Should Butterfly Valve Stems be Heat Treated?

[BronYrAur]

I have an industrial waste water filter system with several 8" and 10" control butterfly valves that cycle filters on/off, and go into backwash, etc.

It's a new installation and have already had 3 valve stems snap. They don't appear to have been twisted or pulled apart - just snapped. See attached photo.

The end user happens to have a metallurgist on staff who examined one of the stems. I don't have an official report back, but I am told that the stem was never heat treated.

Should a butterfly valve stem be heat treated?

 

[stanier]

The stem may indeed have been heat treated incorrectly!. If the grade of steel is such that it can be hardened and tempered by heat treating then the tempering may have been carried out incorrectly.

The heat treating aspect may be irrelevant. the stem has failed and the manufacturer should be contacted for an explanation. This type of application is very common place for butterfly valves. I have never come across a backwash valve stem failing in 35 years in the industry.

“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King

http://waterhammer.hopout.com.au/

 

[gerhardl]

Again I agree with the sensible answers from stanier.

When something breaks it is too weak for the forces applied. This is also implied in the 'heat treated or not' discussion. This is however only relevant if such stems from this factory have a construction norm that says such stems should be from heat treated materieal, and thus metallurgically a bit stronger. If this is the case, manufacturing fault would be part of the explanation: material not correct after factory norm or order/description.

In my opinion it is more likely that one (or more) of the following (common) causes have contributed or caused the failure:

a) Actuator, gear etc applying to strong force for the stem construction
b) End stops, level switches, torque switches: missing, wrongly or not connected or not adjusted properly
c) Other regulating devices or process control placed, adjusted or selected in such a way that signals for open and closing are (unnecessary) giving a too high repetitive numbers of openings and closings.
d) Mechanical connections have slacks or misfits, or actuating forces applied in such a way that slamming forces and shocks are given when starting and stopping. (see also point a )
e) Media or process conditions causes valve to stick. (Eg. valve sealing or valve type possibly not optimal for application) (see also point a) )
f) Valve or valve construction (including stem diameter or material) in general to weak. Eg. actual forces underestimated from user or supplier. (See also point a) and e) ).

Under the circumstances and from your picture it seems unlikely that only a hardening of material will give you a lasting result, the other points have to be checked.

What is the stem material? Possible to change to stronger material and/or increase diameter?

My recommendation is to change the valves to a general stronger construction and use double double offset valves if not anything else improves the situation. Might knife-gated valves or gatevalves (depending of fluid and dry material content) be a better choice?

 

[bcd]

Butterfly valve stems are highly stressed and need to be made from strong material. That is a fact. How strong depends upon how the valve is designed. Is it a one piece shaft or two pieces shaft. How much clearance there is for the shaft to bend within the waterway, manufacturing clearances, etc. It is the combined bending from pressure across the disc plus torsion for the torque to opearate the valve which determines how strong the shaft material needs to be.

Looking at the picture, you have a fracture surface which is perpendicular to the main axis of the shaft. Torsional failures will have a 45 degree plane relative to this axis. So it is not primarily a torsion failure. The failure also appears to be at an area where the shaft diameter changes (stress concentration). Since the failure plane is perpendicular to the axis, and there is also a stress concentration, I am thinking this may be a fatigue failure. Any misalignments and repetative stress would cause a fatigue failure, which is fairly common in butterfly valve shafts. The tell tale signs of fatigue such as beach marks are obliterated by the surface damage caused by the surfaces rubbing together after the failure.



If this valve has an actuator, I suggest looking very closely at the alignment of the valve and actuator shafts. Any misalignment causes the valve shaft to bend eccentrically every time the valve cycles and causes failures that look just like this one.

 

[hsbcn]

What are the valve specifications and process conditions? Of particular interest: valve pressure class and MAST (even if it's estimated); and on the other actuation type and torques applied. Though butterfly valve stems are the most delicate of all stems in the valve range, they're not t h a t easy to snap. Lack of (or wrong) heat treatment could be the culprit, but you also need to check the stem material vs. the pressure/media/temperature key in your process. You might have great stems, and lousy acutators. In sum: take a holistic view of the situation, don't just focus on what's broken.

If you have a flat tire, it might well be that the tire's worn, but you might simply have driving over a huge nail!

My focus is on valve stockists in the us and uk who supply duplex and super duplex valves, and inconel and hastelloy valves, plus titanium, alloy 20 and 904L valves.