Pump selection, VSD and BLDC motor

[stabmaster]

I've tried to educate myself on what will make the proper pump for my application, and I believe to have figured it out but I can't seem to find what I'm looking for.

I need a very small pump (65-100W) to drive hot (180F) water through domestic hot water solar panels. The panels are a somewhat novel design and for our application they tend to lose their novelty when the speed is not controlled based on delta-T accross the panels, so I have a controller developed for said purpose.

The motor simply circulates water through the panels and drains it back into a storage tank "open loop." It overcomes a 13-26 foot head and must achieve 5-10gpm (at no head).

I've learned that a brushless DC motor is the most efficient and robust motor for this purpose; also it is workable to the lower speeds as compared to ac induction.

I've talked to comapnies who produce these circulating pumps for solar panels and most don't care to believe that i need a VSD for these panels. much of the time it is suggested that a DC pump should work off of an inverter and I should vary the dc potential. I would prefer a VSD; most likely by PWM but only because I believe it is the most efficient with no drawbacks over others for my application.

So i've found that none of these pump mfr's know or else care to tell about the motors of these pumps. I've tried to source out each piece modularly so that if there was a part failure, only one part need be replaced. I only found one pump MFR "Laing" who says that they have a VSD which runs off of 120vac but it's still in development.

So in other words there are 3 known parts: BLDC motor, pump, and VSD that need to be pieced together and it is currently not offered, or still in development for some reason. If anyone has a hint as to where to start looking for this combination, i would be very greatful!

 

[BigInch]

I have a very difficult time believing that you can't do this without a pump, never mind without a VSD. It appears that the addition of a pump would severely decrease the overall efficiency of your system. What is the specific difficulty with using a thermally circulated system without a pump? If it is a closed loop system, perhaps your pipe diameters are too small?

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[BigInch]

P.S. That will be such a small motor that you can probably get a suitable VSD kit at your local electronic/ham radio hobby store and build it yourself.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[ccfowler]

stabmaster,

If I understand your system correctly, you shouldn't need a pump if the storage tank is physically above your solar panels. The natural circulation rate should follow the variations in the solar panel's performance fairly nicely.

If the tank is physically below the panels, you will need a pump. Your description suggests that the tank may be lower.

Since you are trying to heat the water, you will not want to be simply dumping the water back into the tank (open loop). You will want to re-intruduce the water very gently to minimize mixing of stratified hot and cooler water in the tank. (If you don't minimize mixing of the hot and cold water, you will both reduce the effectiveness of the solar panels and compromise the elevated temperatures attained.)

You will want to keep the down-flow pipe filled so that the work done by the pump will be primarily to overcome the frictional flow losses but not the physical height differential. This will greatly reduce the pumping power requirements. Simple circulation is usually one of the most beneficial applications for a variable speed drive.

Suitable pumps for this duty are readily available and quite inexpensive. These are normally used in applications for circulating domestic hot water in hotels, apartment buildlings, office buildings, etc. to keep hot water readily available at remote fixtures. There are several manufacturers (Taco is the first to come to mind). These pumps are usually a "seal-less" design with a "canned rotor" located inside the stator. They operate at normal line pressures and temperatures, but be careful to avoid confusing them with similar pumps used in hydronic heating systems. These pumps have all wetted components that are safe for the exposure to oxygen in the water (usually all bronze, stainless steel, and "plastic"). These pumps are highly tolerant to flow rate variations, and normally operate continuously at constant speed for many years without any attention. Commonly, these pumps include an internal check valve with a vary low cracking pressure. These pumps are usually powered by 120 volt, 60 Hz, single phase power in the US.

Since you refer to open loop flow, I assume that the solar collectors operate at only a slight pressure above atmospheric. If so, I assume that there is probably another pump required to deliver the heated water to its intended use. Assuming this sort of arrangement, I would design the system so that there would be a small collection tank and vent located above the solar collectors. The lower main storage tank would be closed and pressurized only by the head of water above it. I would locate the pump to draw cool water from the bottom of the storage tank and pump it to the solar collector.

Considering the small power requirment involved, I would look carefully at the costs and benefits of an adjustable speed drive vs. a control valve. It wouldn't be hard to imagine drive losses being somewhat comparable to the control valve losses. It is important to consider that, usually, most of the benefits of an adjustable speed drive arise from the work not done by the motor, and the drive's losses are simply much less by comparison. (The cooling requirements of an adjustable speed drive arise from the drive's losses.) With such a small load, the drive's losses may be proportionately more significant. I suspect that convenience, simplicity, reliability, and initial costs may prove to be much more significant than savings in pump work done.

I agree with BigInch. This is a very small drive that you may well be able to build for yourself.

 

[ccfowler]

stabmaster,

Another thought. If the small circulator pumps that I suggested are suitable for your system, you may want to consider just installing several in parallel and then configure your control system to run only however many you need to keep the temperature rise within the desired range. No adjustable speed drive needed, and no significant additional hardware. The internal check valves in the pumps should be sufficient to avoid recirculation through the pumps.