure would build up enough to close it. There is always a lot of bubbling when I turn on the solar after it's been off and had time to drain so it seems most of the air comes out of the returns and not out of the pressure relief valve.

Richard

#### chem geek

TFP Expert
Titanium said:
I assume that you are in the same \$0.30 per kwh (or higher) third tier of PG&E electrical costs like I am. Have you ever calculated the break-even point of BTU gain from solar versus the differential cost of electricity to pump the water up to the second story and through the solar panels?

And it make me wonder whether my limited roof space is more valuable as pool solar heating or as photovoltaic panels to get out of this darned third tier, or at least reduce the amount of third tier power used. My gut feeling is that some of my limited roof space may have a better financial payback as photovoltaic panels as versus additional pool solar heating. I'm going to have to sit down and do some engineering calculations...
Yup, I'm in that same nasty 32 cents per KWh tier. And yes, I calculated the cost differential you are talking about and it's not even close. Remember that we try and keep our pool at 88F so that's a big factor in the calculations. At lower temperatures, solar isn't as needed and even a solar cover would probably work, but neither my wife nor I are swimming in 80-85F water and we like the convenience of our automatic opaque safety cover (which isn't clear so doesn't add much to heat, but does retain heat that would normally be lost from evaporation).

During the summer, the solar panels increase the water temperature at a rate of around 0.5 to 1.0F per hour compared to the 200,000 BTU/hr output (250,000 BTU/hr input -- yes, it's inefficient) gas heater which can increase the temperature at a rate of 1.5F per hour. So the solar probably puts out around 100,000 BTU/hr during the peak sun time of 11AM to 2 PM and continues to heat outside that time, but at a somewhat lower rate. It takes about 1500 Watts to run the solar and about 275 Watts to run the pump with the solar off. That's a difference of 1225 Watts which is equivalent in energy to 4180 BTU/hour. So from a pure energy point of view, 100,000 BTU/hr of solar heating beats the pants off of 4180 BTU/hr of increased electricity usage. Now to be fair, the gas costs \$1.39 per Therm which is 100,000 BTU while the electricity costs 33 cents per KHw so the equivalent cost of gas for the heating would be (at 80% efficiency of my gas heater) \$1.74 per hour while the incremental cost of running the pump to drive the solar is \$0.40 per hour. The reason this is so much closer in cost (yet still definitely worth using the solar) is that the cost of natural gas is so much less than the equivalent cost of electricity -- it's equivalent to around 5 cents per KWh.

As for PV vs. solar, this has been talked about in other threads and its getting closer to even (ignoring installation costs) if you get a high COP for the heat pump. When PV panels get more efficient, then it will be less expensive to generate electricity and use a heat pump. You'll need to run the calculations for your particular PV based on its KWh per area. It's the startup (capital) costs that will kill you. If you factor those in, then the solar water panels will be far less expensive. Nevertheless, if I had more roof space, I'd probably dedicate some for PV. If I had to do it all over again, I'd probably also go for the more expensive Gobi panels since we like our water hotter and currently need to use the gas heater at the start and end of the season (like now, for instance) and its the breeze and cooler air temperatures that hurt us this time of year (a Gobi panel is insulated to be like a greenhouse).

Richard

#### mas985

TFP Expert
chem geek said:
Nope. Never tried that. However, my guess is that the pressure relief valve will only be open during negative pressure. If the pressure is near neutral, then I suspect it will close or perhaps it closes at slightly positive pressure. Otherwise, during the filling up of water the valve might stay open though at some point the pressure would build up enough to close it. There is always a lot of bubbling when I turn on the solar after it's been off and had time to drain so it seems most of the air comes out of the returns and not out of the pressure relief valve.

Richard
Well here is why I ask the question.

Once the pipes are primed, there is a net static head loss of 0. However, that just means that at the pad, the static head is net zero but at the top of the panels there is still a pressure change. So if one were to measure pressure at the pad versus at the panels, the panels should be at a much lower PSI than the pad because of the height. If there is not enough PSI at the pad, the PSI at the top of the panel could be < 0 relative to air and the vacuum would break. So here is an example:

2 story house 23' to top of panel.

Solar Entrance Pipe Pressure @ Pad = 14 PSI
Top of Panel Pressure = 14 PSI - 1 PSI (Pipe Dynamic Head Loss) - 11 PSI (Static Head Loss) = 2 PSI
Solar Exit Pipe Pressure @ Pad = 2 PSI - 1 PSI (Pipe Dynamic Head Loss) + 11 PSI (Static Head Gain) = 12 PSI
(Note: I am assuming little or no head loss in the panels themselves)

So net wise, there is no head loss due to the hieght of the panels and at the top of the panels, the pressure is still > 0. However, if a lower speed is used and the pump does not provide enough pressure at the pad, then there could be a vacuum at the top of the panels. So for example, lets say the flow rate is reduced such that the filter pressure is only 10 PSI, then the pressure at the top of the panels is -2 PSI and the vacuum release valve will open.

This effect has been seen by other where the filter gets dirty and the pressure drop across the filter is high enough that the entrance pressure to the panels is not high enough create a high enough pressure at the top of the panels to keep the valve closed.

Given that you have a variable speed pump, my hope was to quantify this effect somewhat and see where that limit actually is. If you are so inclinded, you could try lower speeds and see where the valve will release and record the speed, GPM and filter pressure. Not that you don't have anything better to do :-D .

Either way, this may be a moot point since you need a certain flow rate for the panels to be efficient which should generate enough pressure. However, it would be useful to fully understand what the limits are.

#### chem geek

TFP Expert
mas985 said:
Given that you have a variable speed pump, my hope was to quantify this effect somewhat and see where that limit actually is. If you are so inclinded, you could try lower speeds and see where the valve will release and record the speed, GPM and filter pressure. Not that you don't have anything better to do :-D .
OK, I understand now. It's fortunately sunny today and the solar is on right now so I can try lowering the pump speed and see what happens, but how will I know when the valve releases? Is this when water stops flowing out of the returns? Is there any other more obvious sign I can look at to see when this happens?

Richard

#### chem geek

TFP Expert
Low Speed for Solar Panels

Mark,

My normal solar panel flow is 48 GPM and shows 24 PSI at 1575 W and 3040 RPM. I lowered this to the lowest settable and measurable flow on the pump which is 15 GPM and this showed 6 PSI at 257 W and 1540 RPM. I then manually set lower RPM but could not know the flow though I felt the returns to see if there was flow. At 1400 RPM this showed 4.5 PSI, at 1200 RPM this showed 4 PSI and both of these settings still had a small amount of flow from the returns. At 1000 RPM and 3 PSI, however, I felt no flow from the returns so somewhere between 1000 and 1200 RPM and 3-4 PSI the pump is just maintaining static head -- keeping the water elevated in the solar panels but not enough to flow out of them. If the valve opened, then the water in the panels would have drained into the pool. I figure that the height of the pipe water column feeding to the roof is around 12 feet or so which would represent about 5.2 PSI. My pressure gauge may not be very accurate at the lower pressure readings -- remember that I had strange results that didn't track theory very well at the lower pressures. So it's quite possible the true pressure PSI reading was higher. Also, the pressure gauge is at the top of the filter so a few feet above the ground, so perhaps the height between it and the top of the pipe column to the roof is around 9 feet so 3.9 PSI which is near what I measured (maybe my gauge isn't so bad after all).

Anyway, the bottom line appears to be that one can run the solar panels at a very low 15 GPM flow rate or even lower with no problem. The valve does not appear to open at these relatively lower pressures which might only be a few feet of head. The water in the panels would likely get very hot and be far less efficient as they would lose some of their absorbed heat through convection back to the air. I didn't notice that, but I didn't wait that long between measurements.

Richard