Pressure Surge from High Service Pump Help

[sjohns4]

At the local WTP they are having a surge problem from one of the high service pumps. The pump is a 1600 GPM pump with a swing check and soft start drive. The other two pumps are 2400 GPM pumps with soft starts and electric check valves. The problem seems to be that the pump ramps to full speed before the check even opens, and when it does the pressure spikes from ~110 psi to +160 psi. The drive is an Allen-Bradley SMC flex and according to the tech support folks it doesnt matter what the ramp time is set to, the drive will ramp the motor to full speed as soon it can, which should be easy to do when the pump isnt moving water. We tried to current limit the motor from the drive. When it was set at 200% FLA the motor quickly ramped up to partial speed but when the contacts closed at the end of the set ramp time it jumped to full speed. I changed the current limit to 300% and it ramped up as it always did - with the surge pressure when the check opened. I think ideally what the drive needs to do is quickly ramp to a speed maybe 70%-80% of full speed and slowly ramp up to 100%, but the tech support folks say it wont do that.

Does anyone have any suggustions as to what we can do to fix the surge problem without replacing the drive or installing an electric check valve?

Also, there is a surge relif on the high service line leaving the WTP, if it's working shouldnt this prevent surges from entering the system due to a quick pump start?

Thanks,

Mike

 

[cvg]

fix the soft start? the whole point of a soft start is to ramp up the motor slowly

 

[sjohns4]

I agree, it makes sence to me if you set a 30s ramp time the motor would accelerate from 0-100% linerally across that time, but that's not what the folks at Allen-Bradley (rockwell automation) are telling me can happen with this starter. They say no matter what the ramp time is set to, if the drive can get the motor to full speed sooner it will.

 

[semo]

I had a similar situation many years ago where the soft start drive would ramp to about 80%, then go to full speed. The last 20% would send a hammer through the system that rebounded and increased in intensity. Lots of water heaters blew up. We tried all kinds of settings to no avail. The soft start is not a VFD and won't ramp 100%.

We installed a pressure relief valve between the discharge and the suction to solve our problem. It would open and cut the top out of our spike.

There are a lot of variables involved in your situation as you have 3 pumps of varying flows. I don't know if your problem occurs with the pump when it is alone, when it is with 1 other pump, or when it is with 2 other pumps.

Either way, you need to make sure your pressure relief valve is operating. Then the tricky part is to find the correct setting. If you set it for a pressure with the 1600 gpm pump, your 2400 gpm pump will probably cause it to remain open when that pump is operating or when multiple pumps are operating.

You say the problem is with 1 pump. What about the other 2? Does the same thing occur? If not, what is different about them?

Some other thoughts. Is the swing check sticking? Is it spring and lever or weight and lever? If so, loosen it up, take weight or tension off.

 

[sjohns4]

The pumps are only ran one at a time, the two large pumps have electric check valves so the motor starts with the check valve closed then the check automatically opens slowly and creates no noticable pressure spike (at least within the accuracy of the old chart recorder). I agree I think testing / checking the pressure relif valve is in order. The check on the pump is a cushioned swing check...maybe adjusting with weights will make a difference?

Thanks,

Mike

 

[chief]

Fit a jockey pump and check valve to maintain system pressure. ( approx 5hp motor, 1" discharge line).
This eliminates the water hammer and pressure spikes caused by entrapped air in the piping system.
The jockey pump can discharge to a pressure vesssel with either a precharged bladder or air cushion.
This allows the jockey pump to operate on a differential pressure switch.
Safeguard the pressure vessel safety valve by fitting an orifice in the line to the main system. The main pump flow rate is too high for the jockey pump pressure vessel safety valve,should there be a reverse flow .
Fit a high point auto air vent.
Main pump can be configured to start on a pressure drop below the jockey pump differential settings.

Offshore Engineering&Design

 

[hydrae]

Get a surge analysis done,
If you have hammer on startup you may also have hammer on shutdown that can do even more damage but is hidden until something major breaks.

Hydrae

 

[bimr]

"Does anyone have any suggustions as to what we can do to fix the surge problem without replacing the drive or installing an electric check valve?"

Well, you could always have the operator standby and open the discharge valve by hand.

With that size of pump motor, you should look into the use of a VFD.

 

[cvg]

your existing check valve does nothing to prevent surges, it is designed to prevent backflow through the pump which might damage the impeller. An electric "check valve" would actually be a "pump control valve" which would open very slowly (60 - 300 seconds) to prevent surge in your pipeline.

regarding your existing surge relief valve - these things require tuning to operate properly with every system. If you are not experienced with these, call out the manufacturer to inspect and adjust it.

http://www.pump-zone.com/valves/valves/surge-control-in-pumping-stations.html

 

[semo]

I did not mean to imply that the check valve will solve the problem; but, it can contribute to it.

The pump will not produce any flow until it reaches a specific speed. From that point, the flow will increase as will the pressure.

If the check valve is sticking or hard to open (too much weight, too much spring tension) it will require even more flow to open. This then causes the valve to spring open which does affect the hammer.

Eliminating this won't solve the problem; but, it could help.

If you don't want a pump control valve or VFD, you will probably have to resort to a pressure relief valve. Note that you will have to set it for the larger pumps in order for it not to blow off when they are operational. This will then cause a small spike when the smaller pump turns on.

 

[sjohns4]

All good points. We were trying to not buy a VFD just because the soft start is only a couple years old. The normal mode of operation now is for the operator to manually open the butterfly valve once the pump is already running. I'm of the oppnion that reguardless what we do with the pump, the surge relif needs tested by the manufacturer since no one can remember when it was last done. Guess we'll be getting prices for a VFD and electric check to see if the owner wants to spend the money on one or the other to keep from manual valve operation at each startup.

 

[bimr]

As hydrae stated, the largest surge that typically occurs at a pump station is when the power fails and reverse flow occurs. That is the reason for a good quality check valve.

Surge on startup would have to been fixed using a VFD or mechanical device such as a pump control valve:

http://www.controlvalves.com/series/125/125.html

 

[stanier]

The information given does not describe the pipeline profile so it is hard to gauge. If the pump has a high static head and the pipe rises vertically before going horizontally then there may be a need for a notehr solution.

In this latter case prof ARD Thorley in Fluid Transients in pipelines recommends an air valve with a pressure control discharge be fitted before the check valve.

Check valves for non slam application can be found at

http://www.noreva.de.

The concept of a check valve for non slam is to have low mass of moving parts and short travel distance. thats why a swing check valve with counterweight is not a good solution. What type of check valve do you have?

I suggest you get a specialist transient analsyst invovled to model the system before outlaying your clients money.

A VFD does nothing for pump trip on power loss in mitigating transients.

Look at

http://www.pipingdesign.com

and search for surges. I have some papaers on their on mitigation of pump transients.

Another reference text you may consider if Pressure transients in water engineering by Ellis.

If one could just say what to do to mitigate the waterhammer I would not have to invested tens of thousands of dollars in this software! On top of that spending countless hours modelling different scenarios to cover all the bases on different systems.

An analyst will cost you less than what you pay a plumber for fix a pipe and heaps less than the lawyer you might need if you kill someone.