Negitive pressure on split case pump

[mech3333]

I have a split case pump on a cooling tower. That is vibrating and noisy. While checking this pump out after a total rebuild I noticed that I have a negative suction on this pump. If I open the value on the suction side were the gauge would be and let air in to the pump the noise and vibration goes away. Can some one tell me why this is happening? I checked the strainer and it is clean. It is hydraulic in some nature and I cannot figure this one out. Thanks For any help.

 

[Artisi]

Sounds like insufficient NPSHa and the pump is cavitating,(too much suction lift or excessive losses on the inlet side of the pump) bleeding air into the system softens the cavitation noise but doesn't cure the problem.

More info would help in identifying the cause, maybe the pump is running too far out on its curve.

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.)

 

[pumpman50]

I agree with Artisi. Sounds like the suction line is blocked giving lack of NPSHa. On the cooling tower duty I presume you have a flooded suction?? Try fitting a combined pressure / vac gauge to the pump inlet nozzle or as close as possible. Close down the pump and take a pressure reading and record it. Start up the pump, let it settle and then take a further reading. If you see a negative or reduced pressure and the suction line is fairly straight with no close elbows or valves you will probably have a line or inlet restriction.

The reason why air dampens the cavitation is that air in a liquid is a poor transmitter of sound. The cavititation is still occurring I'm afraid. The damage varies with power. As a rule of thumb, double the power; cube the cavitation damage.

Most pump problems can be resolved by fitting calibrated suction and dely gauges plus a strap on ultrasonic flow meter (if appropriate) and an understanding of pump curves.

 

[QualityTime]

Hi:

I have written about this many times in some of my previous posts about what the problem is. A few years ago, I had an 80 ton cooling tower, which was manufactured by Baltimore Air Coil (BAC), installed on one of my design projects in Toronto. It sounds like you are having the same problem. The cooling tower pumps were located at least 20 ft below the cooling tower so there was plenty of NPSHa but my pumps were not performing correctly. Assuming your theoretical calculations say you have enough NPSHa I will reintegrate my story and how I solved it:

•The local BAC rep set the working water levels in the cooling tower “as they have always been doing it”

•When the system was put into operation, I had flow meters on my pump discharges and the readings were fluctuating wildly.

•I put my ear to the on the pump discharge and it did not sound like it was flowing full. It sounded like water was dripping.

•The motor shaft/pump shaft couplings experienced short life

•I read the BAC O&M manual and it clearly states that the working water level in the tower is critical and if it is set too low you can get entrained air sucked into the pump.(From expererience I knew that you will get cavitation sounds and it can cause the pump to vibrate if is the air bubbles are very severe…you can also experience a very shortened pump/motor coupling life…which I experienced)

•I checked the BAC shop drawings but could not determine what the working water levels should be.

•The local BAC rep gave me the working water levels that they always set their cooling towers at. Let’s call this working water “level A”. It was confirmed by tape measurements.

•I phoned the BAC manufacturing plant in the US and asked them to send information on what the working water level should be. They sent me two published documents. I had to do some arithmetic and found that there was a discrepancy in both documents (the discrepancy was not apparent until one does the arithmetic!!!). One document said the cooling tower working water level should be “level A”. The other document showed a higher working water level which we shall call working water “level B”. I discussed this with BAC and they were not aware of the two discrepancies. They later advised “level B” was the correct setting.

In summary, once the cooling tower was set to “level B” the pipes sounded full, the flow meter readings are stable (except for one which slightly bounces) and the pump couplings are okay.

In order to give you a feel on the importance of the working water levels the following is an excerpt from my email to the contractor telling him of the water level discrepancies

1.The START TO FILL elevation in the cooling tower was found to be 10 ½” measured from the bottom of the pan. From BAC’s cut sheet that was emailed from the factory today it should be 13 7/8”.

2.The STOP FILL elevation in the cooling tower was found to be 12 3/4” measured from the bottom of the pan. From BAC’s cut sheet that was emailed from the factory today it should be 14 15/16”.

This small change in water level settings made all the difference in the world. Check this lead out and tell us what you found out! Send me your address….I’ll send you my bill later…..

 

[QualityTime]

The other thing you shoud check out is to make sure the strainer in the cooling tower is clean and the suction piping is clear!!!!

 

[mech3333]

Thanks i have check the stainer they are clear. I am going down later this week with gauges. I wll check all that out and the water level in the tower and let every body know what i find Thanks.

 

[Artisi]

You should also try and establish the flow rate as well as look for any changes to the operating system.

Can we assume that there wasn't any problems when the system was first installed and operated but has recently become a problem?

We are lacking vital information to give meaningful answers and stabbing in the dark or theorising won't solve the problem, only accurate facts can help.

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.)

 

[QualityTime]

In his first post he said there was a total rebuild of a double suction pump. That is really unusual. What was wrong with the pump and what exactly did you do to the pump?

The top working water level of the cooling tower should be just below the invert of the overflow pipe.

As a followup to my story, I had to get the local Toronto BAC rep to come back to my installation twice to get it up to work at level B. They were unwilling to bring the water level up that high because they had it in their head that at that level it was too high becasue it was so close too the beach plate and one would get water spilling back into the cooling fan. I told them that from what i could see I did not understand thier concern and for them to phone the BAC factory in the US and get it straightened out between the two of them once and for all. They got straightend out.... Makes me wonder how many other cooling towers in the Toronto area (and elsewhere) are experiencing this type of problem and everyone is scratching their head as to how to solve this?

 

[QualityTime]

Does the cooling tower use a mechanical float or water level probes/solenoid valve to start and stop the filling of the tower?

 

[25362]

http://www.bellgossett.com/Press/BG-mystery.asp

 

[mech3333]

I am going down Wed. We had to rebuild the pump with bearings sleeves (were worn) and new packing.Impeller rings and case rings were in toerlance so nothing was done with them. I belive this has always been there but is know getting worse. I am going to look in to a bunch of different thing now that you were able to focus me in on some of the ones i belived were the cause. You will hear from me soon Thanks!

 

[QualityTime]

Any update?.....I am interested what you found....

 

[mech3333]

Sorry for the delay. After two days and no answer i am lost. I put gauges on discharge and suction side of both pumps. Both pumps were the same, I checked the gauges down the line still the same. I added water to the tower til i could not get any more in no change. I am strating to think this has been there the whole time the pump has been in. Just a little louder now, and becasue we worked on it they think it is something we did we are going to let it run and see were it goes from here. Thanks for all the help hopefully i will be able to put this to rest real soon.

 

[QualityTime]

Hi:

I am not sure how familiar you are with pumps and how to do the head loss calculations but I offer the following comments:

• What are the working water levels in the cooling tower

• It would be best if you provided a schematic showing the elevations of the centerline of the pump compared to the cooling tower working water levels and pipe sizes. This way we can tell how flooded the pump is to start and a quick calculation can be made to determine the head loss in the suction piping. From there it can be calculated if the pressure conditions going into the pump suction should be positive or negative

• Provide pressure gage readings on the suction and discharge side of the pump. Please make sure you accurately show where the pressure reading is taken on the piping system and the approximate height the center of the gage is compared to the centerline of the pump.

• Pictures of the entire setup would help

 

[Artisi]

Have you checked flow and head against the original design, taking gauge readings tells you nothing - unless you compare them to something already known.
It sounds like time to employ someone qualified to sort out your problem.

There are many reasons why a pump can be noisy and vibrating.

Running over capacity
Pre-rotation of the inlet flow
Badly designed inlet pipe configuration
Recirculation within the pump
Large quantity of entrained air
Poor pump design

Any one of the above can result in noise and vibration and it is possible to have more than one present in an installation.



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.)

 

[Artisi]

Need to addd-
Running under capacity.

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.)

 

[rmw]

When you opened the air valve, your suction flow stream then consisted of water and some air. A pump is a volume device, so the water flow had to reduce to accomodate the volume of air being pumped. That reduced the water flow and pushed the pump back up on its curve or back far enough to get below the NPSHa point.

Do you have any means of throttling the flow downstream to see if you can quieten the pump(s)?

rmw

 

[Ziggypump]

Has the balancing valve (triple duty valve?) been reset to the proper setting? I know with triple duty valves it is a vary common occurrence to shut down the valve for pump repair, then open the valve all the way, eliminating your balancing ability. This is about the only thing I can think of that could be causing the problem.