Cetrifugal Pump Head Design

[Sacker]

I am looking at a pump design that I have not come across before and could use some help with.

Pump suction approx. 5m head from tank.
Pump discharge approx. 12m head to a hydroclone (say deltaP 10m head for now).
The hydroclone overs discharge then travels down 3m and discharges above the liquid level into a vessel under 400mbar vacuum.

What exactly does the design head for this pump need to be - simply enough to get it up and through the hydroclone and assume the vacuum & gravity will do the rest of the work?

If more information is required please let me know.

Thanks in advance for any comments.

 

[LittleInch]

Can you sketch this in section with heights /lengths etc.

Is pump discharge to cyclone 12m up or is this the discharge head? Is the 10m DP in addition to the 12m or part of the 12m?

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

 

[Sacker]

Thanks for the comment

The cyclone is 12m up and the dP is additional. I've quickly done a sketch, hope it helps.

It's more the logic behind the design I'm after right now rather an exact head calculated - as the lengths and fittings are not yet finalised.

 

[LittleInch]

Just do it all in absolute as head. I've based this on water.

So for you on the inlet side of the pump it's

10+5=15m

Discharge it's

12 + 10(your DP) - 3 + 4( your final pressure)

= 23m

So +7m, but that's basically just to address your DP through the cyclone.

If the pump stops the flow would continue through the pump to a certain extent as the DP through the cyclone reduces unless you've got isolation valves.

You will need to consider vapourisation at the high point if you've got a high vapour pressure.

Pretty sure I've got this right.

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

 

[EdStainless]

Looks right to me LI, now just select line sizes, look at velocity, and toss in the pressure drops in the lines and you should be good.
Yes, flashing in the top of they cyclone looks like a real risk.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Latexman]

Instead of 12m up to the inlet to the hydroclone, shouldn't it be (12+X)m to the outlet of the hydroclone?

Don't forget pipe and fitting losses.

And, take care with the suction and NPSH. If it is water, it looks okay, but the suction is where most pump problems are.

Good Luck,
Latexman

 

[Sacker]

Thank you all for the replies!

 

[Latexman]

I was going to suggest a small vacuum breaker at your high point to stop cavitation, but that would tax your vacuum system quite a bit. So, maybe enough restriction just before entering the 300 mm Hg vessel so the high spot does not cavitate.

Good Luck,
Latexman

 

[chicopee]

The head of 5 meters has to do with the available net positive suction head because the fluid that is being handled by the pump has a required a net positive suction head. If that required net positive suction head is inadequate than as stated above you'll have cavitation. Cavitation is when the state of the fluid at the inlet of the pump becomes a mixture of vapor and liquid which damages the impellers of the pump. To understand at what at pressure the fluid will become a mixture of vapor and the liquid at the inlet of the pump, get a Mollier diagram of that fluid and at the intersection of the isobaric line (temperature of the fluid)and saturated liquid line, there will be a pressure value and the suction pressure at the inlet of the pump must be greater--if not cavitation.

 

[georgeverghese]

As far as I know, there are no hydrocyclones on the market that run on 1bar dp at design flow - should be 50psi or more. Add a control valve to the cyclone reject line going to the light phase drum, and another CV for the heavy phase exit stream also - you need these to fine tune the reject to effluent dp ratio control. If these terms seem unfamiliar to you, talk to the hydrocyclone vendor. Pay attention to how you would operate at turndown and add on further controls as required. Hydrocyclones dont do well at low flow, since centrifugal forces acting on the light and heavy discountinous phases are much lower at low flow.