Pump foundation to eliminate pipe vibration

[tholz]

Folks,
I am working with my building management to resolve a recent issue with the building rooftop water tank refill pipe transmitting a very audible humming noise to an apartment in the building. The pump is located in the basement and tank feed pipes run directly under the 1st floor apartment and make a 90deg turn up a riser shaft located adjacent to the apartment. The pump was recently changed out to a higher Kwh pump because of the NYC mandated installation of a backflow preventer. Prior to the new pump, no hum was detected. Between the pump and 2nd pipe hanger, is a Flex-Hose NND double sphere flex connector. The pump is mounted on a steel base which sits on 4 vibration isolator pads and is bolted to a sizable cement foundation, which is part of the basement floor. My question is: Is it appropriate to have the pump sit on vibration isolation padding, if we really don't care about any vibration transmitted to the foundation and ONLY care about minimizing vibrations transmitted to the outflow piping? Seems to me that we would be significantly increasing the effective mass of the pump if it where directly and rigidly affixed to the foundation, and thus reducing the vibration into the piping.

I've attached a pic of the piping and pump mount/foundation

thanks in advance for your time and thoughts on this.
Tom

 

[Strong]

Pumps have a poor foundation. The suction (inlet) and discharge (out) pipes are labeled wrong for a simple centrifugal pump!

Walt

 

[tholz]

Walt,
you are correct about the label being mixed up. My apologies. But back to my question, should we have isolation pads under the pump if our primary concern is noise/vibration transmission to the roof tank supply?

 

[Strong]

In this case it probably would be better to have pumps on rigid foundation with hollow base plate filled with non-shrink grout. The "vibration pads" are poorly installed, anyway with the base bolts acting as rigid short-circuits. I assume that the vibration/sound frequency is from pump vane passing pulsation (typically 5 to 8x shaft speed). Vibration/sound measurements with frequency analysis (FFT) would confirm this. Some possibilities for cause of problem include:
1) New pump is oversized and not running at best efficiency point, so vane-pass pulsation amplitude is higher
2) New pump has different speed or #vanes, so pulsation frequency aligns with resonant structure for amplification
3) Pipes/hangers/flex-joints were changed and are less effective for isolating new pump from pipes
4) Other structural elements have been modified/added in contact with pipes causing a vibration "short-circuit"

Walt

 

[desertfox]

Hi tholz

A few questions if I may:-

1/ Does the new pump run at a different speed?

2/ How well was the new pump and motor aligned?

A pump and motor which are not accurately aligned will cause vibration and as the pump was replaced it would have to be repeated again.
So I would start to check alignment before I started removing anti vibration pads, also if the pump now is heavier and runs at a different speed are the vibrations pads suitable for this pump?

 

[Tmoose]

Is the first floor apartment occupant complaining about the noise?

 

[tholz]

Let me try to answer several posts/questions here:

The vibration pads are new and I don't think where properly selected based on the loaded/operating weight of the pump. But again, my thinking is that vibration pads are used to reduce noise/vibrations from getting into the supporting structure of the pump (ie the foundation and basement floor), but in this case, we don't care - it's the sub-basement, well below residential level. Walt seems to be indicating to remove the pads and install directly on the foundation to create a rigid mounting.

The pump engineers came out several times to confirm proper alignment and operation of the pump, so I have to trust that its aligned within spec.

The pump is constant speed - It's to fill a water tank 18 floors up. Source pressure is 48lbs (but could vary because it's city supply)

Yes - first floor resident is complaining.

Don't know what Frequency the hum is, we'll have to get that checked out.

 

[Strong]

tholz,

It is very difficult with a forum like this to solve the type of problem you have. If you are interested, can you make a digital recording (smartphone or other device) and attach the file? I will send you back a spectrum analysis (FFT). Send the motor and pump nameplate data as well. If the "pump engineers" were responsible for the pump skid installation on the pads in the photo, then I would be question any information related to: "confirm proper alignment and operation of the pump". Do the "pump engineers" agree there is a problem? If yes, did they offer a solution?

Walt

 

[amanuensis]

You have to think either in terms of TPA ( Transfert Path Analysis )or in terms of SBS ( Structure Borne Sound).
Good luck.

 

[tholz]

Walt,
I don't have the pump info yet - it will take me a little time to get it. I did manage to record the noise. Toward the end of the recording you'll hear the pump wind down as it's being shutoff. Let me know if the recording is clean enough to run FFT on it.
thanks again for all your help
Tom

 

[Tmoose]

Is that wav recorded in the noisy apartment, or near the pump?

 

[Strong]

tholz,

I have attached a frequency analysis. Can you confirm the motor and pump nameplate data?

Walter

 

[tholz]

it was recorded in the apartment.

 

[tholz]

Walt,
The Pump guy (I've down graded him from Engineer) did say the pump was running at 3600rpm, so perhaps a Vane Pass frequency of 5x3550 makes sense.
What conclusion would one draw from this? Should we be looking to altering the impeller? perhaps the vane angle or to a staggered vane?

 

[drawoh]

tholz,

If your anti-vibration mounts are working, your pump is vibrating and transmitting the vibration to the pipes. Your pipes have mass, and they are flexible, so they have a low natural frequency. I cannot make sense of your floor. If it is not rigid, the anti-vibration mounts are not relevant. The pump is going to shake.

Are the studs coming from that plate on top of the anti-vibration mounts, or are they coming from underneath what I assume is a rubber pad? If the studs go all the way through the anti-vibration mount, then it will not work very well at isolating your pump from the floor.

If you do not want your pump to shake the pipes, you need the pump clamped to something heavy, like a concrete floor.

Can you clamp your pipes to something heavy and stiff, using some sort of damping material? I would want to clamp each pipe at three points, so that there is no motion that can be transmitted through.

--
JHG

 

[Tmoose]

Looks like this might work, but maybe the sound needs to be fed in thru the audio card.

http://www.techmind.org/audio/specanaly.html

 

[tholz]

Walter,
Here are the pump specs:
Burks 3100-GA7-1.5. 10HP 3 phase 3450RPM. Impeller diam is 7 1/16

Tom

 

[Strong]

I guess this is the pump:

http://burkspumpcatalog.com/product/3100ga7-1-12

I doubt anyone will redesign the impeller on this small pump! You could ask for a different pump with different number of impeller vanes to change the vane-pass frequency. The pump performance may now be too much with excessive flow rate. Pulsation amplitude goes up when operating away from Best Efficiency Point (BEP) ay high or low flow rate. The flexible pipe (suction and discharge) connections must be as close to pump as possible and certainly before the 1st pipe hanger. Consider special pipe hangers with vibration isolators or remove isolators from 1st floor joists and hang from vertical/A-frame supports setting onto concrete floor. There are other vibration/sound control measures, but the problem must be analyzed in greater detail.

Walt

 

[Tmoose]

How much of the piping or it's mounting was changed when pump was upgraded?

 

[tholz]

Tmoose:
The piping up the flex joint was the only change to the piping.

Walt:
The flex connector is before the first ceiling hanger, but there is a small vertical support from the foundation to the pipe before the Flex connector (Labeled as "rigid mount" in the picture). I agree that adding vibration damping hangers might be a solution, but that will be expensive as it will require some opening of walls/ceilings , as the pipe travels horizontally through the ceiling of the super's apartment (which is below the residence that is complaining) and then goes vertical through a riser shaft to the roof.
I was hoping to try to resolve the noise at the source - hence the focus on the pump, mounts and flex-connector.

Drawoh:
Re: studs - my assumption is that the studs are embedded into the foundation - I can't imagine that they are not - but I'll double check.
Re: Pipe vibration - I was thinking along the same lines - to me, it would make more sense to bolt the pump directly to the foundation mass, rather than have the pump sit on vibration pads - this is because we only care about vibration imparted my the pump to the pipes - vibrations imparted to the foundation don't matter since this is the floor of the sub-basement. Not sure how the pump guys would feel about this.