Trying to keep up w/ Chem Geek

[chem geek]

Yes, putting the panels in parallel will let you move more water through and that will provide more heat, but as Carl mentioned the plumbing is more complicated (the diagram I drew in this post shows a parallel configuration that equalizes the flow to be the same for all panels). It won't be 3 times as much for a variety of reasons, but it will be more. You can never get more heat from these panels than the amount of solar radiation that hits them. There will also be a limit as to how much water you can move since it is recommended not to exceed a flow rate of 10 feet/second in pipe and with 2" pipes that's around 100 GPM, but you probably aren't at that point and any excess can go to a bypass to your pool. I also don't know what sort of pressure these panels can take without leaking, either, but as Carl says you can try configurations to maximize heating until you reach a limit or problems occur. By having a valve to allow for a variable bypass, you can tweak the flow rate through the panels for maximum heating without getting leaking or excessive flow rates.

If your 4x10 and 4x20 panels are similar to the FAFCO ones I linked earlier (or to this link which is for 4x10 and 4x20 panels designed for above ground pools so may be more similar to what you are using), then they have a maximum recommended flow rate of 8 GPM per panel and a maximum pressure of 30 PSI (intermittent up to 45 PSI). With your 3 panels hooked up in parallel, that would mean 24 GPM maximum recommended flow rate. But again, you can try more but may have them leak at some point. Just remember that at 8 GPM per panel you are already getting 90% of the heat you can possibly get out of the panels so at most a faster flow rate will only get you 10% more heating (this is different than Carl's situation because his solar panels are completely different as decking where the flow rates were not even close to the efficient range with so many panels hooked up in series). Depending on the panel, it is possible that too high a pressure (to get higher GPM flow rate) might cause a panel to burst or break which is why I think you should look at the specs for the specific panel you are using and not exceed the maximum flow rate that they specify or certainly not the maximum pressure they specify.

Richard

[CarlD]

Ok,
But let's add this: If you have 3 panels hooked as in your diagram, ShellyAnn, (in series) you will only be getting 8 gpm AT MOST, and probably less, but the water will be hotter.

Actually, I think your max will be LESS than 8gpm, due to tripled resistance. If it was electricity I'd estimate it at 1/3 (less than 3gpm) but this is water and I don't know it as well. But the pressure WILL give you less flow---let's ball-park it at 6gpm.

As Chem_Geek says, hooked in parallel, the panels will give you 24gpm -- or 4 times what I guess-timate the series flow rate to be. Each gallon you flow from the series must be FOUR TIMES HOTTER than each gallon from the parallel to get the same effect.

But what does 4 times hotter mean? Well, that depends on the difference between the pool water temperature and the panel water discharge temperature. If the parallel panels' water is 1 degree warmer than the pool, then the series must be 4 degrees warmer. If the parallels are 10 degrees warmer, the series must be FORTY DEGREES warmer to have the same effect.

Why? Because a BTU is the amount of heat energy needed to raise 1 pound of water 1 degree. So 4 pounds of water 1 degree warmer than the pool is the same as 1 pound of water 4 degrees warmer. Converting pounds to gallons doesn't change anything but the raw amount of BTUs, but not the ratio. (I can't remember if a gallon of water weighs 8.3 lbs or 9 lbs--).

Oh, it's SO much easier to figure in metric and calories...1 liter of water weighs 1 kilo...
24 liters/minute that is 1 degree (Celsius) warmer than the pool adds 24 calories /min. 6 liters/ min 4 deg warmer adds the same 24 calories/min... so if the difference in temp is 10 degrees for the 24-lpm, then it's adding 240 cal/min. The series must be 40 deg warmer at 6-lpm to add the same 240 cal/min.

For reference, 40 degrees celsius is the difference between freezing (0 in Celsius, 32 in Farenheit) and 105 degrees farenheit--a dangerously high fever!

Sorry, you can't bet on gettin' that from solar panels! Not even in Farenheit!

No, it's far more efficient to run them in parallel.

BTW, Richard, I continue to be dubious of manuf calculations: For Fafco to run at their efficiency levels of even 80%, the panels will still be easy to touch even on a blistering day. But you won't get that if you plumb them in series. I STILL go by the rule-of-thumb that if the panel is hot to the touch, or too hot to touch, your flow rate is far too low.

When the panel is hot to the touch, all that good heat is being radiated back into the air. Unless the pool water is already hot, you are nowhere near your maximum transfer efficiency, either your theoretical maximum or your effective maximum.

On a more positive note, manufacturers like Fafco grossly overestimate the number of panels you need to warm your pool. THEORETICALLY, my panels on my deck are far too small in surface area, being only 1/3 the area of my pool. HAH!

My parents had a 13,000 gallon 18x33 oval AG that only got 6 hours of direct sun per day (525 square feet). Even with a solar cover they were lucky on a GOOD day to get 80 degree water--it was usually 78. As they aged that became more and more uncomfortable for them.

I added a 4x20 and a 4x10 panel plumbed in serial--120 square feet with 6 hours of sun per day. All summer the pool ran at 84 degrees, far, FAR more comfortable for them. I had them run the panels whenever the water from them was warmer to the touch than the pool. They lay on the ground and the ground got warm (actually, baked), and after the sun passed would continue to transfer heat BACK to the panels. And this was north of New York City, where swim season is only June thru August.

So, given that, even in series you are likely to get very nice results!

[Daggit]

Ok, I'll toss something in here to go more esoteric. While your discussion on heat transfer and flowrates is in the right direction, its for some of the wrong reasons (or reasoning). Time in the box has a lot less to do with it than turbulent flow.

Water (or any fluid) flowing through tubes very slowly will have a laminar flow pattern, which is to say that the water along the walls stays along the walls and the water in the center stays in the center of the tube. There is no cross mixing. This is a very in-efficient way to transfer heat since only the water near the wall heats up much in the tube...the heat doesn't have time to get to the center.

At very high flow rates the flow is fully turbulent, meaning that the is nearly full mixing within the tube and water near the wall at one point gets moved to the center and vice versa. This allows more heat transfer since more water contacts the tube wall and can get to near wall-temperature.

There is a middle zone between laminar and turbulent called the transition zone where there is partial mixing.

We calculate the amount of turbulence with a dimensionless number called the Reynolds number (I won't include the formula, you can google it). Reynolds numbers below about 2000 are laminar, over about 4000 are fully turbulent.

When we design industrial heat exchangers (one of the things I do) we are carefully to match tube size and flow rate to get Reynolds numbers in the turbulent range to be sure of good mixing within the tubes and good heat transfer coefficient. This is likely the most important factor in determining the overall efficiency in your solar panel, and why published efficiency numbers increase with flowrate. Since pressure drop also increases with flowrate, I would guess that the panel builders design them for only as much pressure as is needed to get to a flowrate that is just above optimum Reynolds numbers for heat transfer.

Oh, BTW, wind chill isn't a very good comparison to what's going on in the solar panel. Wind chill is evaporative cooling determined mostly by the difference between wet-bulb and dry-bulb air temperatures on your damp skin, and not simply the air flow removing heat from you body. The phase change from liquid water to vapor takes roughly 1000 Btu per pound of water, which is a lot of cooling energy.

[CarlD]

Great Post, Daggit!

Water (or any fluid) flowing through tubes very slowly will have a laminar flow pattern, which is to say that the water along the walls stays along the walls and the water in the center stays in the center of the tube. There is no cross mixing. This is a very in-efficient way to transfer heat since only the water near the wall heats up much in the tube...the heat doesn't have time to get to the center.

At very high flow rates the flow is fully turbulent, meaning that the is nearly full mixing within the tube and water near the wall at one point gets moved to the center and vice versa. This allows more heat transfer since more water contacts the tube wall and can get to near wall-temperature.

I JUST saw something exactly like this in the latest Cycle World I got this week. Kevin Cameron in his "Top Dead Center" column, talking about engine cooling says nearly the same thing, but not as clearly. I've been taking the position that the greatest flow rate your system can tolerate will give you the most bang for the buck. I think the real limiting factors are the pressure the panels can tolerate without leaking, and the amount of additional flow the pump can sustain before you lose effective water motion from your regular returns. You seem to reach these limits long before others.

In my Fanta-Sea solar deck panels, leaking is a MAJOR issue--I've just purchased a plastic welder so I can fix a panel with a split seam. As it's a special panel cut to fit around the skimmer, I can't use my other spare panels instead--I must fix it.

BTW, I surprised about wind-chill. I didn't think it included evaporating water--just the flow over the surface. I have ridden a motorcycle at 10 to 20 degrees F for hours on end without a windshield when I was young and foolish and it was one of the most unpleasant painful experiences that did NOT require a trip to a doctor! I still remember spending over an hour and a half in a Burger King in Cheraw, SC in 1980, just thawing out...

My point? I certainly wasn't sweating, but that wind DEFINITELY made it feel like 20 below and it was just sucking the heat out of me, despite multiple layers and wind-proofing.

[ShelleyAnn]

Believe it or not...I actually understood what Daggit was saying. I wonder, though, if you plum in a series, this non-turbulent flow is non-existant as the water flows down one side of solar tubes to a 2" pipe, hangs a right, and then flows up the other side... 3 times.

Hate to ask it, but since we're all looking out our windows at feet and feet of snow and have nothing else to do but TALK about pools, which will get me the most hot water, turbulence (series) or flow rate (parallel)? Actually, I'm going to word it like this...which will make my pool warmer faster, as that is ultimately what I'm after.

Shelley

[CarlD]

Move your pool to Puerto Rico! That will get it warmer faster!

The issue of turbulence is unrelated to series or parallel. Plumb in parallel if you can.

[msm859]

Since we are talking about solar heating, has anyone tried heating their pool with a solar system that is smaller, but is enclosed with glazing and insulation to make hotter water?

[chem geek]

Enclosing a solar panel that is already at 80-90% efficiency in a "greenhouse" of glass will not improve the efficiency or generate more heat. Essentially, the water flowing through the panel pulls nearly all of the energy hitting it from the sun and the panel as well as the air near it in the "greenhouse" will remain cool. Yes, I know that Carl doesn't think that my FAFCO panels are that efficient and he may be right, but more on that later -- just hang with me on this assuming that the panels are operating efficiently.

There will be a benefit with regard to cutting down heat LOSSES if the air temperature is cooler than the water temperature, especially if there is wind.

The reason that solar hot water heaters are enclosed in the "greenhouse" style of glass is precisely for the reason of preventing heat LOSS, not for capturing more energy. Hot water is hotter than typical air temperatures so it is important to keep the temperature around the panels as hot as the water. Insulated multi-pane glass would ideally allow most of the sun's energy through but would prevent heat loss by keeping the temperature of the air around the panels to be the same as the water flowing in the panels.

[EDIT] It might make sense to use the "greenhouse" style for heating spa water since that is so much hotter than air (usually). I've never heard of solar used for spa heating and the raw panels are only specified to go up to around 90F though I can get my pool to around 92F if I don't tell the thermostat not to do that. I don't see why a hot-water solar "greenhouse" style panel couldn't be used for spa water though multiple panels would likely be needed since the spa has more volume (around 300-600 gallons) than a hot water heater (around 50-100 gallons) [END-EDIT]

Essentially, you need to look at this as if there is a limited amount of energy from the sun hitting a square area on the Earth so the best way to capture more of that energy is to have a larger area of panels. This assumes that you are already at high efficiency (80%-90%), which the flat "tube" panels have if you are at their recommended flow rates of 4-8 GPM. The "floor tile" types of solar heating, like the one that Carl has, are apparently not operating near peak efficiency since increasing flow rates in his system (by his splitting into two parallel systems) increased the heat dramatically (and probably cooled the tiles down as well).

Carl, I agree with you that the panels in my solar system should be cool to the touch even on a blistering hot day. If they are not, then they are not at 80%-90% efficiency. This summer, I'll go up to the roof to see if the panels are cool. They are connected in parallel, not in series, and right now have a flow rate of around 5.5 GPM so theoretically should be above 80% efficiency. I've never seen "steam" or "hot air wigglies" coming from the panels, but I'll go up and feel them on a hot day just to see how efficient they truly are. Though I wouldn't normally trust a manufacturer's specs, I worked with the FAFCO folks when I was getting an MBA in college and they seemed to have integrity, but then again that was over 23 years ago so there may be new players.

By the way, my diagram was supposed to show PARALLEL panels with all bottoms of the panel piped together and all tops piped together. I'm sorry if my drawing didn't make that clear. The water (as if you are sitting on a water molecule) flows through the bottom pipe, up ONE of the panels, then through the top pipe.

Just for the heck of it, let's calculate the maximum amount of possible solar heating capability based on the energy from the sun. The amount of solar radiation reaching the Earth's surface with the sun directly overhead (so noontime summer in northern latitudes) is around 1000 Watts per square meter of area on the Earth which is about 93 Watts per square foot (or 317 BTU/hour per square foot). You simply won't get more energy than that no matter what you do. Interestingly, the FAFCO documentation at this link claims over 1000 BTU per square foot, but that isn't BTU/hour so is equivalent to around 3-4 equivalent "peak" hours (in Florida, by the way) so they are quoting a daily BTU rate. Let's see what happens if we assume the manufacturer FAFCO is right about 80% efficiency at 4 GPM. They essentially are claiming that they are heating my pool water at 800 Watts per square meter. My 12 panels have a mixture of sizes, but the total area is 36.74 square meters (effective area). So in theory, they claim that I am getting, at noontime peak during the summer, 29,400 Watts (almost 30 killowatts) which is about 100,000 BTU/hour (about half of the 200,000 BTU/hour of my gas heater's output) and is also equivalent to 25.3 million calories per hour. 1 calorie raises 1 ml of water 1 degree Celsius. My pool is 16,000 gallons or 60.6 million millilters so that means I should expect a peak temperature increase of 0.4 degrees Celsius per hour or 0.75 degrees Fahrenheit per hour.

I have measured the temperature rise in my pool with the solar on and I would say that APPROXIMATELY this peak temperature rise is about right. It is not, of course, a super-accurate measurement, but I would usually see about a 3 degree increase over 4 hours near noontime in June. The pool loses about 2 degrees overnight in the summer (remember that we try to keep it at 88F and have an opaque cover) and the solar typically clicks on around 10 AM (the pump starts at 9 so the solar turns on when the solar panels are warmer than the pool water) and by 1 PM or 2 PM at the latest, the pool is back to its lovely 88F. So I'm pretty certain that the panels are at least over 50% efficiency if not close to their 80%. My actual flow is more than the 4 GPM (it's closer to 5.5 GPM) so my efficiency should be higher (about 84%), but I've got the piping problem I mentioned in an earlier post that is likely causing "low flow" to three of the panels. And you are right that I should shoot for 8 GPM and 90% efficiency if I want to heat my pool a little faster -- I just need to balance that with the tradeoff of pump electricity costs.

Richard

[CarlD]

You may want to check with the manufacturer to see if this is recommended or will void the warranty. It seems to me that most enclosed systems I've seen are in permanent roof-mount systems for household hot water--the "greenhouse" may well prevent freezing. But that's just a guess.

Usually, since it's for summer swimming, it's not an issue. Even if my panels can keep my water warm when it's 50 degrees and windy out, I'm not going in! (unless I have to repair a leak--been there, done that...brrrrr! THAT will convince you that a wetsuit is great to have around!)

[chem geek]

Regarding the laminar vs. turbulant flow discussion, I did some calculations on that a while ago when I started looking at my piping situation and I thought it strange that the FAFCO solar panels specification showed a parabolic curve for the head loss vs. flow rate since that implies turbulant flow. I counted around 200 tubes per panels with spacing around 1/4" so at the desired 4 GPM that would be 4/200=0.02 GPM per tube. Even if the inner diameter of each tube were as low as 0.1", the Reynolds number would be 632 which should be laminar flow (which has a Reynolds number less than 2000).

I also ran some calculations to predict head loss and could not get their results. I wrote to them about this asking them if they actually measured the head loss or if they calculated it. They wrote back (which gives them points in my book -- many vendors don't even respond to questions) and said that

The head loss was physically measured on a sample number of collectors matched against theoretical calculations. The contributors to the overall head loss include the header pipe, small tubes, and metering plenum that evenly distributes flow to the small tubes. It appears the metering plenum is the missing component in your calculations.

They were absolutely right. I had not accounted for the metering plenum -- I didn't even know it was there. So their tubes probably aren't that narrow in inner diameter so the flow is most certainly laminar inside those tubes, but getting from the main pipe into these tubes goes through constrictions designed to restrict overall flow rate and to evenly distribute the water and THAT is where the turbulant flow exists and is probably where the bulk of the head loss comes from (thus resulting in the parabolic curve).

Now, interestingly, FAFCO has a dimpled version of their solar panel (see this link) called "Revolution" that causes the water in the small tubes to spiral (slowly). This makes the panel output about 5% more energy due to its higher efficiency at transferring heat from the sun to the water because the water is better "mixed" in the tubes so that it all gets heated thus having a lower temperature difference between the water and the panel itself (remember, as Carl pointed out, that the highest efficiency is achieved when there is a minimal temperature difference since that keeps the panel cooler and minimizes the radiative losses -- a cool, or at least "air temperature" panel is an efficient panel).

Richard