two-speed with solar panels

[cleancloths]

Well you should only need the head to initially lift the water to the roof. Once you have flow established the height of the roof has no bearing as you effectively are only lifting to the surface of the pool. You create a siphon pulling down from the roof which cancels out the lift.

As for time of day pricing you will probably find that if you put a cover on the pool at night you will have no need for the solar panels in the peak summer months and will be able to run the pump at night as you have done in the past.

[tphaggerty]

Why 2 speed in your case? Why not just wire in the pump so that it kicks on for solar demand, then kicks off when no-demand. You are probably going to want the full 3/4 hp to run solar anyway.

My solar controller has a relay for a booster (110V only, unfortunately) that could do easily handle the 3/4 hp pump (its a Goldline). If your pump is 220v, you could use that relay to trip a 220v relay to run the pump, or get a controller that has a built in 220v relay (I'm pretty sure some of the more expensive pumps do this).

Or, even cheaper, just set your pump timer to run for the 4 or 5 peak sun hours and then at night. No need for a relay!

[semenzato]

Thanks tphaggerty, good suggestions. I already run the pump with the basic Intermatic timer for the 5 peak sun hours---and that's enough. It's an indoor pool, it doesn't get much dirt.

With time-of-day metering, electricity at peak sun hours costs about 50c/kWh. The pump probably uses the full 3/4 hp, which is about 600W, so that's 3kWh/day and $45/month. There is enough sun here (Oakland, CA) to take advantage of solar heat May, June, July, August, September. You're right, maybe I shouldn't bother, but still. According to the literature, I can get 1/2 the flow for 1/4 of the power, so I would save around $150 each summer, plus about 1/2 of that in the winter.

Cleancloths: thanks for the comments. I thought of the siphon effect too. but then I wondered. There is a valve at the top of the panels that lets air in when the flow stops. I believe this is done so that the panels do not get scrunched by the vacuum produced by gravity pulling down the water. I would think this means that the water must still be under pressure when it gets to the top. Siphoning implies negative pressure, but the valve would not allow that (unless it allows SOME negative pressure---but can it be that smart?)

[cleancloths]

I don't know how the valve works so I cannot tell you. I don't think "scrunching" would be an issue, as they would be filled with water as would the pipe feeding them when the pump stops.

Why did you switch to time of day? What is the savings off-peak that makes it worthwhile? I have peak / off-peak here in NJ as well put the rates are something like 18 cents and 9 cents versus a flat rate of around 15 cents.

I did it because I have a 10KW PV array on my roof which generates about 12,000 kwh a year almost all during peak times. I always used to run my pump at night as I have a heatpump to heat the pool so it made sense. But after much thought I am now running the pump during the day as the heatpump is much more efficient when it is 80 degrees out then when it is 30-40 degrees at night. Also, I added a pair of 2' by 20' panels to supplement the heatpump so I need to run during the day for that too.

[mas985]

The solar valves release when there is a vacuum in the line without any pump pressure. However, what cleancloths was describing is simply negative head not a vacuum. The pipe will still be pressurized throughout the whole run. Head is added with a pipe rise but then is subtracted with a pipe fall. Net net is zero head gain due to the fact they are on the roof (the panels and pipe runs will add head). You just need the high speed to fill the pipes but after that, low speed is sufficient. So after priming, the total head loss for solar is the same if they were on the roof or on the ground, assuming the same lengths of pipe.

[semenzato]

Is there really a difference between "negative head" and "vacuum"? As I see it, unless there is still positive head at the top of the panels, the water on its way down is pulled by gravity and produces a "suction" at the top. By suction I mean a pressure lower than atmospheric. If the system is sealed, that negative pressure "sucks" the water up from the inlet. But if there is a valve, the suction will make the valve open. As I said, it's possible that the valve will stay closed with a small amount of suction, but eventually it will open, and I would imagine the margin to be too small to be of use.

Thanks!
Luigi

[mas985]

Is there really a difference between "negative head" and "vacuum"? As I see it, unless there is still positive head at the top of the panels, the water on its way down is pulled by gravity and produces a "suction" at the top. By suction I mean a pressure lower than atmospheric. If the system is sealed, that negative pressure "sucks" the water up from the inlet. But if there is a valve, the suction will make the valve open. As I said, it's possible that the valve will stay closed with a small amount of suction, but eventually it will open, and I would imagine the margin to be too small to be of use.

Thanks!
Luigi

Yes there is a difference between negative head and a vacuum. The primary difference is the reference. Vacuum usually refers to pressure below atmospheric or negative PSI. Your solar valve on the roof will not release until the pressure in the pipes is below atmospheric which requires the pump to be off and water to be draining out of the pipes. This causes suction.

However, with the pump on, there is a net positive pressure within the pipes usually between 10-20 PSI above atmospheric pressure (i.e. your filter PSI reading). This keeps the valve closed. So while the water going down the pipe offsets the water going up the pipe, this is called static head, there is still a substantial positive pressure to keep that valve closed. There just isn't any head added to the system because of elevation. The only pressure difference is the dynamic head loss experienced in pipe and solar panels. The water falling simply makes it such that less pressure is needed to push the water up the pipes.

The easiest way to see this is when the solar is first turned on. You will see the PSI rise substantially and then after the pipes are filled, the PSI will drop slightly but still higher than without solar.

So in formula terms:

Filter PSI Solar on = PSI Solar off + PSI elevation Rise - PSI elevation fall + PSI Dynamic Head Loss Panels and Pipe

PSI elevation Rise - PSI elevation fall nets to 0 so there is no PSI impact due to elevation.

[semenzato]

Oh, I see what you mean. The "suction" from the falling water reduces the pressure at the top, but not quite enough to make it negative (that is, below atmospheric). The term "negative head" confused me. I would think of it more in terms of "head reduction".

However, I am still not sure this works. You're saying that I would use the pump at high speed to fill the pipe, then I can switch to low speed. But you also agree that even with low speed there is higher-than-atmospheric pressure in the pipe at its highest point (where the valve is located). That means that at low speed the pump still has enough power to fill the pipe, and I don't need two speeds, just the low speed.

Anyway, thanks for your help and advice. I will have to do some measurements and calculations and see if I can get by with a smaller pump.

[mas985]

Oh, I see what you mean. The "suction" from the falling water reduces the pressure at the top, but not quite enough to make it negative (that is, below atmospheric). The term "negative head" confused me. I would think of it more in terms of "head reduction".

However, I am still not sure this works. You're saying that I would use the pump at high speed to fill the pipe, then I can switch to low speed. But you also agree that even with low speed there is higher-than-atmospheric pressure in the pipe at its highest point (where the valve is located). That means that at low speed the pump still has enough power to fill the pipe, and I don't need two speeds, just the low speed.

Anyway, thanks for your help and advice. I will have to do some measurements and calculations and see if I can get by with a smaller pump.

Not quite. Static head, which is basically the rise in elevation, does not depend upon the rate of water flow (GPM). Dynamic head (pipe friction loss) does depend on the rate of water flow. A pump at high speed has a pump head curve such that it can take care of both the static and dynamic head together (i.e. high head curve) quite easily. See here:

http://www.poolplaza.com/WhisperFlo-techspecs.shtml

The pump at low speed, which is basically half speed and half flow, only has enough power to handle the lower dynamic head (notice the head curves are much lower). So at low speed, the flow rate is 1/2 of the high speed so the dynamic head is 1/4 of that at high speed. Since the pipe is primed with full speed, at half speed there is no longer any static head so it will have no problem with the head loss that remains.

Here is what would happen in a typical pool.

No Solar with full speed filter PSI = 15, head loss about 45 feet
With Solar at full speed filter PSI = 20, head loss about 52 feet
With Solar at low speed filter PSI = 5, head loss about 13 feet

To ease your mind, people with pools and two speed pumps have been operating on low speed with solar for years so no worries.


Here are a few more links you can learn about pumps and hydraulics.

http://www.powermat.com/pools/index.html
http://poolplaza.com/pool-pump-sizing-2.shtml
http://www.zoeller.com/zep/techbrief/JF1article.htm
http://www.eere.energy.gov/industry/.../pumpintro.pdf

and I have much more technical version when you are ready.

[semenzato]

I think I have a valve because after the pump stops I can hear water coming down from the panels for a minute or two. When the pump restarts, for the first couple of minutes only air flows down from the panels and into the pool through the jets---and there is no doubt about that! The explanation of why the valve is needed also makes sense, so I have never questioned this.

I switched to time-of-use because that was required by our power utility (PG&E) in order to sell them energy. We just installed 30 Sharp 170W PV panels (which by the way appear to produce a peak of 4300W, according to the inverter).

I like the cost of electric energy in NJ! But at that price I don't think I would have installed the panels.