Flexible connections at pump inlet at discharges 2

[Luizmi]

Dear all,

I am searching some general recommendations to study in what cases the installation of flexible connections at the inlet and discharge flanges of a pump (centrifugal) is advisable to avoid noises and vibrations in the piping.

I thing that the power of the pump could be a good parameter to decide it, but, what is the minimum power of the pump to be advisable the installation of the flexible connections?

Thanks to all

Luizmi

 

[quark]

For suction, it is better to go with flexible connections irrespective of pump capacity. For discharge, I don't like them and my experience has been bad with flexible connections. (As far as my memory is concerned, majority of members opposed flexible connections in a past thread)

Regards,

 

[Montemayor]

Luizmi:

Your question is not specific as to what you propose and you seem to have the idea that you can resolve centrifugal pump noise and vibration with a “flexible” connection. Refer to thread407-94199 for a discussion on one type of flexible connection.

Flexible connections have absolutely no ability to eliminate or lessen pump noise. Their acoustical attenuation value is nil. So this reason would lead to a waste of money. Since you don’t state the origin of the noise and vibrations, I have to assume they originate in the pump. This is a bad installation/application/operation if that is so. Centrifugal pumps should make no noise nor have any vibrations. In fact, the majority of centrifugal pumps that I have installed and operated in over 44 years have run so quietly that the coupling guards have had to be removed in order to see if the shaft coupling was rotating. And my installations have been typical of most others. A well-designed and applied centrifugal pump will generate no detectable noise or vibration. So something is wrong with your installation if that is occurring.

However, noise and vibration can be transmitted to a pump by upstream or downstream equipment. If this is the case, you have another type of problem that should be resolved first because it will affect the mechanical operation of the pump if allowed. No process equipment – outside of that expressly designed to vibrate (shaker screens, vibrators, etc.) – should be allowed to vibrate to the point where it affects critical equipment such as a pump. This is not proper engineering and there is no solution except to cure the cause – not the effect.

There are various types of “flexible” connections: bellows hoses, rubber hoses, omega loops, etc., etc.. You haven’t identified what type you are proposing. I would warn you that in the majority of cases the hose or flexible connection is weakest mechanical link in such a configuration. And in the case of bellows, the flexible connection is subject to a reduced life due to constant flexing.

If you are talking about a contractor’s portable pump skid, designed for mobility and temporary use, that is a different situation – especially if the fluid in question is benign (like water). But if this involves a serious, long-term installation in an industrial processing environment then vibration and noise effects should not be allowed to exist. Any hose or flexible connection you install as a “remedy” to the effects is subject to extensive stress and will fail with time. I am not opposed to flexible connections where needed, but in the case of operating pumps I would be very concerned about the cause (and hazards) of any produced noise and vibrations.

I do not believe that horsepower is a measure of the vibration/noise expected of a centrifugal pump. I am accustomed to a constant, steady “hum” of a centrifugal pump – but no vibration. Consequently, I find no justifiable reason for using flexible connections on process centrifugal pumps – suction or discharge. Perhaps you have other reasons for your proposal; if so, please indicate with specific details and information.

I hope this experience helps you.

Art Montemayor
Spring, TX

 

[PUMPDESIGNER]

My company has consulted with two consultants and one manufacturer concerning the use of flexible couplings.

Both consultants agreed with each other 100%.
The manufacturer of vibration pads also agreed with the consultants completely.

Based on those consultations, and on my companies experience, we have adopted the following policies with great success. We realize however that in some industries our policies will not work.

1 - Mount everything rigid as possible. This forces all vibration frequencies to a higher pitch which helps.

2 - Do not use flexible couplings at all.

3 - Do not use vibration pads at all.

We feel that if a problem occurs it can be corrected in the field. We have never had a problem in spite of some seriously tough situations where the pumps were within 100 feet of people sleeping.

All pumps that we work with are in the 450-8000 Ns range, and always less than 9000 Nss range. These types of pumps are highly resistant to vibration problems. Use of other pumps with higher Ns and Nss values would definitely change things because with high Ns and Nss values the operating ranges of the pumps are tiny, and the pumps react to low flow rates very badly.

PUMPDESIGNER

 

[jmw]

An interesting question and comments.

In short fast sample loops, there has sometimes been a problem with 1" gear pumps in close proximity to some vibrating element sensors.

The engineers on site seemed always to be convinced the problem was due to transmitted vibration and installed flexible couplings with successful outcome.

I don't suggest that they
[ul square][li]understood the problem fully[/li]
[li]understood how what they did resolved the problem (if it did)[/li][/ul]

In some such situations I'm afraid commissioning engineers must adopt practices similar to car mechanics i.e. just go through a sequence of things they have done before until the problem goes away. Quite obviously, from the comments above, it wasn't the use of flexible couplings per se that solved the problem, but some other aspect of what they did when they installed the couplings.

As to noise etc. this has always proven an excellent web site for pulsation dampers etc.:

http://www.pulsation-shock.com/

JMW

http://www.viscoanalyser.com

Eng-Tips: Pro bono publico, by engineers, for engineers.

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Luizmi]

first of all, thanks to all!,

Montemayor, the question is not based on an existant installation, is a question for a new installation design. In ships, where i worked before, the good practice learns that it is better to install flexible couplings (rubber PTFE or similar at the connection flanges of the pump) to avoid the transmission of vibrations to the pipes and, trougth them, to the structure of the ship where they are attached.

Now i am working on energy plants and i supose that the reasons for install flexible couplings are the same, but the cases in where they are advisable are different.

As i understand you, and as a conclusion, flexible couplings are not the solution for vibration and noise problems and the good practice to design new installation is not to install these accesories.

 

[quark]

One reason for my recommendation of flexible connection for suction side is because I mostly deal with chilled water piping. Inspite of welding the lines to plumb, after circulation of chilled water through them, I found stress generation and it becomes difficult to refit the piping to the pump(particularly as the pipes are welded to steel tanks or grouted in concrete tanks). Sometimes this led to damage of pump suction flanges.

Regards,

 

[PUMPDESIGNER]

Quark as usual brings out a good point, movement due to temperature expansion/contraction, which my company does not normally deal with.

PUMPDESIGNER

 

[saxon]

luizmi, If the pump is properly bedded, anchored, aligned, balanced, and the piping is not forced onto the nozzles (except in cases of coldspringing) there should not be a problem with vibration. If there is, something is wrong with the pump or it's installation. As for flexible connections (expansion joints), these are normally used in high temp. service where the dimension of the pump changes due to thermal expansion and would lead to excessive piping stresses on the pump nozzles, which then throws the pump out of balance or alignment or the out right failure of the nozzles.

Hope this helps.
saxon