Choke Valve:: Tungsten Carbide Tip Broken 1

[bmoorthy]

Recently I came across Choke valve Failures. The valves were in service for less than 1 year. In 2 cases the hand wheels were rotating freely, so the valves were opened and it was observed that the Stem tip made of Solid (HIP) Tungsten Carbide had broken. The tungsten carbide was attached to 410 stem by silver brazing. The brazing had not failed, the WC tip had broken.
Query (1) :: The only opportunity for any load to directly act on tungsten carbide (In form of torsion or shear ) is when the Choke is closed fully. Is it OK to close the Choke valve fully and tighten such that it almost becomes isolation valve. Why should one close the Choke? is the use of Choke not just for pressure reduction, is it also common to close the choke frequently?

Query (2) :: What is the Maximum allowable shear stress and maximum allowable torsion (Something like MAST) for HIP tungsten carbide? Is it necessary to conduct Wake Frequency calculations (as per PTC 19.3) for Choke Stems?

Query (3) :: Although Tungsten Carbide is widely used in Choke valve industry, it is observed that NACE MR0175/ISO 15156-3 does not list this in its table. So it is OK to use tungsten carbide without qualification testing for Stem tip (noting that it is a non pressure component)

 

[MikeHalloran]

The only form of WC with which I am at all familiar, and I'm no expert, is a sintered powder metal part, bonded with some other metal whose name I forget.

I can't understand what benefit the HIP process would bring to a WC part.

Somebody please help me understand the many ways in which I must be wrong.





Mike Halloran
Pembroke Pines, FL, USA

 

[gerhardl]

bmoorty:

You seem to presume that mechanicaal force only is the cause of the damage. How does the surface of the broken tip look? Could the cause be cavitation?

If yes you will have to look at dimensioning, flow and pressure (delta P), type of valve and use/practise/throtteling degree.

 

[bmoorthy]

Even in the event of extreme throtelling, the stem tip is not supposed to have any pressure acting on it since the valve is open. Only load that acts on the Stem tip is the flow impingement

That is why i was wondering about the applicability of Wake frequency calculation.

I did see some manuals (Weir and CCI) that suggested Tungsten Carbide is not suitable for cavitating service.
Is it not incumbent on the Choke manufacturer to know that the Well will cavitate and design the valves to take care of it.
The surface does not appear to have ductile failure ridges (Using A 370 pictures for comparision on the type of failure).
It is smooth and uneven with small (Only a couple of them) flake at the edges.
It seems that HIP WC had some micro cracks (Since in HIP presence of cavities are extremely rare due to the very process of HIP) and these cracks may have propogated, but the question is, how can a non pressure component that get subjected only to flow induced vibration (no other load whatsoever) break?
So it was a inferred from the nature of discovery of failure, that someone attempted to close the choke (When it was already closed and broken) and the handwheel was freely rotating when attempted to close it further.
API 6A does not suggest any NDE for the Stem tip, so what triggerred the failure. Is it the choice of the material itself?

 

[gerhardl]

bmoorty:
Could you possibly post a picture of the fracture?

 

[bmoorthy]

Photograph of broken Tip attached.

 

[gerhardl]

bmoorty,

seeing the picture I have to completely agree with you. This is no cavitation break, and seems to be caused by purely mechanical causes, perhaps aided by partly mechanical weaknesses.

To your original questions.

1) A choke, or regulating valve,is normally not used, (if not built for it!), as closing or isolation valve alon. Some chokevalves are built for complete thightness and closure. All choke valves might under operation go to fully closed position, and should not normally be caused to fail by this position.

2) The cause of breaking seems to be (as you implicates) too high shearforces for the construction, probably by turning the cone down into the seat, thus locking the cone, and then turning the stem further until break starting from rim of metal boss (heating/crimping when soldered? hairline scratches? other?). The 'swirl' of the break implies this.

Any normal closing of this type of valve would in my opinion not require a closing forche high enough to break the material in such a way. The distance between normal closing torque and torque high enoogh to give this type of damage is in my opinion 'high'.

The cause is then simply false operation forcing way past normal torque. Difficult perhaps to control by figures, or best described as limited to 'normal' handforce. Prevention: torque or movement limitation. If hand operated: mechanical limitation or correctly set limit switches on gearbox. If no gearbox: mount one.

 

[bmoorthy]

gerhardl
Thanks, your input reinforces my thoughts.

 

[jbeckhou]

Hello Bmorrthy

The choke shown in the picture is an H2 style choke. This type of choke isn't designed to shut off flow, the choke is designed to decrease pressure in fluid flow situation, as a result, the velocity of the fluid increases. Abrasive particles in the high velocity flow can cause excessive and premature erosion.

You will also damage the needle and seat if the valve is forced closed. The stem material and carbide tip are designed to NACE and API standards. The tip is considered "trim" and can be easily replaced. The carbide tip could have been damaged prior to use. After the tip is attached to the stem, the whole stem and carbide tip should have NDE preformed, such as LP. This is a requirement of API.



Petrotrim Services

http://www.petrotrim.com

 

[gerhardl]

jbeckhou:

Even though the choke is not made to shut off the flow, any chokevalve I know of have a seat and a mechanical movement making the cone/closing member reach the seat and thighten against the seat sealing. If this movement is not limited in any way (torque or mechanical limitation), my point is that normal mechanical material stresses could easily be higher than allowed.

I see no abration or cavitation wear on the attached picture. A question mark for earlier weakened material along rim of metal boss as mentioned above, and as a possible cause as you mention.

I read LP as liquid penetration, and LP alone could possibly give an unclear indication if a small hairline crack were hidden exactly at the rim of the joint metal/tungsten.

Anyway, in my opinion, it will require a considerable twisting 'overforce' to break the material as given.

As an afterthought: no very high temperature differences of wetted surfaces by fluid or process (out of allowed limits) to give excessive crimping/expanding forces for the two joined materials?