Ball valve stem sizing 1

[Wintechcad]

Hi,

Does anyone know of a calculation or 'rule of thumb' process for determining the breakaway torque of a ball valve? I am designing a class 1500 ball valve from scratch and am using ASME B16.34 2009 for body bolts, wall thickness and general design, but there seems to be nothing on stem sizing. I need to know what torque I will be applying to the stem in order to determine the stem diameter.

I believe it must be a function of the presure acting on the ball, thereby pushing it into the seat and the resultant rotational friction to overcome this force.

Any help will be greatly appreciated.

 

[gerhardl]

Hi Wintechad,

I suggest you take a look at the previous threads below. They will give you some of your answer.

Perhaps you could use this to 'narrow down' and perhaps detail your question a bit more?

Thread408-157477 thread408-253314

 

[rneill]

You didn't mention if this is a floating or trunnion mounted ball but at 1500 it is probably trunnion.

The trunnion will hold the ball in position and the seat loads should be relatively independent of system pressure providing the steam/shaft is adequately sized to resist deflection. I think most of the friction will be dependent upon the design characteristics of your seating and the initial loading you've designed in.

 

[Wintechcad]

Thanks for the replies.

The ball valve would be floating as it is in the size range up to 2". The posts indicated seem to show torque values for existing ball valves from different manufacturers. I am trying to determine how the stem is sized when starting with a blank sheet. My experience as a newbie to the valve industry (originally a burner designer) is that a lot of data is merely copied from leading manufacturers and I myself had problems when designing burners with eroneous valve Cv values etc. from some of the smaller manufacturers.

I assume that there must be a way to size the stem given the pressure (B16.5 ratings) and seat material. Being a floating ball valve there may be an element of pre-loading (clamping of seats), but this would increase significantly when 3000psi pressure is applied. I fail to see how a valve can be designed without firstly determining the expected torque to be overcome by the stem.

May be there is a ball valve 'bible' which gives the answers, but I am yet to find it.

 

[gerhardl]

I am no constructor, but good quality valves will usually have oversized stems.

The reason is the battering a valve can get, and the trouble experienced by changing if damaged, even if this has happened during extreme/not normal conditions.

Example. We sold 20 years ago thousands of 2" flanged SS ball valves, the whole package with firesafe design and anti-static device, for the North- Sea concrete oil-platforms. The valves were used for concrete transport under the platform casting procedure, used twice, and ending up concrete-filled as part of the conrete wall armoring. Torque after first-time use? Up to ten times normal?

In addition to torque comes possibillity of water-hammer, solidity for stem sealing constructions (air emission), stability if tested firesafe (deformation at high temperatures), excessive force torque operations, and general stability of stem bearings (sidepressure decreasing with larger stem diameter) etc.

The point is that 'the safety factor' should be several times more than normal for a mechanical construction for the stem, and the evaluation of this factor for your target market segment is more important than finding the exact 'normal' torque figure, even if you have to start somewhere.

 

[Wintechcad]

I understand what you are saying gerhardl, but I still need to be able to calculate the normal torque to which I can apply a safety factor.

My question is still, how do you calculate the expected torque in order to correctly size the stem when designing from scratch?

 

[rneill]

With regards to gerhardl's comments, it would not be uncommon to get sand or grit between the seats and the ball which would cause additional friction and may end up driving the torque values up much higher.

The worst case might be a valve with a large oversized actuator that slams the valve shut. The stem might see significant load after full travel but before the power to the actuator is released - the actuator would be turning the valve against a stop. What's the maximum torque that an actuator can apply?

I have a number of books on valves but they are all from the end user perspective and none get into the design details for things like the stem.

 

[unclesyd]

I just talked to a person involved with a ball valve problem we had with class 300 6" ball valves. The MOC of the valves was 316 SS with Teflon Trim. After the manual operated valves were installed it was decided that some valves needed mechanical operators. This turned out to be a disaster as the existing 316 SS stems could carry the torque. All the stems twisted and valves became inoperable. We took some of the valves to test under pressure using water and using stems made from D2. On three valves the actual torque to operate the valves was determined. From this we were were able to determine that in the present valve configuration it was going to take a little increase it strenght for the actuator to work . We were able to use some Carpenter 7-Mo we had on hand for the stems. The manufacturer did change the diameter of the shaft which brought on some other problems with the packing.
We had essentially the same experience on some 3" meat seated ball valves that operated at 600F,

He stated that as far as his part in resolving the problem he didn't do any calculations or remember anyone else doing any. He also stated that he can't recall seeing any calculations from the manufacturer.

What you might do is look at your handle with say a 100 ft lbs input. I believe a good operator would stop and ascertain the reason for a 2" valve not operating with this much effort. The 100 ft lbs is a SWAG as most small valves operate with a handwheel instead of a lever now and it would be hard to get that much input on a handwheel. But as your valve is a Class 1500 one would expect it to require a high torque to operate.

 

[thought2007]

Wintechcad,

To find out max torque requirement to actuate the ball is completely iterative process.I have done these calculation by considereing max breakaway toraue can be expected when the valve is in closed condition.Its power loss due to friction between kinematic elements like Ball & seat,If its trunion type arrangement pressure force which is applying on closed surface of ball will be submitted into bearings ie. friction loss between bearing & stem & one more parameter which drives is dyanamic seal which is used to control leakage.To summerize three components drives max torque requirement

1.Frictional loss between Ball & seat
2.Frictional loss between Bearings
3.frctional loss between dynamic seals

Hope this will help you

John