Use of Expansion Bellows 2

[Pipeline1972]

Hi everyone!
I'm doing a bit of research about the use of expansion joints in stress induced piping systems. From what i know and read, bellows should be the last option to consider. does everyone agree with this? What are, in your opinions, the pros and cons?
I've seen several threads about this topic, but i thought it'd be a good idea to put all opinions and ideas in one thread.
The reason i'm doing this research, is because the Maintenance dept. of my company (chemical) wants to apply bellows on all pump nozzles, as to make it easier for them to move the pump for maintenance reasons.
I don't agree with this, so i want to counter them with all the reasons why a bellow should not be used.

Thanks!

Draftsman - Designer Industrial Piping

 

[monkeydog]

What are you reasons for not using bellows?
Seems like placing a bellows at the pump would allow for compliance in the line due to vibration from the pump when the pump is running.

 

[Scipio]

The main reasons I don't like seeing bellows on pump nozzles are;

a) Can lead to pretty lousy alignments since the guys in the field will pull and expansion joint as far as they can and frequently don't even know the design deflection limits.

b) The eddies caused by flowing over the bellows can cause uneven flow into the pump suction.

c) Bellows expansion joints can, depending on size and pressure, generate a lot of axial thrust, which can push a properly alignment pump out of alignment.

That said, there are situations where they are absolutely necessary, extreme operating temperature swings, for instance, but if one of my stress engineers wants to put a bellows on one of my pumps, they need a lot of justification for it.

 

[mshimko]

IF you do use bellows, be sure to specify IN DETAIL the alignment requirements. As mentioned above, there will be a tendency for the installers to assume that one purpose or benefit of the bellows is to reduce the need for alignment.

 

[Montemayor]

Pipeline1972:

You are taking the correct, practical, and engineering stance in challenging those who propose to apply bellows on all pump nozzles. Scipio, an experienced and veteran forum member, once again contributes his insight and sharp engineering ability to this thread. I can only contribute the additional points:

1) Bellows inherently contribute highly stressed metal areas and crevices (the bellows formed) to potentially corrosive chemicals and fluids. Additionally, the bellows are "natural" deposit areas for solids, slurries, and other compounds to further block flow and possibly restrain bellows action. Bellows are usually the weakest link and the one point more prone to corrosion in a piped circuit.

2) Bellows are destined to fail - either by corrosion or by metal fatigue. Consequently, they must undergo a strict and methodical periodic maintenace inspection. This is the tradeoff that naturally has to be expected when one obtains a presumed "free ride" on expansion and vibration applications. As a result of this fact (coupled with the risk of potential chemical leakage), most -if not all - major chemical processing companies will refuse to install bellows expansion joints unless there is no other possible alternative. And once installed, the bellows are subjected to rigorous and methodical, documented maintenance inspections.

3) Bellows normally can tolerate NO vertical loading forces. All adjoining piping has to be independently supported and secured. Safety precautions have to be taken in order to avoid any external forces imposed vertically on the bellows - such as personnel using them as a step ladder during maintenance.

With the chemical leakage risk due to potential rupture, the maintenace time and costs involved in establishing mechanical integrity, and the special design and care that has to be incorporated to the adjoining pipe, you will find bountiful reasons as to why bellows joints should be avoided - not installed - at the vicinity of any chemical handling equipment (especially pumps!)

I hope the above experience helps you out.


Art Montemayor
Spring, TX

 

[Pipeline1972]

Thank you very much all for your insights. I will keep your opinions in mind. I am now even more convinced that the Maintenance's reason for using bellows (it's so much easier to dismantle the pump that way!), is a lame one. The lifetime of the piping and the pump, the risks for leakage or failure, and hence for serious accidents, is infinetely more important than the "comfort" of dismantling an equipment.

Draftsman - Designer Industrial Piping

 

[Thomasjl]

Hi there,

I presume we talk about a centrifugal pump. A bellow makes the flow turbulent and has possible air traps, one is even worse than the other.

The only advantage of bellows is that there are no forces on the pump nozzle, which extends the life time. A proper running centrifugal pump does hardly vibrate, so there is no advantage there either.

If the maintenance has problems with removing and putting back the pump I think the alignment is not very good in the first place.

The only reason that I would allow a bellow is when there is absolutely no room for an expansion loop.

Good luck.

 

[DSB123]

Thomasjl,
"No forces on the pump nozzle" What about the pressure thrust which needs to be considered?? This can be significant depending on the pressure and line size.

 

[clooney]

Hi there,
If only turbulence is your concern, you can always ask for a bellow with a "liner". For metallic bellows, liner is inside the bellow. It is actually a thin-walled pipe (prohibiting direct flow against the inner wall of the bellow) welded to the assembly only at one end. That is why knowing the direction of flow is important when installing a bellow with liner. Since the liner is only connected at one end, it does not interfere with movements of the bellow.
This thread is not intended to promote use of expansion joints. I am still against liberal application of expansion joints for all the other reasons you see posted.

C

 

[KernOily]

Maintenance is never the first reason to install expansion joints. As far as removing the pump or piping to ease maintenance, there are other options. For example, a drop-out spool in the line, stainless steel flex hoses, or a rubber (elastomeric) expansion joint (e.g. Garlock) in the suction only, to name a few. A well-experienced piping designer can help you with this. The reason we install expansion joints at equipment nozzles/flanges is to alleviate a load that is applied to the flange that is in excess of the manufacturer's allowable load.

Expansion joints must be properly designed, specified, and installed or you will create more problems that you started with. As a pipe stress engineer and former plant engineer, I can say without hesitation: Been there, done that, and got the scars to prove it. There are some spectacular and tragic plant accidents that were caused by failure of a metal bellows expansion joint.

Here are some considerations:

Firstly, do NOT install ANY metal bellows expansion joint without a complete stress analysis of the system to ensure that you have accounted for all loads (forces and moments) that will be applied to the joint. There are several types of joints out there. The correct one must be selected to address the situation at hand.

Secondly, as Art said, the joint WILL fail at some point in its life. I believe the design life for most metal bellows joints is 2500 cycles, or something like that. Since it will fail, you must account for that in your plant MOC and maintenance management efforts and this in itself is not a trivial issue. Will you still personally be there when it's time to inspect and replace the joint? Will anyone remember? Is your CMMS and MOC prorcedure good enough to ensure this issue doesn't "fall through the cracks"? In my experience, most plants fall pretty far short in this department.

Since you asked, my two cents is - stay away from these things unless you have first done the stress analysis and can find no other better solution. Pete

Thanks!
Pete