Solar Thermal Heat Capture

What type of system do you have?

Give us a little more info please.

• Model number or home made?
• Capacity?
• What is the medium. Anti freeze? Water? Oil?
• Pump size and specifications. GPM? HP?
• Piping size?
• Piping material?
• Piping color?

There are many...MANY variables that you need to verify, then you can start to calculate your delta T.

Model 6STC Rocket dome Solar rocketdome.net

My inground tank is 200 gallons

water and 1/2 in copper pipe coated with selective surface.

I send 1/2 gallon of water through the 125 lf of copper each minute. ( I want the maximum btu's through the system as possible) That is why I am considering a three speed controller and pump.

Wilo has a pump and when used with the Resol controller, I can specify three different pump speeds. Righr now I sell the unit with a March BR HS pump and a IMC solar eagle controller. I have no control over the water flow unless I use a ball valve and close or open it periodically.

I open it as the delta T increases between the tank temp and the collector temp.

And close it as the difference decreases.

Rock

put a differential switch in the tank and collector........ have a slow pump activate when there's a difference. whenever heat is there you'll collect it. a slower pump has more time to collect than a faster rate.

With all other variables being equal, and ignoring heat added by the pump, greatest heat gain will be at the lowest delta T (fluid T out - fluid T in), as long as the fluid temp is less than that of the collector.

Heat transfer is driven with difference in temperature. Your lowest delta T inlet to outlet will allow the greatest average delta T between the fluid and the surface it contacts. Keeping the surface heated by the sun cooler will also reduce heat losses out of the system.

Additional fluid speed will also improve heat transfer by reducing boundary layer effects/increasing turbulence.

Much of the additional energy needed to pump the fluid faster will end up heating the fluid as well.

.

Efficiency needs to be clarified. If efficiency means heat captured per energy input to pump the fluid, then the highest efficiency will be with the pump off. If efficiency means capturing the most heat for the setup, the highest efficiency will be at the highest pump speed.

It is probably best to leave the ball valve completely open and not throttle at all.

"greatest heat gain will be at the lowest delta T (fluid T out - fluid T in), as long as the fluid temp is less than that of the collector."

"Additional fluid speed will also improve heat transfer by reducing boundary layer effects/increasing turbulence"

I disagree as these statements do not apply to all solar collection systems.

In every solar system as dictated by design, the optimum heat transfer between the collector and the circulating fluid has a well defined flow range which directly relates to pump speed.

The difference in temperature between collector inlet and collector outlet is a direct measure of the quantity of heat collected from the sun and absorbed by the circulating fluid therefore a higher delta T between the inlet and outlet is indicative of higher efficiency.

If the fluid flow rate is too high (dictated by pump speed & impeller size) the difference in temperature between the collector inlet and collector outlet decreases thereby decreasing efficiency of the system.

Increasing turbulence results in cavitation wherein the fluid does not maintain contact with the collector surface(s) thereby decreasing efficiency.

Flow rate outside the optimum range defeats the efficiency of the system whether up or down.

The only way to effectively determine the optimum heat transfer from the collector into the fluid in any solar system is by measuring the difference in temperature between the collector inlet and outlet.

Heat transfer and fluid flow do not come in special varieties for different heat sources. Having a solar collector as the heat source does not alter thermodynamic principals.

.

While greater delta T at the same flow rate indicates more heat gain, the same cannot be generalized to different mass flow rates..

.

For the same incoming solar radiation, a higher flow rate will cause a lower outlet/inlet delta T. On this we agree, right? If you will also agree that this leads to a lower average collector temperature, we should have no problem getting on the same page.

.

When exposed to the same solar radiation, as long as the absorption characteristics are not significantly altered by a temperature change of a few degrees, doesn't a cooler collector indicate to you that more heat is being removed by the fluid?

.

Even without considering the greater heat loss to ambient from a hotter collector, it should be apparent.

.

Turbulent flow and reduced boundary layer effects from increase flow rates will improve heat transfer in most typical cases....it is a long way from there to cavitation hampering heat transfer.

.

The real question is how much energy do you want to use pumping fluid for a given collector. If you keep increasing pump size, soon you will be adding significant amounts (compared to the solar input) of heat from the pump, and as this raises your average collector temp, you run into similar losses as low flow regimes.

.

By the way, the preceding is very simplified, but even as you look closer, the facts only become more apparent. Consider in lower flow regimes, how temperature in the latter part of the flow path approach the collector temp....and how little heat transfer occurs in those parts.

Some of the hang-up may be with the use of the word "pump" when in fact "circulate" is the better term. Since this is a closed system, the only real resistance to flow is the static head from the frictional losses in the flow path. Many solar systems do quite nicely with the use of those tiny Taco circulators for hydronic heating systems. Most consume between 9 and 42 Watts, hardly enough to add significant heat energy to the fluid.

While you ask us to ignore the endpoint collector temperature and possible solar re-radiation, those are among some of the parameters that need to be controlled to maximize the efficiency of the system.

What I don't see being mentioned here is the actual path of the fluid piping across the collector surface, nor the orientation of inlet and outlet ports. I'll leave it for the ME heat transfer experts to prove me wrong, but you get the best overall performance by using counterflow heat exchanger principles to get the best LMTD (Log Mean Temperature Difference). All other things being equal, a parallel run of pipes into headers versus a single serpentine loop is a better choice, as is placing the cooler fluid inlet at the hottest point of the collector surface, the top, with the hotter outlet at the bottom.

These principles led to the use of embossed/expanded aluminum plate collectors where the fluid flowed down through the parallel channels of a 90% wetted collector which greatly reduced the heat transfer losses of the tube to collector bond while providing a much more uniform average temperature across the collector surface.

Fabrication costs are higher, and out of the reach of most DIYers, but what power engineering project doesn't include economic as well as engineering trade-offs?

I am not asking you to ignore collector temp. I only suggested you consider a simplified model. Removing the simplification only strengthens the argument.

.

For the same total heat input, if you wanted to lower the temperature, would you need to remove more heat, or less?

Your question ignores mass and the quality of the heat. Which would provide more heat input, a 1,000kg block of granite at 300ºK that is placed into a liter of 298ºK water, or 100kg of granite at 3000ºK into the same liter of water?

Both blocks have the same number of BTUs in them, but one is going to warm the water and the other is going to boil it away.

Same conundrum with solar collectors, on a full sun day at high noon with no flow the collector surface can easily exceed 400ºK and you'll be pushing plenty of energy into the fluid, while on a cloudy day the collector surface could be 2º above the inlet water and very little energy will be transferred even if you 200 times the surface area of the first scenario.

Solar collectors are a variable entropy engines, and entropy governs the availability of heat to do work.

Bad analogy. Stay on target here. We don't have different heat sources here, only different mass rates of coolant flow.

Simplified somewhat, it goes like this:

For steady temperature, heat in needs to equal heat out.

If you will agree that higher flow rate and thus lower delta T will yield lower collector average and high temps, I'm pretty sure I can talk you through the rest in a way that will make sense.

Do you agree that higher flow rate and thus lower delta T will cause lower average and max collector temps, all other things being equal?.....or is out disagreement before that point?

Here is another approach.

Let's assume the incident solar radiation does not change over a few degree temperature difference and neither do the reflection/absorption characteristics of the solar collector material.

For temperature to be stable, heat in must equal heat out. Heat in is the solar radiation absorbed by the collector. Heat out is the heat transferred to the coolant plus heat lost out of the system to other than the coolant.

Slower flow rate and thus greater delta T will result in higher collector temps. Just for a moment look just at system losses. Heat transfer is driven by temperature difference. A simple but useful common way to describe heat transfer characteristics of a material in terms of power per area per degree delta T across. Thus higher delta T will lead o higher heat transfer for the same material area. Since the collector is at a higher temp in the lower flow regime, losses to ambient via conduction, convection. and radiation will be higher.

.

Since the input energy is constant (the sun shines no brighter on a hot collector than a warm collector), the increased losses to ambient at higher temp, mean at lower flow rates less heat is transferred to the coolant.

RAM: If you can make that cue ball fit into (and out of the monkey - the liter of water) then you, sir, are truly a wizard and I salute you. Please check your units of measure.

Also, I suspect that granite might undergo a phase transition before 3000 °F is reached, and/or that the heat capacity will change too much to allow a heads up comparison.

'....All other things being equal, a parallel run of pipes into headers versus a single serpentine loop is a better choice....'

.

.....which is effectively a higher flow rate and leads to lower outlet to inlet delta T. Your words there make a good point.

I think I am going to disagree with you on the parallel pipes with headers. If I did not have a great amount of turbulence then I might agree, but I can gather more heat even at a lower temp that parallel pipes because of the turbulence. In straight pipes you have the incidence of laminar flow and as a result the possibility for less heat transfer.

Laminar flow is indeed bad for heat transfer. Still, there are lots of ways to insure turbulent flow.

.

Cutting your flow path from one long one into multiple parallel paths could improve the efficiency, but at added cost/complexity.

.

There is at to like in your design, not the least of which is the aesthetic appeal. If it is working well, and can be produced inexpensively, there is no reason to 'fix' what isn't broken.

.

Congratulations on your accomplishment.

.

I would leave the ball valve fully open.

.

I'm not a thermo guy, so this is a real question! When is the greatest heat quantity transported? When a lot of fluid is heated a little (fast flow, small delta T); or a smaller amount of fluid is heated a lot (slow flow, high delta T)?

I agree that cavitation would be undesirable.

Just ask yourself; how do you cool something down quicker....by increasing flow of the coolant or decreasing flow of the coolant? To cool a hot spoon of soup, do you hold your breath or blow on the spoon?

The higher rate of flow will remove more heat, unless something, perhaps severe cavitation begin to hamper heat transfer.

There is no "one-size-fits-all" when it comes to optimizing a solar collector, you'll have to determine what's best for your particular configuration. One thing you have to do is define where you're measuring your delta T. Is it across the inlet and outlet, from the outlet fluid to the surface of the collector, etc., etc.?

Rather than a multispeed pump, you're better off with a differential controller that adjusts the pump speed/flow rate based upon your desired delta T. As you probably know, the conditions of your system are always changing throughout the day, and you'll have to come up with a control strategy to maximize your gains.

Tell us a bit more about your set up and maybe we can guide you, it's not a trivial exercise.

Ram, Excellent reply. I was trying to figure out how to say exactly what you just said. Thank you for reading my mind. There are soant variables that he needs to optimize the controls for his situation.

I don't think it matters that much....It would be interesting to play around to see if you can find a formula that increases performance, but you would have to start with just the single speed for comparison.... I think it would be easier to just add another collector...

http://www.greengaragedetroit.com/index.php?title=Solar_Thermal_Collectors

Adding a collector is always a possibility but most people are stuck with a fixed area they can cover and are looking for ways to maximize the gain from what they have. In most cases, the question of control strategy goes to the very core of how well a solar system will perform since there are three control loops, each with their own (sometimes opposing) requirements.

The first subsystem is the collector where your goal is to collect as much energy as possible, but what does that mean since there are two opposing requirements. If your flowrate is too fast your heat transfer fluid will not get very hot and your storage system will take longer to charge. If your flowrate is too slow then the fluid will get hot too quickly, as will the collector surface, and you will be re-radiating heat from the collector surface back to the atmosphere.

At the storage end (the second loop) the large delta T works to your advantage because the greater the delta T between the fluid and the storage, the sooner your tank heats up; i.e., the better the quality of your heat. If this is a domestic hot water system the third loop isn't a player since the system is demand driven, but it will matter if you don't have a secondary heater to provide an even delivery temperature.

It's the old "I can give you three things, pick two" problem. You can have high temperature, high flowrate, quick heating (time), but not all at the same time. The worst part is that all three are functions of time and a random, but statistically consistent, variable, the solar incident energy on the collector.

Depending upon many variables regarding the collector design, you may find that the "best" pump control strategy is one that measures the insolation (the available incident energy on the collector surface) and uses that in combination with the delta T between inlet and outlet temperature, to vary the speed of the pump. It's basically a "collect it when it's available" and "don't give it back when it's not" strategy.

By knowing when the collector surface is hot you push more fluid through to collect that heat, and less (or none) when there's less useable insolation and/or your tank is approaching it's desired temperature.

I would think that drawing water from the bottom of the storage tank to feed the collectors and depositing the heated water on the top would be the best strategy....Drawing water for usage about 1/4 way down and fresh water being supplied to the suction line at the bottom for the collectors...this taking advantage of the thermal layer to maintain maximum efficiency....

Solar heat capture does not depend on delta T. It depends on Availability of Solar radiation.

The solar radiation is captured and converted into hot water, in case of solar water heater. The efficiency of solar heat capture depends on the design, and for heating water, evacuated tubes are more efficient than conventional copper tubes.

How you utilize this captured heat depends on many factors like insulation quality, circulation of the water/liquid in the system etc

You are mostly right, but collected energy does depend somewhat on delta T (for the very reasons you later mention). At larger delta T where the fluid toward the end of the path is close to the collector surface temp, not much heat transfer occurs.

Also as delta T increases and the high temperature portions of the collector reach greater temps, more heat is lost to ambient through a given insulation, and more heat is radiated away.

You can play likewise with the capacity by making a multiple parallel pump installation--of course automated.

Q = mass flow(spcf.heat of Water) (delta Temp) , you can vary mass as well as delta Temp. It depends on collector size and magnification.

• If there is more energy coming from the collector than it costs to run the pump, then run the pump.

The sizes of the equipment are known locally, so the above criterion can be assessed locally.

I have included a picture this time. I have plenty of turbulence to aid in heat transfer and I am using a copper tubing that has an inside surface area twice the external surface area ( imagine the tardis from Dr. Who). I like all the answers you have given me and appreciate your feedback. I can vary my flow rate with my ball valve but that is problematic as I am not always available nor would I want to be. I am getting a very good throughput to my in-ground tank and can heat it quickly. Yesterday for example I captured about a 25 degree rise in temp in the tank; from 100 F to 125 F for the day. I am checking it again today. I only loose about 4-6 degrees over night in my 200 gallon polyurethane insulated tank and that is quickly recovered the next day. It is 10:30 am and my DT is 9 F. By this afternoon it will be closer to 20 F. I am switching to a Resol controller that will let me control the speed of the pump based on DT and will probably start with 15 F and record the performance and then change to higher and lower temps to see just how much it changes. I have a data recorder attached that will download to Excel where I can determine the best settings over time.

I presently have this at several beta sites and it has performed admirably. But like all products you can never stop making improvements. This Black unit will debut in Las Vegas on July 1 at a first of it's kind store. Via com is a retail solar store with a format like an Apple store and we are being featured in the store with a full sized unit.

I appreciate all the help you have given me and I hope this picture comes through.

Did somebody mention TARDIS?

Wow that's cool looking....that?heats a 200 gal tank to 125°?!?...or are you not using the water?

This is a test unit and we do use the water but only to wash our hands during the day in the shop. I have the same system at home where we use the water regularly and I can keep the temp at about 120 F. There are two of us in the house and we shower daily.

Amazing, "What's old is new again", I remember seeing a very similar design back in the "solar craze days" of the '70s, I'll look through my files for some pictures that I have. You're using internally ribbed tubing to increase the surface area, does it also have a slight twist to increase the turbulence?

Leave the ball valve wide open, the only throttling that you need is the speed of the motor. Since you have a data recorder, your next goal is to experiment with the controller to find a good balance between delta T of the fluid and the temperature rise of the collector surface. You may find that the delta T goal needs to be adjusted on a cycle that tracks the temperature of the collector.

Yes, the tubing has an internal twist. Along with the concho spiral configuration I have plenty of turbulence.

Michael,

I live in Las Vegas. Where is this store?

It is the new Via Sun Store at 4245 S. Grand Canyon Dr # 103 I think it is in Summerville( not sure about this) but has a Vegas address.

If you are in Vegas next week I will be a the Hampton Inn near the store.

Michael C Rock

There is an area called Summerlin, it's on the West side of Vegas.

I'm here all the time. I live on the East side (Near Henderson) and work on the Southwest side. (Blue Diamond and Rainbow)

It's possible that I could run by the store after work next week and take a peek!?!

In the early 80's I made some molds for the solar water heating industry.

If we are talking about the simple quantity of solar heat stored in a water reservoir (underground), then 15-20 degree F rise is probably adequate to do some space heating or provide warm/hot-ish water.

If we are wanting to turn around and make a heat engine, the only dT limit is how much temperature the materials used can sustain, and not be damaged beyond belief, so the baby can sleep without hurtings.

Keep it simple.

Any decent pump will transfer the heat quicker than it's being generated, so it's pretty much a non problem

You don't need fancy controllers, just a simple "Is the water at the collector say 8 degrees (or whatever) hotter than the water in the tank?" If yes run pump.
You need to allow for thermal loss as the water is pumped round and enough delta T at the tank end else there won't be enough differential to transfer the heat.

(I think I used about 8-10 degrees C in my system)

It will self regulate to a great extent. Eventually the water in the tank will end up as hot as the collector one way or another. The thing that changes will be the pump cycle time.

I favour a fairly high delta T with a bit of lag on the pump to stop the pump hunting.

I made the beginners mistake of having my sensor on the wrong end of the collector coil initially, so it was quickly cooled by the incoming water.

One way to look at it is:-

The collector gets hot. How long will it need the pump to run to shift that slug of hot water round into the tank coil?
You don't want to just shift it down the pipework, so a bit of pump over run (hysteresis) is probably a good thing.

I think the simple on off system will be better than a fancy controller on the grounds of cost and reliability. You still have plenty of things to adjust, delta T and pump speed. You can always add some gate valves to twiddle too

But like I said, it will tend to sort itself out anyway.
Del

No, sorry, no one can answer the question as posted. But I can tell you how to determine the answer...

There is a curve of values that ranges from maximum capture at zero degrees delta-T, and lots of fluid flowing in the system. But that heat would be unusable, resulting in 0.0% 'net efficiency' for the system. As the flow decreases the efficiency will increase up to some maximum which varies based upon the parameters collector and what the heat will be used for. Then the curve declined towards zero again as the temperature in the collector rises but the flow is diminished such that very little heat is being brought into the the system where it can be used.

It's similar to the Laffer curve in economic theory. At zero percent tax rate the government received zero revenue no matter how great the GDP; at 100% tax rate no one would produce and again the government would receive no revenue.

Until the use for the heat is established, at which time the desired temperature can be established, *then* you can determine if the overall system would be more efficient on the high side of that temperature or on the low side of that temperature based on where the losses occur. That would be the optimum temperature, then the delta-T would be obtained by subtraction.

How high does flow have to be to enable a 0 delta T with non negligible heat input?

.

How is it that the word 'solar' makes so many forget the fundamentals of thermodynamics? This is not a complicated problem, and yet so many insist on answers that are completely off base.

Shed loads of overthink here.

It is a self regulating dynamic system.

Just think about how the temperature at either end will shift.

Say you have high flow and the transfer liquid doesn'y have time to pick up much heat, but is efficient because it is colder than the collector.

The liquid won't be much hotter than the tank and won't transfer much heat, so it will just flow round again gainig more heat until it is hotter than the tank and it will transfer heat.

The pump will keep running 'cos the collector is hotter than the tank.

But the tank then gets hotter... and so on until the whole lot is at about the same temperature.... Or to be more precise it will keep going off and on until it stabilises with the collector just less than delta T hotter than the tank!

Did I mention KEEP IT SIMPLE?

Del

Couldn't have put it simpler myself.

Rock

Michael,

Did you say you were going to be in Las Vegas this week?

Bryan

yes, I am here now at the Hampton inn in summerlin. Rocket Dome still en route on a Road Runner truck. hint: never use Road Runner freight. They have given me bad delivery dates twice now. Expect it tomorrow. Will be at 4245 S. Grand Canyon drive in store Via Sun on the entrance to shopping center directly across from the Hampton.

my phone is 864-216-3489

Rock

Michael,

No can do today, I have Heaven Can Wait coming to my house to drop off some baby kitten traps. A Momma kitty had her litter of 5 kittens under my house. I've already trapped 3 but the other 2 are giving me the slip. I will try to get there Wednesday after work. About 2:45 pm

Call me before hand. I am not scheduled to be back in the store until Thursday to train installers. If they are there Wednesday we can get in.

864-216-3489

Rock

Posting you number in open forums might have unwanted repercussions.

Thanks for the heads up. I will be more prudent in the future.