Multiple Compressors on Common Foundation 4

[privateer]

I've started at a new company that does a lot of compressor stations in pipeline facilities. The company has had a lot of success installing multiple compressors (3 to 5 per) on common concrete block foundations - meaning, lined up in a straight row on the common concrete (3-4 feet thick) foundation. There has been no vibration issues. They even install pre-fab metal buildings on the same foundations (even with bridge-cranes) that cover the compressors, without any vibration issues in the structures too. The compressors used are in the 5000-6000 HP range, weighing 110,000-140,000 lbs. (entire package, including the structural skid).

My project manager asserts that this is the way he has always done it, and has had success for about 30 years this way. As a design engineer, I do not dis-believe him - actually I would like to prove him correct somehow (it's not enough to be successful, I need to understand WHY/HOW we are successful..)

I have done compressor foundation analysis before, but by no means am I an expert. But the foundations I have done were a single compressor on a block foundation, and the overhead structure foundation was separated (separate column footings and isolation joints..)

In my research, I have found no design guide to address multiple compressors on a common block foundation, other that several "references" that comment that this is actually a favorable condition - that the vibrations "cancel" or dampen because vibration frequencies are scattered by the multiple machinery running at the same time....

Any thoughts/comments/ideas....????

 

[kslee1000]

Have you checked the compressors, and sure there is no vibration absorbers? I am not convinced the "dampen" effect, but the thickness (massiveness) of the block foundation may have played important role.

 

[kslee1000]

Also check with the compressor supplier, the "dampen effect" could be achieved by design through cycling control.

 

[Ron]

It sounds like good ol' common sense trumped engineering analysis on this one. The mass of the foundation and the mass of the compressors has likely damped the vibrations. The spacing of the compressors likely has been such that the harmonics produced by each are either very low or out of phase so that they don't do any damage.

You can analyze and prove it, but sometimes it's just as good to listen to a finely tuned ear!

 

[privateer]

Ron, kslee - Thanks for the quick replies.

kslee - I'm not sure what the manufacturer is doing to dampen the vibrations. The unbalanced loads are rather small. And the manufacturer did mention these were very well balanced compressors. Thanks for that lead and I will follow up.

Ron - I certainly understand that point, and yes, I agree that "common-sense" can trump the calc. One of my professors used to say "Engineering is not an exact science," and I used to think that was a crazy statement for him to make until I grad. and started working.

A big concern here is this project manager will not be around forever and when he passes the torch, I don't won't people to just copy what he did without any thought or understanding.

Yes, a lot of mass we're dealing with here. The common block foundations are typ. 42' wide by 94' long and 3-4 feet thick with say 3 compressors. If the compressors are all tied to the block foundation, then wouldn't their own collective masses also contibute to the dampening? Also, according to the soil report, we have fat clays in the area.

 

[kslee1000]

Per your last statement:

We usually do not consider the mass of equipment, which has its own frequency, and is external to your foundation (tied or not).

Soil condition varies from site to site, it does have dampen effect, however, it is usually excluded from structural vibration analysis, because of its complexity.

I couldn't pin point the literature at this moment, but suggest to search ASCE for publication on "Foundation for Vibratory Machinery". It may not cover your case exactly, however, it provides the basis and quite a few common sense type stuffs. Also, asking around internally wether there is "rule of thumb (may not in the literature)" type design guide, that recommends foundation mass/dimension to avoid lengthy vibration analysis.

Finally, if your boss has 30 years of experience, and you are thinking to carry the torch, start from now, talk to him one case/question at a time as frequent as you can/he's available, and write down the notes. It will carry you a long way. Communication is the key. Good luck.

 

[Ron]

I believe in this case you would have to consider the mass of the compressors, even if indirectly. The compressor is the driving vibration. It's internal vibration is damped by the housing or framework, which transmits the damped vibration to the foundation. If multiple compressors are running at the same time and out of phase with respect to harmonics, then one or more compressors will damp the effects of one of the compressors, so at least the mass of one of the compressors could be considered to be damping the vibrations of the other two at any given time.

I would suggest that you measure the vibrations at various locations concurrently, then back into the analysis. That should tell you a lot about the "how and why" of your conditions. You know they work, but I agree that a full explanation of "why" they work is important for the long term.

 

[concretemasonry]

I worked in the early 1960's in the aerospace industry and was responsible for the design and modification of many facilities including rocket test stands (corporate investments), equipment (pumps, dynamometers, etc.) and work on the the U.S. stands at Edwards AFB.

What I quickly learned is that no matter how much technology and support you have a available, no amount of sophistication of analysis and instrumentation will never overcome the effect of pure mass, since it is very cheap and readily available. If the equipment is not firmly anchored anything you get from instrumentation is really not worthwhile if you do not know the relationship to the foundation and the soil response. Bigger was better, cheaper and more repeatable. - We did not get paid for future analysis to determine the cause, effects and responsibility. - We just did was necessary and it worked!

This is in contrast to the analysis of existing rocket test stands (buildings without sides on a side of a mountain supported by concrete foundations, holding up several huge tanks of fuel up to 20,00 gallons where the dynamic loads went up instead of down and were cyclical). Te vertical loads were from F1 Saturn engines that had a thrust of 1,500,000#. We were fortunate enough to have the resources and short time line to monitor the structures with accelerometers at all joints and use a sister companies air frame analysis/design program to look at the structural performance on a member-by-member basis and compare to the current AISC code.

I make this post to point out that it is fine to go into the details and theory, but there is a time and place for the detailed analysis and design and there is a place for the common sense and "seat of the pants" conclusions.

If you have a horrible soil condition the approach of using massive concrete (cheap)is difficult to beat, especially if you have multiple pieces of equipment operation on an unknown sequence.

Dick

dick

 

[privateer]

Ron - If I understand your point, say Compressors A, B, and C are on a common block foundation. If all the compressors are set properly, then we can say that the static masses of compressors B and C help dampen the vibrations of compressor A (along with the mass of the common block concrete), right? And the same for the other compressors, etc. Also, we would have to assume they are operating out of harmonic, because it would be very hard (if not impossible/impractical) for them to coincedently operate within harmonics of each other....

concretemasonry - I feel you are converging with my project manager's stance. And that's OK.

But we need to keep in view that the client may want to put a steel frame (pre-fab) bldg. over these compressors in the future. A major concern here is that the foundation vibrates and causes excessive vibration in the steel structure, causing a lot of displacement and shaking things loose, causing things to fall and conk the poor operators on the head. But all other projects like this (same arrangement, same facility) have had success - meaning, no vibration in the steel structure(s).

I think a theory is forming here - This situation may work for the specific conditions and applications for this certain project and at this facility, but may not necessarily work just anywhere....

BYW, I give you all stars.

 

[BigH]

Soil dampening, if I remember correctly, is only a small fraction of the foundation mass dampening - and embedment, even a small bit, helps considerably. Why don't you monitor the vibrations on some of the existing foundations to see what kind of vibrations you are getting - then, if you are worried about such vibrations beating up on a steel frame structure that "might" be put on the same foundation - you would have some data to support your case one way or the other.

 

[galiano]

I think that concretemasonry gave the best advice about massive concrete foundations. If you don't have vibration problems (high amplitudes) with existing type of foundation blocks, you shoul'd leave it that way. Foundation dynamics is critical only when is too flexible (low stiffness) and when it works close to resonance (assuming that the dynamic forces from compressors are within tolerances). Dampening effect of structure depend only for dampening coefficients of dynamics elements. And multiple compressors may work in constant phase difference only in same driving shaft. Working with different drives each, they create "beats" which frequency relate to rotational frequency difference between them. As for your clients...They have to order dynamics analysis of foundation before any major modification.

I hope this helps,

Galiano

 

[Bribyk]

Talk to the engine and/or compressor manufacturer, if you're packaging their equipment they should give you an Application and Installation Guide. Your company's rule of thumb seems to be inline with manufacturer's recommendations.

One OEM's rule of thumb is FD = W/(D*B*L) where:
FD = foundation depth (ft),
W = total weight of equipment (lb),
D = density of concrete (lb/ft^3),
B = width of foundation (ft), and,
L = length of foundation (ft).

For a 110,000 lb package on a 12' X 20' pad, the depth works out to 3'!

The mass of your foundation should be equal to or greater than the equipment weight (above formula), you have to consider allowable soil bearing loads, and you have to consider the dynamic load of the equipment (depends on whether isolators are used).

 

[Bribyk]

Also, same OEM does touch on foundation issues with paralleled generators and that stronger foundations are required. Depending on the compressor control strategies being used, this may apply to your installations.

 

[Ron]

Privateer...yes, you got my point.

As for the foundation shaking a cover placed on the system, you can isolate columns from the vibration such that you get minimal transfer. Based on the sizes of the equipment, the sound pressure levels might be an issue with enclosures.

 

[privateer]

Ron, how can we isolate the columns?

 

[Ron]

Depending on the size and loads, you can use seismic isolators or heavy machine isolators. I would use the machine isolators...they are usually composite springs between plates. Uplift can be handled by cables if the isolator doesn't have enough capacity. For equipment isolators, load ranges generally go to upwards of 10 kips, which is usually enough for light structural covers.

If you put structures up, you might want to consider aluminum as the frames are light and tend to transmit vibration a lot differently than heavier steel frames.

 

[kslee1000]

Check with the compressor manufacturer, it's not rare that the vibration obsorber, as described by Ron, is in the package already.

 

[privateer]

Ron - Aluminum may be an option - I don't know if they've looked into it in the past and ruled it out. The building structure does need to support a bridge crane (see my initial post), so my guess is it should be a steel structure.

How are the isolators installed? Are they placed under the column base plates? Could you point me to some info on the net?

 

[swearingen]

I wonder about a case that I expect you will run into in the future: three compressors going in a very tight and oddly shaped space.

I bring this up because I did many vibrating machinery supports using old rules of thumb from my superiors (Fdn. wgt. 3 to 5 times machine wgt., fdn. depth 1/5 width and 1/10 length, hgt. to c.g. of machine 2/3 to 1 times fdn. width, etc.) These served me well until I ran into a situation with a 12' diameter I.D. fan that was vibrating like crazy. Granted, the main problem was build-up on the blades, but the original foundation was very small and the client wanted us to change it out. This thing was hemmed in by columns, footings, undergound duct banks and the like and I really had to roll up my sleeves.

The final design had an odd shape and was thinner and far less massive than I would have used by the old rules. I had set up a huge spreadsheet based on text and examples from "Design of Structures and Foundations for Vibrating Machines" (Arya, O'Neill, & Pincus), as well as some data from other sources. I did sensitivity tests on all of the variables and concentrated on the ones that made the biggest difference. In my case, the water content and compaction of the soil were two of the most sensitive, so we used very rigorous testing criteria before, during, after soil placement.

I think it was the sensitivity analysis that did the trick. As noted by some other posts, there are a ridiculous amount of variables involved. If you believe you have accurately modeled them or have even accounted for all of them, you are kidding yourself. However, if you can weed out the ones that really make a difference and focus on them, you can at least bracket your problem and be on pretty solid ground.

This type of problem is what sets engineers apart from layman. A layman could take the rules of thumb given to us and start a company that builds these foundations in perpetuity with no problems. Until - that is - he comes across one that doesn't fit the premise for the rules he was given. This is the basis of all engineering - being able to take our knowledge and apply to things that haven't been seen or done before...




If you "heard" it on the internet, it's guilty until proven innocent. - DCS

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

 

[baw1]

You might want to post your question on the ME Acoustics and Mechanical Vibrations engineering forum. Those guys get really scientific with vibration theory. I would think you would get some interesting responses.