Jet/Eductor Pump Stupid Question

[KernOily]

Hi guys. I have a question and the vendor can't answer my question (!).

Here is the question. Do I have to add the motive fluid flowrate to the suction fluid flowrate when calculating the discharge head? I say yes, because the motive fluid blends with the suction fluid inside the pump, and now the blended stream of combined rates is passing through the discharge line.

I asked the vendor and he said "In 22 years of sizing jet pumps I have never had anyone ask me that...".

Here's the situation. I need to pump 70 gpm of 200 F water using an eductor/jet pump. The required discharge head (pressure) is 15 psig. This 15 psig is the sum of the destination static head + the friction loss in the discharge line due to the 70 gpm. The jet pump selected to do this job requires 315 gpm of motive fluid.

So I ask the vendor: Do I add the 315 gpm of motive fluid to the original 70 gpm of suction fluid and recalculate the deischarge head based on 385 gpm, since the friction loss part of the total discharge head just went up by the additional 315 gpm of the motive fluid? He couldn't answer my question.

Am I missing something here?

Thanks! Pete

 

[zdas04]

Yes, you are missing something. Eductors and jet pumps are devices to pump liquid (you used the terminology correctly, not many do). You can compare the suction pressure to the exhaust pressure to find the total head across the pump. Use that number and the suction flow rate to find the effective hp used (that number is the numerator in an efficiency calculation).

Think of your 315 gpm power liquid exactly the same as the electricity to drive a mechanical pump--it is input power. You can use the exhaust pressure and the 315 gpm to determine how much mechanical power will be required to get the stream back up to power gas pressure (that number can be the denominator in an efficiency calculation).

The only time the 385 gpm is important is calculating the exhaust piping size required to get from your exhaust pressure to your sink pressure.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The main difference beteen humans and apes is that we have cooler tools

 

[insult2injury]

If you are not comfortable making the distinctions, there is no problem with providing the vendor with a system resistance curve and having him select the correct size pump for the application.

I2I

 

[KernOily]

Guys thanks for the replies. But it appears you both missed my question. I was afraid that was going to happen. I will rephrase and I attached a sketch.

Please see the sketch. Here is the question.

The pressure required at the eductor discharge to move 70 gpm is 15 psig. If I add 315 gpm of motive fluid to the original 70 gpm, what will the discharge pressure be?

 

[zdas04]

I think I understood your question. If your design exhuast pressure is 15 psig, then you need to provide enough downstream pipe size to allow 385 gpm to move from 15 psig to the sink.

It isn't an eductor question it is a pipe question. The eductor will be designed by the manufacturer to provide 15 psig at the exit of the diffuser. What happens from there depends on your piping.

If your discharge pipe is too small, then the eductor exhaust pressure will increase and put the unit onto a different pump curve. The exhaust pressure is always a function of the system you are exhausting into. Saying

The pressure required at the eductor discharge to move 70 gpm is 15 psig

is really not accurate. It would be better to say

In order to move 70 gpm from __ psig to 15 psig I need 315 gpm of power liquid at __ psig

See the difference?


David

 

[insult2injury]

I too understood your question. If you give a system resistance curve, it will show the discharge pressure required across the whole range of discharge flows. It is then up to the vendor to select the correct operating point on that curve to move the quantity of liquid specified. Be sure to be clear that the flowrate is total discharge flow so that nothing is overlooked.

I2I

 

[KernOily]

Thanks. Let me try again.

Do you agree that the motive fluid mixes with the suction fluid inside the eductor?

 

[KernOily]

OK I think I get what you're saying. The problem is that the required motive fluid rate is not known a priori when sizing one of these things at the outset. Therefore I can't include it in the eductor load when sizing it.

The original 15 psig is based on a 2" discharge line (it's real short).

If I understand David correctly, what I need to do is increase my discharge line size until the discharge pressure at the eductor exit is 15 psig when 385 gpm are flowing.

All of life is trial and error, no? Thanks!

 

[MikeHalloran]

Yes, the fluids mix; neither of them just disappears.

You recalculate the size of the discharge pipe based on 385gpm, not 70 gpm.

Or you recalculate the operating point of the eductor based on the flow resistances of the pipes you have selected.

Yes, you are working the problem the hard way, but you should eventually get there.

The eductor salesman may have misunderstood the question because you didn't phrase it exactly the way he's used to hearing it... or he may have been doing it wrong for decades.





Mike Halloran
Pembroke Pines, FL, USA

 

[KernOily]

Right on, thanks guys, I think I got this - sometimes you have to beat me about the head and shoulders with it ;-) As Eric Idle/John Cleese would say - ' A priviledged glimpse of the bleeding obvious'

I think if I'd had a performance curve a la centrifugal pump it might have been obvious. But the eductor folks publish performance tables. Same thing, different format - guess I had blinders on by trying to make it fit with what I'm used to seeing all day.

Thanks! Pete

 

[Artisi]

Saying that you require 15psig at the eductor exit is putting the cart before the horse.

It is possible of course this figure has been quoted as the residual pressure available at the eductor exit (NOT a required discharge pressure as you advised) after allowing for the driving force to entrain a Q1 of 70gpm and discharge Q2 of 385 gpm against a nominated head imposed on the eductor you advised the supplier.


When using eductors the important parameters are
1) the flow required (Q1),in your case 70gpm
2) the amount of flow required (Q2) as the driving component, in your case you report 315 gpm.
3) the pressure required (P1) at the inlet to the eductor to drive Q2 (315 gpm) entrain Q1 (70 gpm) and overcome the discharge head produced by the flow Q3 of 385 gpm plus the friction and static head losses etc in the discharge pipe work.

2 and 3 are usually available from a curve or tabulated data once you have calculated the head loss in the discharge pipe -- remember there is a residual head (pressure)at the eductor exit which must be subtracted from your friction head calculation - unless the tables are set up in such a way that head imposed on the eductor is already tabulated.

Trust this hasn't confused the issue any further.

 

[zdas04]

I just finished a detailed design of an ejector (gas as the power fluid) and I had a bunch of options of what to let float and what to fix. I found that the answers were much more robust if I fixed suction mass flow, exhaust pressure, and power pressure and calculated power gas mass flow and suction pressure.

Any design that doesn't treat exhaust pressure as a given is asking for a lot more work than is really necessary. It is really easy to take a sink pressure and a total mass flow rate to determine a pipe size that results in an acceptable exhaust pressure. We've all done these pipe sizing exercises. Designing a thermocompressor (either an eductor or ejector) from that point (with a fixed suction flow rate) is a lot easier than any of the alternatives I tried.

Still, going back to first principles to design a thermocompressor is not really as easy as you would hope.

David

 

[KernOily]

Good points guys. Perhaps my error was that I initially approached this eductor sizing problem as one would size/select a centrifugal pump: the required discharge pressure (and TDH) are calculated based on the job one is trying to do. The best-fit pump is then selected. In my typical case, the suction and discharge pipework, the design flowrate, and the design TDH are all usually fixed, so there is usually one best-fit pump, all other things remaining equal.

What was throwing me off was how to deal with the motive fluid. I had blinders on, being used to centrifugal/PD pump sizing, where I usually have a fixed discharge line size based on velocity sizing contraints or overall dP and I vary the pump selection to fit the job. This situation was a little different: select the pump, fix the discharge pressure, then vary the discharge pipe size to meet the required discharge head. So I learned something. I'm a little disappointed in myself; I should have been able to figure this out. Thanks!

 

[zdas04]

I see that a lot. Pump guys take the piping in/out as a given. Pipe guys take the pump capacity as a given. The only way to look at these things is to look at the total system where the performance of each is absolutely dependent on the rest.

David