Pump support steelwork vibration 2

[malbek]

Hi All,

I have recently been looking at some jobs involving pumps and electric motors supported on structural steelwork. With regard to vibration of the structure, I know the general requirement is to try to ensure that the fundamental frequency of the structure is not too close to the operating frequency of the machine. However, it would seem logical that for small pumps or motors below a certain size relative to the structure, the dynamic effects will be relatively small enough to say that a dynamic vibration analysis is not required.

So my question is, does anyone know of any guidelines which can be used in this type of situation to indicate that a pump or motor can be considered small enough relative to the supporting steelwork that dynamic analysis would not be needed? Or any other "rules of thumb" which can be used in these types of cases? I know there are guidelines for pumps supported on ground bearing footings which say that performance should be ok as long as the footing is a given multiple of the pump or rotor mass. If there are similar guidelines for pumps supported on suspended steelwork I would be very happy to find these, as I am having to tell clients we need to do dynamic analyses in situations where the pump appears to be quite small relative to the supporting structure, as I don't know of any simple guidelines which would indicate when the pump or motor is small enough not to require this.


Thanks in advance,
Tom

 

[swearingen]

It's a little easier when you talk about concrete supports because the variability isn't there. When you're considering a pump up in a steel structure, there are many things that can happen.

Standing back and looking at it rationally, however, as you know, resonant frequency increases with stiffness and decreases with mass. Therefore, if the rotor's mass is small compared to the long span beam that it sits on, you're probably OK (high mass + low stiffness of long span beam = lower resonant frequency, likely to be significantly less than rotor frequency). Otherwise, you may want to do some quick calcs to see if it is in the ballpark. Outside of that, any complex layout and decent sized motor and you'll have to sharpen your pencil.


-5^2 = -25 ;-)

http://www.eng-tips.com/supportus.cfm

 

[malbek]

Thanks Swearingen. What you say regarding the rotor mass relative to the beam mass sounds like a good starting point. I am currently looking at a pump package weighing 3600kg, critical operating frequency is that of the electric motor rotor which is about 90kg. The whole package is supported on supported on two 5 metre beams each weighing about 280kg. Rotor is therefore only about 2% of the weight of the combined mass of the pump package and supporting beams. In this case it sounds like the rotor weight might be sufficiently small not to be too concerned about potential vibration issues? Be good to hear your thoughts regarding that one.
Regards, Tom

 

[WARose]

When it’s on the ground, (as you said) there are some pretty good rules of thumb that will let you skip a dynamic analysis. (I.e. if the machine is about 5 kips or less in weight and less than 200 hp, then just select an appropriate mass ratio for the foundation to machine and some appropriate dimensions and you are good to go).

But when it’s on steel (or other flexible type framing…..all bets are pretty much off as to what can happen. There are so many different d.o.f. that could be excited, you just can’t take the chance of not investigating it (unless the manufacturer says otherwise).

 

[BigInch]

The problem is that that mass ratio must be concentrated and provided very closely to and be very firmly attached to the pump skid and the surrounding mass must be firmly attached to that.

A 3510 kg pump on steel support definitely needs some structural dynamics work. Don't look to rules of thumb to help with that one. Steel design often has relatively high frequency resonance when compared to soil-concrete foundation systems. Still you may have intolerable vibrations transmitted to other appurtenances within the structure. Don't forget to check short circuit loads too.

you must get smarter than the software you're using.

 

[malbek]

Thanks for your comments folks.

The pump itself is not a cause for concern as it operates at a frequency sufficiently lower than the first mode of the supporting structure. The electric motor is the issue here as this runs close to the structure frequency. However as I said above, the rotor is only 90kg out of the total package mass of 3600kg (which comprises 400kg motor with the 90kg rotor I refer to, plus reciprocating pump (not critical), pipework, valves, skid etc). Therefore the rotor is only 2.5% of the pump package mass, or about 2% of the combined mass of the pump package plus supporting beams. Given BigInch's comments above about mass being concentrated and attached to the pump skid, this would seem to apply here.

Therefore if you are looking at ratio of rotor mass compared to the overall mass needing to be mobilised as an indication of whether dynamic analysis might be required, in this case the ratio is low at about 0.025. If I was forced to make an intuitive judgement I would think that this was probably low enough. If you then take other factors into account such as that it is not in an "occupied" area, and there is no other sensitive equipment in the area, then that might suffice? I suppose the crucial aspect here would be the ratio of the masses. I would be grateful for any more thoughts on this.

Thanks again,
Tom

 

[WARose]

I don't know (given this thing will be on elevated framing) that I would eliminate any piece of equipment from a check based on where the frequency compares to the "supporting structure". Is that the overall building it will be going in or is it the framing the pumps/equipment will actually sit on? The other thing to remember with this sort of thing is how it can cause unintended excitation elsewhere. Some years back I got called in on a vibration problem where some members of a roof truss were shaking like crazy from a piece of equipment 2-3 bays over.

In the past, I've put in the scope (in the assumptions; for stuff on the ground) something like "it is assumed that dynamic analysis will not be required for [stuff on the ground that fall into the criteria I gave above]". But since you are up on elevated framing (like I said) all bets are off. I can think of numerous situations (in offices and even in industrial settings) where I have been asked to address vibrations from equipment that you wouldn't think in a million years would cause a problem.

But if you don't have the time/budget to model the whole deal, perhaps its best to tell the client up front: "I've looked at the overall structure and the framing this stuff actually sits on and we are ok.....but I cannot promise that something X bays over may start causing annoying vibrations. Is that ok?" The key part is that last sentence. Some things will be fine with some manufacturers.....some not ok with others. If we are talking a paper manufacturer (or chemical) I've gotten away with some pretty high vibrations. But in the semiconductor business, I use to get held to some pretty strict guidelines. So it just depends on who it is and what they are willing to tolerate.

 

[malbek]

Thanks for that WARose. The motor sits on a skid frame which will be bolted on to the deck of an oil platform. I have already done frequency analysis of this, however that doesn't really tell me enough as it doesn't take any account of the magnitude of the forcing function. I will now be doing a time history analysis with the relevant forcing function input.

What I was really trying to get a handle on was, is there a point where (say) a motor can be said to be so small compared to the mass of the supporting structure that a full dynamic study would not be required. It would appear that there are no clear guidelines for this, which is understandable due to the potential amount of variables. It looks like the only situation where you could neglect dynamic analysis would be where you were looking at something maybe the size of a domestic washing machine.

Regards,
Tom

 

[TLHS]

On the subject of this, can anyone recommend a good book on forced vibrations in steel structures (machinery is what I'm mostly interested in)? I have a few articles and excerpts and have taken a class here and there, but every book I've read is either incredibly theoretical and doesn't end up connecting the theory to actual design problems or is just a shallow overview without reasonable depth. It would be nice to have a good reference. I have some reasonable stuff about foundations, but not so much about superstructures.

 

[WARose]

I can’t think of too many text books that cover this subject specifically. The best book I have on vibrations is centered around foundation vibrations: ‘Design of Structures and Foundations for Vibrating Machines’, by: Arya, et al (1979). But it is useful in any application for vibrations because it’s one of the few texts that gives guidance in estimating unbalanced forces for various types of machinery (as well as charts for acceptable levels of vibration for both people and equipment.)

As you noted, you’ll probably find far more books that deal with foundation vibration (or structural dynamics in general) as opposed to steel vibrations……because there are so many more unknowns with foundations (i.e. estimating spring constants, damping, etc). As long as you have a good understanding of the concept in general, there is nothing magic (IMHO) about vibrations in steel structures as there is no question as to what stiffness and damping to use (especially in the lower modes).

 

[dhengr]

Tom:
You say.... “It looks like the only situation where you could neglect dynamic analysis would be where you were looking at something maybe the size of a domestic washing machine.” Be careful, I’ve seen washing machines cause a floor system or the bldg. unit next door to jump all over the place. A good general ref. book is the “Shock & Vibration Handbook,” edited by Cyril M. Harris and Charles E. Crede, pub. by McGraw-Hill. It does not have a bunch of worked problems specifically in steel. Doesn’t AISC have any material on the subject?

 

[malbek]

dhengr,

I don't know if AISC has guidance on this, however I have now done a STAAD time history analysis of this which confirmed that vibrations were within the required limits. With my original query I was looking to see if there were any "rules of thumb" which could be used to say if a motor or pump was small enough relative to the structure so that dynamic analysis could justifiably be neglected. I recognise what you say about the washing machine, my point with using that as an example was that (in the UK at least) kitchen floors are generally designed for a standard "residential" floor live load and no account is taken of dynamic loads. Clearly this performs adequately in the majority of cases, however I have come across situations where the washing machine seems to cause excessive vibrations in the floor.
It look like probably 99% of cases of pumps etc fixed to steelwork it will be necessary to do at least some basic structural dynamic checks.