How to ramp-up pressure slowly on an RO unit

[ProjectEng]

We have 3 reverse osmosis units, each fed by a 200hp centrifugal pump capable of approx 1600gpm @280' head and 150psi. The actual vertical run between the pump and the inlet to the RO units is about 30'. I have verified the head/flow requirements and we're ok there.

There is a 6" V-ball flow control valve on the inlet to each RO unit. As soon as the flow control valve comes off the seat, the pressure spikes up from 0 to ~130psi.

How can I economically modify this system to allow slow ramp-up in pressure, say 10psi at a time?

My thought is to put a 1 1/2" bypass w/ control valve around the existing 6" control valve. This way, I can use the bypass on start-up and slowly transition over to the 6" valve when the system was at operating pressure.

ProjectEng

 

[BobPE]

ProjectEng:

As you probably already know, spiking a RO is not a good thing, and I just wanted to restate that for everyone elses benefit.

I think your bypass sounds good, but the flow rate in the 1 1/2 line may not be great enough to allow a smooth transition when the 6 inch V-Ball valve is operated. Being that these RO units are so suseptible to transients, I would look to invest in a properly designed flow control valve. A 6 inch V-ball valve is just not the right valve at all for these types of system. The flow control valve would be a variable area unit to simulate pump rampup while maintaining a smoothly increasing flow to the units. This will eliminate transients which I would imagine is your goal.

Good luck...

BobPE

 

[budt]

I work in oil hydraulics so this idea may not be any good for your application.

How about a second 6" bypass valve teed into the line to the RO process that is connected to the source of the pumps fluid. If the 6" bypass was open before the 6" feed valve all pump flow would recirculate at startup. Closing the 6" bypass in a controlled manner would bring pressure up slowly with good control? Bud Trinkel
Certified Fluid Power Engineer
HYDRA-PNEU CONSULTING, INC.
812-853-3234
fluidpower1@hotmail.com

 

[quark]

What about a 3-way modulating valve?

 

[Kailee]

ProjectEng,
Greetings. Sounds like a good candidate for a VFD. Some quick calculations and a little programming should alow you to step up the speed of the pump to prevent any spiking in the RO. Once the unit is up to pressure/speed, control could be fully assumed by the valve.

Hope this helps.

 

[ProjectEng]

Kailee.. VFD is out of the question because the motors are 4160v.

Bob.. I've talked with a couple of reputable control valve manufacturers and they seem to think that the V-ball is the best valve for this application. I don't want to take a large pressure drop across this valve.

budt.. I thought about doing what you suggested after I had made the initial posting and that remains an option. Would I really need a full 6" recirc though? I think I could go smaller. I'd generate some heat initially but only for the time it takes to switch over to normal operation.

This pressure spiking was a big problem previously because it was a high-pressure RO system, operating around 350psi. I'm being told now that I shouldn't have to worry so much about this issue because the pressure will only reach a max of 150psi. But I think the pressure will just spike up to the 150psi now instead of the 300psi or so before.

 

[trihydrate]

ProjectEng,

Are you solving an existing problem or solving a problem that might exist? The reason for that question is, there are thousands of RO systems out in service that do exactly what you describe, start and spike from 0 to 200 psi without identifiable problems. Is it the best situation? Of course not! Is it doing harm? Apparently not.

Do you have a transfer system delivering the water to the high pressure pumps? If so, could you delay the start of the high pressure pumps until the RO system pressure has risen to the normal pressure of the transfer system, say 50 psi, that way your pressure spike would only be from 50 to 130psi.

Good luck.