Hastaloy C-276 bellows failure 3

[Alwynb]

Hi there,

We recently had a failure of a bellows in a control valve. The chemical involved is a in the HF acid family and acc to my knowledge, Hast. C-276 is OK.

The question is whether this material is a good bellows material and whether it is more suceptable to fatigue. The valve was in operatrion for 1 month but the valve PID settings were incorrect and thus it opened and closed quite alot. I estimate about 50,000 cycles was done.

Any comments would be appreciated.
Regards
Alwyn

 

[hacksaw]

Check you valve maintenace manual.

Bellows are typically high stress depending on the amount of extension.

One manual suggest that its bellows are rated for 10,000 cycles at full stroke, 100,000 cycles at 1/2 stroke, and 1,000,000 cycles at 1/4 stroke.

Appears that your valve had greater than nomal life expectancy. They also indicate that TFE bellows have a much greater life. That does not mean that TFE is suited to your case. The valve manf. will have to advise you about their design.

 

[DeanA]

The valve supplier should have a design criteria for the number of cycles. If you are requiring extensive cycles then advise the supplier and they should be able to change the design of the bellows to accomodate additional cycles. This could be by using an extended bellows and therefore only using a small amount of the available trave and restricting the fatigue life of the unit.

 

[bljnv]

From my experience 50000 cycles is excess than the designed cycle life of bellows. You can ask the bellows supplier for the design life expectancy.
Inconel 625 can be another alternative if your service accepts.
(Normally the bellows would be designed for 10000 to 15000 cycles).

 

[iainuts]

I believe the Hastaloy should be very good for the HF application (chemical compatibility is good).

Rather than ask the vendor what life the bellows is good for, I'd suggest telling them you want some high cycle life. It's not impossible to make it infinite, the stresses simply have to be low enough. For a bellows that fails after 10,000 cycles, the bellows is actually yielding very slightly with every stroke! I think it's sad that manufacturers actually get away with this, allowing yielding.

I've spec'ed bellows for valves and told the manufacturer to make it infinite cycle life, and they simply keep the stresses down. Cost hasn't been a big factor. A welded bellows assembly as opposed to a hydroformed one is generally a great improvement in life as the stresses can be kept down while the assembly stays in the same dimensional envelope. I suspect any bellows assy not getting around 100,000 cycles or more is probably undergoing some minor amount of yielding with each stroke.

 

[bljnv]

EJMA standard will provide more details on estimating the design cycle life. I think this standard refers to convoluted (hydro formed ) bellows.
There are suppliers not vendors who do design according to given specs and they can help to estimate the cycle life of bellows.