I always thought that throttling a pump would hurt it as well. I have found this not to be the case. Throttling a pump to low flow will cause more thrust load, and radial deflection of the impeller. The thrust load has already been discussed. The pump just thinks it is in a deeper well, and the thrust bearing is designed to handle these loads without problems. There has never been a thrust bearing destroyed by a CSV in 15 years.
Radial deflection of the impeller is also not a problem. With an end suction centrifugal pump, the impeller sticks out on the end of the shaft, and is supported by the radial bearing in the motor and the pump shaft. As long as the bearing can handle the load, and the shaft is strong enough not to flex, radial deflection is not a problem. Only pumps with cheap bearings and thin shafts would have a problem, and I have yet to see one that wouldn't last a long time running this way.
With a submersible or turbine pump, there is a bushing on top and bottom of each impeller. This holds the impeller true, even if it does have some deflection. So radial deflection will never cause a problem with a sub.
Cavitation happens at high flow rates, when the NPSHA is less than the NPSHR. At low flow it is more recirculation than cavitation. Recirculation can cause the same kind of wormhole looking damage as cavitation but, is usually in a different area of the impeller. Impellers made with a higher tensile strength material are resistant to this type wear. So a pump would have to be very cheaply constructed for this to ever cause wear. Again, we have never seen a single pump with this problem.
Pump companies will tell you that low flow will cause a problem but, most of them have never tested a pump this way for extended times, and they are just guessing. As most pump installers, I learned these things the hard way, trying them for myself. I have run 10 GPM pumps with a 1 GPM Dole valves, because the wells only made 1 GPM. When the pump lasted an average of 15 years, two pumps in 30 years, I decided that low flow is not that hard on a pump. I also tested many pumps at really low flow rates for long periods of time, then inspected the pumps to find absolutely no wear. Turns out that pumps can handle really low flow rates much easier than they can handle cycling. Reducing or eliminating cycling has proven to triple or quadruple the life of many pumps, even when running at very low flow rates. So low flow rates are not a problem.
I have tested many pumps, including small subs running at 1 GPM for many many years without a failure. I have also tested 50 HP, 1000 GPM subs running at 5 GPM constantly for a year. Inspections of the impellers and thrust bearing showed absolutely no damage what so ever. Big centrifugal pumps have been tested for long time periods, at even lower flow rates than 5 GPM, without any measurable wear. I find that most pump companies have engineers right out of school, or that have never even seen a pump installed, so they do not have the experience to go against what the book says. Pump installers have been trying these kinds of things for decades, and can tell you what will really work and what doesn't. Pump installers are smarter than the manufacturers. You have had to decide on your own what works and what doesn't over the years, and using valves is no different.
A CSV will hold a constant pressure as long as you are using more than 1 GPM. You do have to wait for the tank to drain from 60 down to 40 so the pressure switch can start the pump. Then the pressure quickly increases to the 50 PSI of the CSV, and will stay at 50 PSI constant as long as a faucet is running. When the faucet is finally closed, the pressure will slowly increase to 60 and the pressure switch will shut off the pump. The CSV holds the pressure at a constant of 50 when the water remains running, but will allow the full use of the draw down in the tank before the pump starts. So even a 20 gallon tank that holds 5 gallons of water, will allow you to use all 5 gallons of water before the pump starts. The pump does not have to come on for a glass of water, to rinse a toothbrush, for the icemaker to fill, or for about three flushes with the 1.6 gallon flushers.
With a variable speed pump that is trying to always maintain 50 PSI, there is no pressure bandwidth to be able to use water from a tank. If the pressure is always at 50 PSI, you never put any water in a tank, or you never get any out. No matter how large a tank you install, the pump must start every time a faucet is cracked open. With a constant pressure system, little tanks are fine, but you need to be able to at least use the small amount of water that is in the tank.
My conscience is what keeps me doing this. After you see all the benefits to using a CSV, your conscience won't let you install another pump without one. Cycling cause nearly every problem with a pump system. Pressure switches melting, capacitors and relays burning up, bladder in tanks going bad, pressure fluctuations, water hammer, and most importantly burning up motors, are all problems caused by cycling. You just have to get use to the fact that throttling is actually good for a pump, and all these problems go away.
About 8 years ago, Goulds had some 100 HP subs on a chicken farm, that were cycling themselves to death on a regular basis. After having to warranty the pumps several times they installed CSV's to prevent the cycling. Even though these pumps can now run as low as 5 GPM, they have yet to have a single failure since installing the CSV's. Here is a little of a letter from the distributor.
April 24, 2003
It is alarming that anyone would be making the statement that Goulds Pumps/ITT will not stand behind their product if it is used with Cycle Stop Valves. As a Goulds Distributor, we have used Cycle Stop Valves on Line Shaft Turbines, Submersible Turbines, 4" Domestic and Centrifugals of all sizes and of different horsepower. The fact of the matter is Goulds has recommended customers to Global Pumps & Equipment in the past for the installation of large Cycle Stop Valves on systems that continued to experience premature failure due to system cycling. Example: 3 - 100 HP Submersibles on a chicken farm in South Texas. 3 - 5" valves were installed just over 3 years ago, no problems since.
We have had zero problems with any system that is properly sized and installed using your valve. Neither Goulds, nor any manufacturer, has ever said they would not stand behind their product if used with Cycle Stop Valves. Anything else is BS.
General Manager / VP
Global Pumps & Equipment, Ltd.
Sometimes a recirculating line is used from the discharge side of the pump to the suction. Especially with hot water, a certain amount of flow, recirculating or going downstream, is needed to maintain the cooling for the pump and seal. Without some water being let out downstream, the recirculating line will continue recirculating the same water, until a critical temperature is reached and the pump will fail. Recirculating lines are usually used to maintain a certain flow for cooling, for a short amount of time, when the discharge from the pump is temporarily lessened or shut off.
Recirculation within the pump itself is a different thing. There is some recirculating form the discharge of the impeller back to the eye of the impeller. Some flow slips between the impeller and the wear ring, (high pressure side to low pressure side) and re-enters the eye of the impeller. This is dependant on the clearance between the impeller and the wear ring. This clearance becomes larger, and recirculation increases, as the impeller and the wear ring wear down over time.
There is also recirculation within the impeller. As the discharge flow is lessened, flow within the impeller also lessens. When there is very little flow through the impeller, some of the fluid will just bounce around in certain pockets and eddies, and stay inside the impeller.
Both of these kinds of recirculation will cause cavitation like wear on the impeller. The wormhole effect you see, will be in different areas of the impeller, depending on the type of recirculation.
Sufficient flow is usually the best way to reduce recirculation. When pumping cool water, very little flow is required to prevent excessive recirculation. You must maintain enough flow so that the temperature of the water does not increase very much. The higher the temperature, the worse the cavitation like wear. Except for some really large pumps, recirculation at fairly low flow rates, are usually not much of a problem.
Large pumps or small, if recirculation is a problem, impellers of higher tensile strength will resist the wear better.