Pressure tank installation 1

[ADAG]

Hello,
I have the following situtation regarding the installation of a water pressure tank and I would like some advice:

I have a well with a submersible pump, pumping water to two 1000 gls storage tanks adjacent to the well site. About 1100 feet away from here there is going to be a shed with a pressure tank to supply water to a few surrounding facilities. I am a little confused about what to do to feed this pressure tank. I could install a pump besides the storage tanks but the problem is it is quite far to send the signal from the pressure switch at the pressure tank to the pump at the well site. Then again I could put the pressure switch at the pump outlet at the well site, with the pressure setting for the pressure tank plus the friction losses in the 1100 ft pipe. I am not sure however is that is going to work allright. Any suggestions or comments are most welcome.

Thank you,

 

[itsmoked]

Are we talking level from tanks to the pressure tank?

The pressure switch at the pump at the tanks, will probably work fine. Thing is it may not be overly accurate, but a 10PSI difference from some desired pressure at the pressure tank likely will make little difference anyway.

Remember that the pressure switch at the tanks with the pump will see, (as you've mentioned), a higher pressure than the pressure tank would see,(only when the pump is running). When the pressure switch sees the cut off pressure and stops the pump the pressure will step drop quickly to the pressure tank pressure without the pumping loss(static pressure). If your pressure switch has too little hysteresis, and your pumping line too much resistance, then the this delta change will get you. But if you make the interposing line large enough so the dynamic pressure change is reasonable, and use a switch with separately adjustable on-off settings you will be fine.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Artisi]

I see no problems with fiting your pressure switch at/near the pump discharge. As Keith has pointed out, you might need to play round with the pressure setings to get the required system pressure at the final delivery point/s

 

[Valvecrazy]

There is a problem with a pressure switch at the pump since it is a long way to the pressure tank. When the pump starts, the pressure switch will click on and off rapidly as the pressure at the pump will be very high until the flow gets moving. Even pulsation dampeners don't help much with this problem. I would use a Cycle Stop Valve on the pump and a 1 gallon pressure tank with the pressure switch. The CSV will make the pump start and stop at a very low flow instead of full pump capacity. This will eliminate the pressure shock on the pressure switch and the switch will not bounce. This way you can also use a much smaller than normal pressure tank at the well house. Also the CSV will fill the tanks at a very low flow rate, which will eliminate friction loss between the two tanks, making them fill equally.

 

[ADAG]

Thank you all for your comments,

Itsmoked aswering your question, the storage and pressure tanks are pretty much level, being the latter seated maybe 3 ft higher.

Valvecrazy, the cycle stop valve sounds good to me, and for my pumping capacity (around 30 gpm) not very expensive. Altough I did not understand what is the use of the 1 Gal pressure tank. Perhaps I did not explain myself well:
On the well site I have a submersible well pump feeding two atmospheric 1000 gals storage tanks; this pump will start/stop responding to low/high level signals from the tanks; just beside these tanks will be another pump, with a pressure switch, which will pump water to the pressure tank located 1000+ feet away.

 

[itsmoked]

Thanks for the info.

If there is electricity at the pressure tank/s You could use a radio transmitter between the pump and the pressure tanks. There are a lot of fairly inexpensive units that can essentially pass a switch contact wirelessly. It could cost less than the valve and (local to the pump) small pressure tank solution Valvecrazy has sagely suggested. It gets you essentially what you want. No wires and pressure tank feedback.

Example: First one I saw on a search. Many others may be better. Just showing what I mean.

http://www.globalspec.com/FeaturedP...gy/ShortRange_Wireless_Switch_System_/32549/0

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Valvecrazy]

For only a 30 GPM pump, you can find a CSV a few different places on the web for as little as $70.00. I understood your system requirements, I do these kind all the time. The pressure switch must be with a pressure tank to keep it from chattering. Because the CSV reduces the flow to 1 GPM before the pump shuts off, you can use as little as a 1 gallon size tank at the switch, and both tanks will fill equally. You can also safely use a main pressure tank that is no larger than 20 gallon actual size. This may even fit at the pumps location, which would also solve your problem as well as more than offset the price of the valve.

Otherwise you need to use something like the radio control that itsmoked suggested, to send a signal from the remote pressure tank to a relay at the pump.

 

[itsmoked]

Hey Valvecrazy; Would you please walk us thru the little P tank and the hookup of the CSV? Any restrictions, like to the little tank or the sensor after a restriction? Anything going on like that?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Valvecrazy]

When using the CSV only a 20 gallon size pressure tank is needed. So I would recommend installing the tank as shown in this link under "CSV1 with Jet Pump".

http://www.cyclestopvalves.com/applications_1.html

However, you can still install the 1 gallon tank as shown in this picture and install a larger tank anywhere downstream.

You could install a submersible in the storage tank or a jet pump outside the storage tank. Either way the set up would be pump, CSV, tee with pressure tank and pressure switch off to the side, and other side of tee goes directly to distribution line. The CSV will make the pump start and stop at 1 GPM, which with a tank of any size attached to the pressure switch, will eliminate the chattering of the switch. Once the pump is running, the CSV will vary the flow to exactly match the usage by maintaining a constant pressure. IE; with 40/60 switch, the 50 PSI CSV will maintain 50 PSI as long as between 1 and 30 GPM is being used. When all water outlets are closed, the CSV will fill the tank from 50 to 60 PSI at a 1 GPM rate. Then the switch shuts the pump off at 60 PSI.

Filling the tank at 1 GPM is why the remote tank will fill at the same rate as the tank at the switch, because there is no friction loss at 1 GPM. Starting and stopping the pump at 1 GPM is what keeps the switch from chattering, because there is no water hammer or transients. Varying the flow to match the usage is why a tank as small as 20 gallons size, with only 5 gallons of draw down, is all that is needed.

 

[SteveWag]

I have done this many times with a local pressure sensor and a remote tank, some times as far away as 2-3 miles. If there is little static difference, a time delay on start and stop will let the system stabilize on start or stop. If there is a very large static difference, I install an altitude valve at the tank.

Steve Wagner

 

[itsmoked]

Thanks Valvecrazy.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Valvecrazy]

A time delay on start and stop is also the only way I was able to make remote pressure or elevated tanks work with a pressure switch, before I started using valve control. The time delay just gives time for the transient pressure waves to subside. After seeing a graph of these transients, I realized how devastating they actually were.

The valve and small tank control eliminates the transients and the need for any time delay. The demand is first met by the water from the little pressure tank. This gets the water moving down line before the pump starts. Then the pump starts at 1 GPM against the valve, and the flow makes a smooth transition from the being supplied by the little tank to being supplied by the pump.

On shut down, the demand must stop completely. Then the pressure tank is topped off at 1 GPM before the pump shuts off. No destructive transients, no need for time delay.

 

[ADAG]

Thanks valvecrazy,

With that low 1 GPM flowrate to the pressure tank, am I not going to have my transfer pump running almost constantly?

 

[itsmoked]

Not sure why Valvecrazy has picked that value other than as an example. No doubt you can adjust it up to the highest value that prevents the chattering situation.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Valvecrazy]

Actually the 1 GPM figure is derived from the minimum flow required to keep that size pump from overheating. It is basically two halves of a hole that come together as the valve closes, to create a fixed but non clog-able orifice. Different pumps would create different amounts of differential pressure across the valve. Pumps that have a lower shut off head will push a minimum of ½ a GPM through the valve. Pumps with a high shut off head will push a minimum of 1.5 GPM through the valve. That works out well as the pumps with higher shut off head need more flow for cooling, and pumps with lower shut off head only need ½ GPM to remain cool.

Pumps are made to run. They will last longer if they run continuously than if they cycle off and on, any way you do it. Ideally, every demand should be set for 30 GPM, so the pump runs continuously anytime there is a demand. With the CSV, as long as the demand is more than the ½ to 1.5 GPM (depending on the particular pump) the pump will run continuously instead of cycling.

The counter intuitive part, is that the amperage or power consumption of most pumps will decrease in proportion to the flow rate. In most cases there is little if any difference between the power consumed by a pump that is throttled with a valve, compared to reducing the RPM with a drive. In other words, even though the pump may run continuously, the power use is proportional to the flow required.

Not only will the continuous running increase the life of the pump system but, it virtually eliminates transient pressure waves, as they mostly happen from starting and stopping the pump.

When there is no demand or less than the valves minimum flow, the tank will fill and the pump will shut off.

As described earlier, when the pump does start or stop, the 1 GPM minimum through the valve in conjunction with a small pressure tank, also eliminates transients which will bounce the pressure switch as well as cause other problems.

 

[Artisi]

The consideration for minimum flow has to be based on more than the flow necessary to prevent overheating, overheating and minimum recommended flow are two seperate issues and cannot be interchanged with each other.

When considering minimum flow, you must account for internal recirculation, radial and axial loads imposed on the pump shaft which leads to higher bearing loads, deflection of mechanical seal faces, and in some cases shaft failure etc etc.

Also, I am in complete disagreement with you on the blanket statement that "They will last longer if they run continuously than if they cycle off and on, ........"

A pump that is operating below its recommended minimum flow is slowly destroying itself and will suffer a lot more damage and give a lot more trouble than a correctly sized unit which is cycling within the limits of its design.

 

[Valvecrazy]

Minimum flow for most pumps can be much less than most people think. Yeah, yeah, recirculation, radial deflection, bearing loads, mechanical seal, etc. etc. None of this is a problem with most pumps. The pumps that do have a problem is because the manufacturer has used the shaft slenderness ratio to make the smallest diameter shaft he can use and bearings to weak to handle the load. (Cheap pumps) Still most pumps can tolerate these conditions much easier than cycling itself to death. I only have a few hundred thousand pumps working out there to prove the point. Some have been operating this way since 1992 with no problems to date. The biggest problem is that nobody has tried it because it says in the books that it won't work. I tried it thousands of times, it does work, and most of the books are wrong about minimum flow rates.

 

[Artisi]

I think if you like to look at the quality manufactured pumps round the world, the shaft slenderness is not an issue and is usually far more than adequate for the normal range of operation within the acceptable hydraulic range for which they are designed. These pumps will normally have a minimum flow specified by the manufacturer for very good reasons.

Then you are probably talking about "toys" within the pump industry who's shaft size is usually oversized for the hydraulic duty, resulting from the need to be able to assemble with ease and not having to have the skills of a watchmaker.

So back to your blanket statement - what is your recommendation for a 3000kW boiler feed pump minimum flow, or a 2000 kW axial flow cooling water pump.

 

[Valvecrazy]

My "blanket statement" about minium flow is only for centrifugal impeller pumps, moving fairly cool water. Boiler feed pumps at 90C need a certain minimum flow or recirculating line to keep the water from flashing. Axial flow impeller pumps don't normally decrease amps as flow decreases, and therefore don't lend themselves to valve control. For systems such as these, VSD control may be the only viable solution. There are so many negative side effects that the only time VSD should be used, is when it is the only viable solution.

"the shaft slenderness is not an issue and is usually far more than adequate for the normal range of operation within the acceptable hydraulic range for which they are designed."

Acceptable hydraulic range for which they are designed means the shaft and bearings are made to minimal standards. A nickel more worth of shaft and an extra dollar on the bearings and radial deflection would not be an issue. Heavy duty bearings and a stronger shaft would make the pump last longer, and I believe that is the major reason most pump manufacturers don't do it.

Anyway this thread is about a 30 GPM pump moving cool fresh water. That pump can be throttled back to as little as 1 GPM, and can run that way continuously without damage.

 

[Artisi]

Seems that we will have to agree to disagree on the matter regarding severe throttling of a pumps discharge to below its recommended minimum flow.
Regarding shaft design, I must say that the pump companies I have worked for over many years do not/did not design pump shafts to the bare minimum - which happens to cover pumps from a few kW upto many thousands of kW.