Dear Friends,

For untied rubber bellows, which typical values of axial,lateral & torsional stiffnesses do you use?

regards,

sam

Your best bet here is to contact a vendor, or obtain the values from a catalog.

Rick is right. I experienced having trouble guessing these figures to lower loads on equipment connection. When I finally realized we are using Garlock, I contacted somebody from their Technical Department and she mailed me hard copies of these figures including stiffnes in torsion, exact I.D., etc.

Gimini55

Dears,

Regarding rubber bellows, one of our known bellow vendors told that like pressure drop, the axial/lateral stiffness of rubber bellow is negligible. But, regarding torsion, he told to avoid such loading on rubber bellow. Does it mean reasonably high torsional stiffness ?

If anyone knows about these stiffness values & share, we, the CAESAR II users will learn & have a harmonized approach for rubber bellow application.

regards,

sam

I would be highly dubious of the claim that a rubber bellow has "negligible" stiffness values for axial and lateral loadings. I'd see that for a teflon joint, but a typical rubber joint is a multiply, reinforced element and you can easily get a few hundred pounds per inch for stiffness. Obviously, it's not 1E12, but I wouldn't treat is as zero either, especially for something sensitive like a pump/compressor nozzle.

If torsional stiffness value is neither zero, nor 10^12 ft-lb/deg, what it should be for, say, 10" NB tied rubber bellow of design pressure 4 barg & design temperature 60 C in N.m/rad - Some typical figure, at least, to start with ?

Regarding the veracity of data from some vendors , I am giving a real example I faced many years ago. For a suction strainer of fabricated body in the suction piping of boiler feed pumps, I asked for the weight from the vendor. The vendor gave the weight as 25 kg, the weight of the strainer element, while the weight of the body was 400 kg, later found out in strainer GA drawing while we discovered nearby spring hanger bottoming out in pump's start-up walk-down. I had to replace the spring hanger at once for higher rating. My fault was my inexperience not to doubt & check the data with some one else experienced enough. I learnt from that mistake. For each input data from vendor, one should have some benchmark figure to compare with.

In today's environment, deep domain knowledge without proper experience & interaction with vendors sometimes create similar problems.

regards,

sam

Sam,

I'll find from my file what you may need and if I can find it you could send me one of your email addresses and I can send you stiffness of rubber bellows from Garlock I used 4 years ago. There was a name of a lady supervisor too I had talked to and she was the one who send me the data and you may talk to her too. Hopefully the size you need is in there. Let me know.

Ed-Lamigo

Thanks to you, Ed-Lamigo, for your response.
My email address is sambhu_snb@rediffmail.com.
For 10" NB tied rubber bellow of design pressure 4 barg & design temperature 60 C, I need typical values of lateral & torsional stiffness in N/m & N.m/rad respectively.

regards,

sam

Sam,

I already scanned-to-email the info you may need from Garlock. If you need more info, contact that Product Specialist.

Ed-Lamigo

We often get asked by clients to model bellows, when they have no idea as to what they will be buying, and typically use the values from the Pathway Catalogue. Many other bellow catalogues don't seem to give values for axial, lateral, angular and torsional stiffness, which this one does.

We also try a range of bellow stiffness values to see how the system is behaving. If you have no information re what bellow is to be used, it would be a good idea to make it very clear in your stress report to your client that the results are based on assumed values.

Dear Madam,
You are right in approach that we make some assumptions when we have no firm data,which we should clearly mention in stress report, say in Caesar-II title page.

But we, as responsible engineers, are accountable for resonableness of these assumptions to our clients. In the case of engineered solution with bellows, bellow stiffness properties are important.

Fortunately, in the realm of metallic bellows, EJMA is there & there is uniformity of approach among major bellow vendors. In Caesar-II itself, we have a few bellow catalogue data. Bellow vendors are generally co-operative in finding out a right solution.

In case of rubber bellows, we expect same co-operations from bellow vendors

regards,

sam

Sam,

The one i sent you were untied (no tie rods) and you can still contact Garlock if you need a maybe recent figures if they have. For tied as I just noticed in your later post, I remembered using 100,000 lbs/inch for tension (compression should be of course for the rubber) and for lateral, I used 2,000 lbs/inch. I believe the manufacturer cannot give you this figure same wasy as for torsional so make assumptions as experts here are suggesting. Consider FEA for exact figure to determine exact stiffness. One thing, in my previous experience, they use bellows to isolate pumps to the piping system to avoid vibration transfer. If your problem is to lower loads on pumps nozzle, it is always best to get first pump's allowable loads and moments and then piping in the inlet and outlet could be designed flexible enough to reach the allowable loads and moments. I saw tied bellows used on the discharge side of the side-discharge pump and to me is meant purposely for lateral deflection (to absorb piping system load that creates lateral force components and moments to your nozzles. In the later years, working with much expert stress engineers I have worked on system without this bellows and infact hotter and with much higher pressure than those I worked earlier of my experience. Hope to keep this thread rolling.

Ed-Lamigo.

Dear,

Thanks to you, Ed-Lamigo, for your esponse.

Please resend the scanned-to-email to my address sambhu_snb@rediffmail.com. I still didn't receive it.

You are right in your approach that bellows are not a preferred solution in many cases, especially with high pressure in piping. But, in case of vibration isolation or very large size piping operating under relatively low pressure where it is difficult to provide a number of elbows/bends in piping, bellows provide some acceptable solution.

regards,

sam
For tied bellow, we provide high stiffness both in tension and compression, but you referred to keep the bellow unrestrained in compression. Such will be the case if you provide some gap at the inner side nut of tie rod, or provide no nut in the inner side.

Dear SAM,
Are you using the bellow for isolation of vibration? I think for vibration isolation bellow should not be used because of limitations in number of life cycles.

I missed the "h" in sambhu. Sorry Sam, just came back and I'll email it one more time.

Ed-Lamigo

Thanks to You, Ed-Lamigo.

In response to Manoj Sarkar's email:

We use rubber bellows for vibration isolation. This discussion thread started with the rubber bellow stiffness data. It is true that metallic bellows should not be used for vibration isolation.

regards,

sam

Dear Sam

If we allow tied bellow with no nuts in inner side. I feel it will allow compression at the same time it will act as un-tied bellow i.e. it will not take care of pressure thrust which is main reason to use tied bellow instead of un-tied bellow.

Habib

"The joint will only contract when the piping load overcomes the pressure thrust, which will disengage the limit rods.

As far as the pressure thrust on the nozzle goes, the flange should only see the thrust on the bellows area that is outside of the pipe ID. The pressure thrust inside the pipe ID area goes into the pump casing and the impeller.

You will need a pretty stout anchor downstream of the joint as it will need to carry the full pressure thrust of the effective diameter of the joint. "

This I copied for Habib's reference from the replies made by Kavin Monroe & Edward L. Klein in April 16-17,2002 in this discussion forum.

Thus, tied bellow with no nuts in inner side, will still be a tied bellow, so long a positive gauge pressure exists within piping & piping load has not overcome pressure thrust fully.

regards,

sam