High pressure loop 1

[Leonardop318]

Hi everyone

I'm working on a project where I have to design a closed-loop calibration rig.

It has to circulate a fluid (water) through it and I need to be able to raise the pressure in steps (10 bar, 30 bar, 50 bar, 80 bar.), I was told that I could do that with a back pressure valve but I'm not sure how that can affect the flow (or if it would work at all), since flow has to be constant.

Another problem is the pump, I was convinced I could use a plunger pump but I just read the manual and it says that "high-pressure closed-loops must be avoided".

Do you guys have any suggestion?
Any help will be highly appreciated.
Thanks in advance.

 

[Artisi]

It will need to be a positive displacement pump, if the maximum discharge pressure is to be 80 bar and it's within the pump rating there is no problem - - as a safe-guard you can fit an over-pressure relief valve to relieve at 80+ bar, so don't see any major problems.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

There are two ways of doing this.

Either as artisi says use a variable speed PD pump to vary flow and vary pressure using a control valve.

If flowrate is big you can heat up the water really fast and then need coolers etc as the control valve will dump back to a tank usually.

Or have a closed loop and vary pressure using a presurising pump and a low head centrifugal.
Problem is usually finding a high pressure low head pump.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[1503-44]

Chose the right pump and backbressure valve. When the valve has the proper Cv range vs %open for your desired flowrate and pressure range, it can be made to deliver equal flow even with changing pressures.

Lets try to keep it simple.

Closing a valve increases pressure drop and reduces flow, however if the pump increases pressure correspondingly, see pump's head vs flow curve, flow across the valve will increase again, so simply make sure that the pump can deliver the same flow at its increased outut pressure and it should work out well. A variable speed pump is not necessarily needed to do that.

 

[Artisi]

Simple, positive displacement pump with the correct pressure control valve, or am I being too simple.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

What is your planned flowrate range?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

LI, constant, but you didn't consult your crystal ball, did you?

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Leonardop318]

Thanks for your answers guys,

About the pump I was considering a PD pump but I read they had problems with low-viscosity fluids, maybe there's a type of PD pump I didn't think of.

As for the valve, I couldn't find a pressure control valve with the range I need (10-80bar), but I managed to find this back pressure valve

https://straval.com/products/back-pressure-regulators/bps05-600flg/

that I think I could use.

The flow range can be anywhere between 0.7-17m3/h.

Thanks again.

 

[LittleInch]

Well an 8:1 pressure range with a 20:1 flow range might really be a bit much to do with one pump and one valve.

I think if this is a closed loop you would be better to use a vessel say half filled worth water and vary air pressure on top using an air compressor to vary your pressure.

With flow you probably need two or three PD pumps with VFD drives to get your flow range.

How low is low viscosity? & what type of pump ate you thinking of.

17 m3/hr isn't too high so you should find some high pressure metering and injection pumps.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Leonardop318]

Sorry I meant flow rate, not flow range. What I meant is that it can work with a flow rate of 0,7m3/h or 17m3/h or anywhere between them, not in the whole range between them.

The fluid is water so between 0.4 and 1cP depending on temperature.

Well I checked gear, lobe and screw pumps

 

[Artisi]

Are you saying the flow can be whatever you choose to use, but it will be constant irrespective of pressure?

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[1503-44]

A loop does not have to have much of a pressure drop, only enough to sustain your flow rate around the loop at any given time. Loop average pressure could be independently controlled up or down, maybe with LittleInches air pressure tank. Then, if you will not have much pressure drop due to loop flowrates, your pump power and pressure control valve requirements may be minimal. You might not even need a pressure control valve. If you will have up to an 80 barg pressure drop in the loop, thats quite different and you might need two valves to make such a large pressure drop happen, if there is nothing else providing resistance to flow. Is the loop horizontal, or is there elevation change as well. Since calibrations are usually related more to running at different specific temperatures and pressures, rather than pressure drop, I guess that large pressure drops in the loop probably wont be necessary, unless you need to push that 17 m3/h through a loop pipe diameter that is way too small.

As you can see, how you want to operste the loop may drastically influence your design. If you need more help, telling the full story now would be beneficial. You make us guess to much.

 

[Leonardop318]

I'm sorry, I'm pretty bad at explaining myself so I'll try my best.

The loop will be vertical, it will have an elevation change (around 1-1.5m). the idea is to circulate a fluid through it at 20°C and 10bar, take the readings I need from the meter I'm calibrating, raise the pressure to 30bar take the readings and so on until 80 bar, then raise the temperature to 60°C and start again with pressure 10,30,...,80. the pipe would be of 1inch

The flow rate would have to be constant for a set pressure and temperature, not for all conditions, although the less it changes the better. (I mentioned the flow rate being constant only because I don't know how the backpressure valve affects the flow)

For temperature control, I will use a heat exchanger attached to a section of the pipe and fed by a cooling/heating circulating bath.

Thanks again for your answers and your time.

 

[Latexman]

Sounds like an application for a multi-stage plunger pump. I'm familiar with those on pasteurisers/homogenisers.

Good luck,
Latexman

Engineers helping Engineers

 

[Compositepro]

Perhaps I am not understanding something but this seems a simple problem. Loop pressure and loop pressure drop are two independent variables. To get constant flow in the loop use a gear pump and control its speed. To control loop pressure a separate pump is needed to add or subtract fluid from the loop. This pump can be very small and operate only when increasing pressure. An expansion tank would be very helpful in keeping the loop pressure constant/stable.

 

[Leonardop318]

Sounds like an application for a multi-stage plunger pump. I'm familiar with those on pasteurisers/homogenisers.

I thought I could use a regular plunger pump but as I said above I read they shouldn't be used for high-pressure loops, is it different for multi-stage ones?

To get constant flow in the loop use a gear pump and control its speed. To control loop pressure a separate pump is needed to add or subtract fluid from the loop. This pump can be very small and operate only when increasing pressure. An expansion tank would be very helpful in keeping the loop pressure constant/stable.

I don't seem to find a gear pump suitable for water, do you know where I could find one? and what kind of pump would you suggest as a second pump?

 

[Latexman]

It depends on how rough (single stage) or smooth (multi-stage) you want the flow/pressure cycles. As long as you have a properly sized PSV I see no problem.

Good luck,
Latexman

Engineers helping Engineers

 

[georgeverghese]

The flow /pressure range would only be suited to pd pumps, but screw, lobe and gear pumps may not be suited to water due to low viscosity. That leaves only plunger pumps, but flow pulsations in the pump discharge may be a problem when calibrating this flow meter(?). One way to iron out these pulsations would be to install an intermediate gas filled surge tank (with a compressed N2 regulator). Another gas filled drum may be required to iron out pulsations on the pump suction. These tanks can then act as the expansion drum also for this closed loop. The pump suction drum may need to run on gas space pressure to keep level constant(?). It may be simpler to run this plunger pump on VFD to get the required flow.
2 surge drums may be reduced to 1 if you would switch out to some oil as the circulating medium which would allow the use of a screw or gear pump, since you wont have to deal with pulsating flow.