Bellows after pump discharge 2

[mmmumuw]

Dear Colleagues,
My boss keep telling that there shouldn't be any bellows after pump discharge. He says there are a basic rule stating that we can not put bellows right after the discharge and inducer.
Thanks,
Cheers
MW

 

[Scipio]

I don't like using them for a few personal reasons, first under most normal services there should be no reason that the piping can't be designed to induce acceptable loading on the pump nozzles. Second, bellows joints are a future fatigue point which will eventually need to be replaced, and unless you have a lot of faith in the knowledge of the guys installing them, they'll be installed like something that they know will have to be replaced anyway. I've seen them installed and pulled more than three times out of their design deflection to accomodate misalignment between piping and equipment. Finally, depending on the size of the bellows joint and the operating pressure, they can actually produce a significant axial thrust due to the geometry of the convolutions which might exceed the load a properly designed piping system would.

I guess the most fundamental rule for applying these things would be, make sure you really need them, because they're just one more thing that can go wrong if not done properly.

 

[Montemayor]

mmmumuw:

I can't add anything of worth to Scipio's valued comments and recommendations except that it is my personal experience that what he states is engineering reality: the flexing and cycles that a bellows joint undergoes in some services will take it to the destructive failure mode eventually - something that you may not be able to control or tolerate.

You haven't identified your reason for employing them in the first place. You, as an engineer, have to have a solid and serious reason for using them; is it because of misalignment, vibration, flexing, thermal expansion, etc.? For whatever the reason, I still (like Scipio) would strongly advise against their use - especicially if the contained fluid is hazardous or can cause a safety problem if allowed to leak or spill. I have used them with success and with failure(s) as well. I would not venture to use them on an application without experienced, knowledgeable, and trusted pipefitters and millwrights involved in the actual installation and startup.

Even if you succeed in a good installation and application; you still have to maintain a strict operating log and inspection routine on the bellows if you want to go home and relax without any worries. They require serious and strict maintenance on a regular and controlled basis. Then, also, there is the question of using the correct metal or alloy as well as the correct mechanical design and fabrication in order to avoid built-in stresses at each fabricated bend. I hope by now you can appreciate what your boss is trying to tell you.....

I hope this experience is of some help.

Art Montemayor
Spring, TX

 

[PUMPDESIGNER]

I have consulted with two vibration engineers on bellows and such. Both of them were kept ignorant of the other engineer's conclusions.

In both cases, the vibration engineers recommended hard solid fastening at all times.

Measures such as bellows were recommended only if there was a problem, and each case would then be analyzed and solved as a distinct case not applicable to other cases. In other words, if you need some sort of vibration alleviation on one system, do not make that a rule on all future installations.

PUMPDESIGNER

 

[rmw]

If it is a vibration issue, the multi-ply bellows has better fatigue life than a single ply, and, they create less spring forces on the pump.

My opinion is that short of having a specific reason for putting a bellows, as mentioned by Montemayor, you are better off just doing a good job of designing and supporting your connecting piping so that it does not impose any loads on your pump, rather than depending on bellows to cover sins in the design phase.

While I have seen my share of them, I have seen many more pumps without them than with them.

rmw

 

[PUMPDESIGNER]

rmw brings up a valid point and a common problem - using the pump volute to support piping.

The reason is that in many cases there is no problem allowing the pump to support piping in specific cases, on small pumps it is done all the time. But then no one is looking at that to see if it is correct, assumptions are made (presumptions really), then a problem occurs.

PUMPDESIGNER

 

[stanier]

Inmy experience the use of bellows on pump discharge is directly proportional to the level of engineering expertise applied to project. They are prevalent in the building services where little engineering is applied to stress analysis or vibration design and rarely seen in high tech industries like petrochem.

Where piping is run by "plumbers" you will find the ubitiquous rubber bellows either side of a pump. To get them to put tie rods on the discharge unit to prevent loading the pump from the pressure thrust is like asking for gold.

There is absolutely no need for them unless there is a thermal case to design for. If the piping is a thermoplastic material then there well may be a need if the piping layout doesnt facilitate the use of elbows to take up strain. As most "plumbers" install PE/PP/ABS like steel or ductile iron piping they have ended up with bellows.

 

[Artisi]

I think it has all now been said but ultimately the decision to use or not to use is up to you based on the information that has been given.
My input to this discussion is to highlight the fact that extremely high axial thrust from the expansion bellows can be applied to the pump casing.
Be warned.

International College
Naresuan University
Phitsanulok
Thailand

 

[electricpete]

It is interesting point mentioned by Artisi. I believe it refers to the fact that in presence of a discharge expansion joint, a force equal pressure times area of discharge pipe may be applied to pump casing. i.e. given much higher E of steel than the bellows, whatever movement occurs is not enough to put create any significant axial tension in bellows.

At our plant we found that one application had high vibration transmitted from the piping to the pumps, causing false brinneling of the bearings. Expansion joints were the solution to this particular problem.




=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Artisi]

If unrestrained, the force is equal to the area of the bellows which is usually a much larger diameter than the pipe diameter. For example - a 6" pipe bellow type expansion joint (doughnut)could have an effective diameter of something like 8" -- at a discharge pressure of 10 psi this is equal to approx 5000 lbs force at the pump flange if unrestrained.

International College
Naresuan University
Phitsanulok
Thailand

 

[Artisi]

Correction to my previous msg. force on the inlet flange will be 500 lb (not 5000lb)

International College
Naresuan University
Phitsanulok
Thailand

 

[mmmumuw]

Yes, thanks all,
If the reason is thermal expansion, is it right enough? The problem happened in thermal oil pumps in our boilers. Our pumps' mechseal and bearing fail too shortly. It is not misalignment case, because we recheck the alignment in hot temperature. It's OK.
The pump size is 100-250. Yeah, until my boss notice the bellows. What is the indication if our pumps suffer pressure thrust from bellows? The volute casing is not broken. If mechseal and bearings fail, isn't it because of misalignment?
Thanks and Regards,
MW

 

[Scipio]

Pressure thrust from the bellows would most likely result in coupling misalignment and, in extreme cases, casing distortion. What kind of pump are you using, anyway? I'd normally suspect shaft imbalance, vibration due to recirculation, and hydraulic radial or axial loading on the impeller if you're seeing seal and bearing problems.

 

[Artisi]

You are aligning the pump at the operating temperature with the pump stationary - therefore there is no pressure in the expansion bellows and no thrust being developed which would result in misalignment.

I suggest that you check alignment as usual and then position a dial indicator in an appropriate position on the pump bearing housing and put the pump online, any movement on the dial gauge will show you if in fact you have any movement of the pump during operation.


International College
Naresuan University
Phitsanulok
Thailand