Foundation Wt vs Pump Wt for Dynamic Analysis 1

[rmc2008]

I have two pumps that each weighs 6850 kgs (15100 lbs) and runs at 3600 RPM. Owing to certain restrictions the foundation on which my pump sits is only 650 mm thick (13.3m x 9.35m) with the two pumps centered on a 7.7m x 6.5m x 150mm thk pedestal. This big pile cap 13.3 x 9.35 is supported on 21 nos -500 dia concrete piles 10 m long embedment.The weight of foundation is approximately 15 times weight of pumps together. Dynamic analysis of the foundation was performed using DYNA. In the machine performance curve for a node at the motor location the resonance frequency was found to occur at 100% normal operating speed. See red curve in attachment. Since the foundation is so much heavier than the pumps should I be concerned about this?

 

[JoshPlumSE]

Yes, I would be a little concerned.

But, it looks like you've done a good amount of analysis to demonstrate that the vibration (even though at or near resonance) will not cause vibration amplitudes sufficient to cause faults with the pumps. Right?

What about operator comfort? Will there be maintenance workers there for any significant period of time? Will the vibrations cause problems for them?

These sorts of analysis are very dependent on stiffness assumptions and damping. I assume you've played with these values a good amount already. Assume some degree of cracking in the slab. Assume variations in pile stiffness.

My guess is there will be significantly more damping (due to contact with soil) than is really modeled. Which might help.

 

[SlideRuleEra]

This big pile cap 13.3 x 9.35 is supported on 21 nos -500 dia concrete piles 10 m long embedment.

How thick is the pile cap?

http://www.SlideRuleEra.net

 

[JoshPlumSE]

Owing to certain restrictions the foundation on which my pump sits is only 650 mm thick (13.3m x 9.35m)

 

[WARose]

I've said it a bunch of times on this forum: the foundation dead weight to equipment weight ratio being high is great (as a staring point for design).....but ultimately, what your analysis tells you is what you should be concerned about.

 

[SlideRuleEra]

Thanks, Josh, I missed that.

rmc2008 - The "thin" pile cap is probably the root cause of the vibration issue, the outer perimeter piling are probably not rigidly loaded due to pile cap deflection (deflection will seem trivial, but for vibration trivial matters). Also, I agree with Josh, resonance at operating speed is not good.

On startup, the machines need to pass through any resonances and settle at steady-state speed with minimal resonance. Our steam turbine/generators have massive pedestals on thick, pile supported foundations. The concept is to not only ensure the piles are rigidly and equally loaded, but to have sufficient pile embedment in the caps to mobilized piling mass for damping and transmit vibration efficiently to underlying soil. This way a good bit of soil mass is also engaged for damping. Despite these features, the piling/foundation/pedestal/machine are designed as "tuned" to contribute to cancelling out remaining steady-state vibration. It is that important.

IMHO, do what it takes to make the pile cap "thicker", a lot thicker and deeply embed the piling... there are no shortcuts.

http://www.SlideRuleEra.net

 

[rmc2008]

The portion of the foundation that the pumps sit on (7.7 m x 6.5 m) is 800mm thick and the rest is 650mm thick. Underside of concrete is the same all through. The concrete pile is embedded 150mm into the pile cap. The allowable overall displacement limit at motor shaft is 0.8 mils (from vendor manual) and from my DYNA analysis I get 0.1933 mil. Inspite of this low value, is resonance still an issue? I have a factor of safety of 2 for damping.

 

[SlideRuleEra]

1) The concrete pile is embedded 150mm into the pile cap.

2) The allowable overall displacement limit at motor is 0.8 mils (from vendor manual) and from my DYNA analysis I get 0.1933 mil. In spite of this low value, is resonance still an issue?

1) I would expect embeddment of, say, over 500 mm. Increasing embeddment goes hand-in-hand with increasing pile cap thickness The pile cap as designed has a length to thickness ratio over 20:1. Doubling pile cap thickness (from 650 mm to 1.3 meters) begins to become reasonable... 1.5 meters thick, or so, is likely better.

Pedestal thickness probably should be increased to allow two rebar mats in it and meaningful anchorage to the pile cap. Pedestal not just for "housekeeping" purposes.

2) The issue is that resonance is happening at operating speed. The key is to get all resonance frequencies well away from steady-state speed. Avoidable vibration is the enemy of reliable, long term machine operation. Your client may not make the connection between a questionable foundation design and repeated, ongoing machine anchorage, bearing and alignment woes... but if they do, you don't want to be responsible for it.

http://www.SlideRuleEra.net

 

[Handy_Andy]

60 Hz seems like a very high resonant frequency for a concrete pile cap. Have you verified all your inputs into DYNA? You would need a very low mass for this to be possible.

EDIT: Just did a quick check because I am sick and bored. Definitely need to add thickness, if you can't go up, go down. If the piles are already in place it will be a lot of shoveling but better than questions down the road about why there are always problems with the machines.

 

[Klitor]

Beat me to it andy, seems like a faulty input
For OPs geometry and mass, soil should be some kind of rock. But then...why 10m piles?

 

[WARose]

I have a factor of safety of 2 for damping

And that isn't enough. Your actual damping value could be off by even more than that. (several times that value in fact.)

If you are gambling that resonance (amplitude) won't be an issue because of damping.....I'd re-think that (in fact, being near resonance at all). Damping is always the big "X" factor in this.