Design against gaseous cavitation 5

[ColonelSanders83]

Hello All,

At my company we recently had an interesting issue with cavitating pumps.

Two 100% pumps were installed in a chilled water plant and were operating quite happily until the late addition of a thermal energy storage tank (TES, large tank, open to the atmosphere).

Pump Flow rate = 6000 gpm
Pump media = water
media temperature = 55-75 F
NPSHr at Flowrate = 28
NPSHa at flowrate = 38

Once the TES tank was brought online the pumps began cavitating, even though the NPSHa was above required. This was confirmed with a pressure measurement in the field.

The pump manufacturer installed air taps to allow small amounts of air entrainment into the pump suction, however the client rejected this as a long term solution due to long term corrosion concerns from the injected air (oxygen). The final solution is to install VFD's and operate both pumps at 50% where the NPSHr is only 15 ft.

All sign point towards the pumps suffering from gaseous cavitations due to dissolved air in the water coming in and out of solution in the pumps areas of low pressure.

My question is how does one design against the gaseous cavitations failure mode in chilled water cooling tower and TES tank applications? All the information I have found to date states gaseous cavitations will occur in these applications, but there is very little direction on how to prevent its occurrence during design. Allowing its occurrence is not an option due to customer perception issues even if the risk of damage is low. Any guidance would be greatly appreciated.

 

[ione]

BigInch,

Cannot really understand whether you’ve found the paper of some interest or whether you’re simply pulling my legs. Anyway it doesn’t matter, I just wanted to give my small and probably trivial contribution to this thread which is becoming somewhat intriguing.

 

[ColonelSanders83]

Gentleman,

As I have stated previously we have a suction pressure gauge.

When the TES tank is operating and the pump is "cavitating" the suction pressure gage reads 0-2 psig (it bounces around)

When we Isolate the TES tank the "cavitations" go away and the suction pressure gauge reads 10-11 psi.

Ione: the 70 degrees quoted is the warmest the water ever gets, due to volume comparisons in the system and the fact that water at this temperature is a sponge for air, the system quickly reaches equilibrium. There is no place in the system where the air can ever come back out of solution, except at the low pressure points of the pump.

According to "Cope with dissolved gases in pump calculations" by C.C. Chen the work previous to his approximated an equivalent vapor pressure based on experiment to be approximately the average of the vapor pressure of the liquid and the pressure of the gas it is exposed to. C.C. Chen’s work went through and detailed the equations and step necessary to accurately calculate the equivalent vapor pressure (working on this, however it will take time).

For our system the NPSHa was calculated to be about 31', however this was based on the vapor pressure of pure water from the steam tables, (I previously posted a link saying why that was incorrect). For our case pure water has a vapor pressure of about 0.5 psia whereas the air rich water has an equivalent vapor pressure of 6.73 psia. If the equivalent pressure is used in the calculation of NPSHa we find that we only have 15.5 ft of head available and this fails the NPSHr of the pump (25ft), and leads to the noise and flow issue we are having. This shows we will suffer gaseous cavitations in the pump when operating the TES tank.
When the TES tank is isolated our suction pressure goes up by about 10 psi, using the same method above this yields 38.1 ft of NPSHa, well above the required 25 ft of the pump.

It is important not to confuse vaporous and gaseous effects in the pump. they must be accounted for separately using the appropriate vapor pressure, although it is apparent to me that for chilled water systems, the gaseous requirements will govern.


Always remember, free advice is worth exactly what you pay for it!

 

[BigInch]

ione, I meant exactly what I said. The error is now gone and there's the star I promised.

Looks like the NPSHA is starting to add up, or not, as may be the case. Very interesting.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[ione]

So in this specific case it seems that the presence of the TES affects the fluid we have to deal with (dissolved gas produces a different “mixture” than those of the scenario with TES isolated) by increasing the vapour pressure, and thus leading to a vaporous cavitation (NPSHa < NPSHr) induced by gas presence.

 

[ColonelSanders83]

Ione: I don't believe that is correct. The dissolved air doesn't come out of solution the moment the TES tank is isolated. It is still dissolved in the water and would behave exactly the same way, causing the onset of cavitations as calculated by the use of equivalent vapor pressure values. There will be no change is the dissolved water through the system overall, once the water absorbs all the air it can up to its fully saturated state the water and air will be in equilibrium. The only thing that can change that in this system is the change in pressure at the pump suction. Once the water passes through the pump to the discharge side the higher pressures will cause the gas to go right back into solution.

The difference is one of an open loop and a closed loop with the subsequent increase in suction pressure associated with the closed loop.

As I stated above, with the open system suction pressure is 0-2 psig and when the system operates from a closed loop the suction pressure is 10-11 psig.


Always remember, free advice is worth exactly what you pay for it!

 

[BigInch]

Great. A red herring? (I withdraw the star. No, keep it. Its a good thing to think about, maybe.)

Well then do you think we're finally getting back to something relating to the piping configuration to and from the tank? You must be losing some head there.

What's is the tank outlet valve's head loss and the tank's outlet nozzle flow coefficient? Please don't tell me there is 1000 feet of pipe there too.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

All very academic and interesting - but gave me a headache reading it.

I'm staying with an NPSHa problem at this stage.

Where on the curve is the pump running in relation to BEP?

 

[ColonelSanders83]

Big Inch, I don't think it’s a Red Herring, just the wrong conclusion. The water is saturated with air, going back to a closed loop does not change that.

Please see my above post detailing that the TES tank is 500 ft away. This is what accounts for the differences in suction pressures in the two configurations (closed loop and TES tank operation).

The real issue is that the pump sounded like it was cavitating when the NPSHa to NPSHr comparison said it should have been fine.

The use of the equivalent vapor pressure for the air saturated water shows that the pump will indeed operate noisily from gases coming out of solution when the TES tank is operating. It also shows it will be adequate when it operates from a closed loop.

The papers I have referenced, esp. the one from C.C. Chen details out how to correctly calculate the equivalent liquid vapor pressure. I can then use the equivalent vapor pressure to calculate the correct NPSHa for this system, which will be substantially lower the NPSHa calculated using the values for a pure liquid (see above).

This allows me to design against the failure mode of “gaseous cavitations” or “noisy pump operation”, which was my original question. (I admit gaseous cavitations is probably a misnomer but I didn’t make the name).




Always remember, free advice is worth exactly what you pay for it!

 

[BigInch]

Are we finished then?

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[ColonelSanders83]

BigInch, I believe so. I thank everyone for their sometimes lively and spirited contributions to this a-typical and (in my opinion) very intriguing problem.

If there are any further thoughts on the matter feel free to add.

Always remember, free advice is worth exactly what you pay for it!

 

[BigInch]

I'd like to see the spreadsheet some day (to offer a critique the head loss calcs), but if you're happy, I'm happy.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[ione]

Following ColonelSanders83's advise I add this paper.

Pls take a look at page 119 of the attached paper (it’s quite recent).