double acting actuator concept w/r pulling force

[Danlap]

Good day fellow valves professionals,

May sound like rookie question, but does someone have any idea or a better way to explain for double acting valve concept in terms of how it remains “locked and back-seated”?
Below photo is the remaining of spindle for 8”-Class 1500# high rate valve. In the event of overpressure, this valve should be opened (and backseated) from fully close position within 3 seconds

Aside from sulfide stress cracking contributing factor, I am wondering whether constant pulling load against back seat is also a factor. Or is the impact&pulling load is one time (fraction of seconds) only, after the positioner “see” valve is close, the hydraulic stop its attempt to move and pressure remains
Not about the subjected valve, but the design more is more or less similar like this.

Why it seems I am lazy:
-This broken stem valve is located 250 meter above ground
-The powerpack is located on the ground underneath the valve. Its pressure is 200 bar
-Plant was built in the 70’s with so many missing or unclear information / schematic. Even Indiana Jones Holy Grail map is better
-Control room is like 500 meters away. This where the on/off button is located
-And only two tubing in/out the actuator. To trace what lies in between the actuator and powerpack or the positioner to control room is quite a challenge. I can see some double block and regulating configuration on the tubing part to separate the actuator shall required

Thank you for the insight.
Kind regards,
MR

https://nosuchvalve.com

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected

http://www.eng-tips.com/faqs.cfm?&rat1=2&fid=376

 

[gerhardl]

From your description of the situation you seem to search for several tips.

1. Reason for spindle breaking in addition to cracks.
Unclear 'what is what' in pictures but several factors might contribute. Most likely excessive forces over time. Could be wear, support/bearings too slim (to long and firm linked together parts giving skew loads), not correct adjusted or moved end-stops.
The stop impulse at 3 seconds will probably give greater stop forces on valve than a normal spindel is dimensioned for.

2. Requirement and equipment for holding force.
Try to determine from the process what the safety requirement from this operation is today and how critical this is for the total plant operation. Safety would today normally be resolved by redundant solenoids if necessary.

3. Suggestions for improvement:
Support for actuator/piston. Loose link/fork between spindle top and cylinder piston to not give skew loads. Very precise adjustment of stroke.

.......... something else you are looking for?

Best regards

 

[Danlap]

Dear gerhardl,

thank you for the explanation. As usual, it is insightful.
point 2 and 3 are noted. We've missed to consider this
point 1, correct me if I am wrong "excessive forces over time...give greater stop forces on..". So once valve reach its (adjusted or non-adjusted) ends stops, the actuator will stop "pulling" the back seat stem against back-seat body?? despite the power-pack remains live
not so good technical drawing, but the the first picture above is about about the spindle's link with the wedge. What we did at the moment is make the "shoulder" with radius and make sure the material heat treatment was correct and hardness was assured. It breaks on location indicated by the blue arrow. Due to time constraint of 3 days to fabricate new spindle, we were abruptly decide to minimize the stress by make it radius and the stronger the better (re-heat treatment the spindle) and measure its hardness

Thanks again,
Kind regards,
MR

https://nosuchvalve.com

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected

http://www.eng-tips.com/faqs.cfm?&rat1=2&fid=376

 

[LittleInch]

To open a valve in three seconds including back seat is just asking for trouble.

Back seating is a concept that is often misunderstood, but in normal operation your valve operation should stop just short of the back seat contact. Back seat cannot and should not be used on line, and is a bit old fashioned IMHO.

This broken stem valve is located 250 meter above ground

what!!!! That's not far off the Eiffel tower?

I'm, not really sure what your question or point is though.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Danlap]

Thanks for the remarks LittleInch,

My excuse for false information, the valve is located 110 meter above ground. Now, I've check that against the civil isometric. Sorry for the exaggeration, maybe because I was using the stairs.
My point of question is to weigh whether the initiating crack was due to sulfide stress cracking (which is already confirmed btw); impact upon valve was back-seated (happened once in every 3 years and during emergency); and or continuous pulling force (this is my question). In addition to gerhardl's remarks.

For the last 30 years, on the same valve, similar incident had occurred three times. And we just changed the spindle material with "stronger" ones without fully addressing and understanding all the possibilities.
Hard to described, but seeing how operator on field and in the control room communicates, and the only activity done by the field operator is by opening the spindle's "interlock" (mechanically stop the spindle from accidentally closed), I am starting to think that indeed this system is old fashioned.

Regards,
MR

https://nosuchvalve.com

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected

http://www.eng-tips.com/faqs.cfm?&rat1=2&fid=376

 

[bcd]

Looking at the fracture surface and drawing of the part where the failure point is noted, my first thought was bending fatigue failure. For one, the fracture surface is flat, which is consistent with bending fatigue. Also, you have a fairly rapid change in diameter where the fracture occurred (i.e., stress concentration). Any misalignment between the actuator and valve stem can cause bending stress on the shaft. If the valve is installed horizontal as is shown in the picture, it is highly likely that if anyone removed and reinstalled the actuator, there would be misalignment causing bending stress. The bending stress would be additive to the tensile and compressive stress every time the valve operates.

 

[gerhardl]

When a mechanical part break down in the same way, under the same use, and at regular times this is obviously a systematic fault cased by too weak components.

Some points, not necessarily the correct ones:

a) I believe that you could somewhat improve the performance by altering the spindle in addition to other improvements (support and non-stiff linking). Also have a new spindle ready to be changed as a prevention before normal breaking time.

Why is the spindle hardened, would a softer material be better? Other material (titanium)? Why is the spindle given a (weakening) grove, or do you mean a radius positive up from the shaft diameter and not down. In that case I agree.

b) If you measure I doubt that you really shut down in three seconds. Is your requirement really 3 seconds? With a longer periode you could better adjust and control the movement, or perhaps select other type equipment

c) The valve itself it not mentioned. Is it OK without wear and closing tight? The problem could be caused by wrong selectded/constructed valve solution.

d) After the repair now I suggest you analyze the whole safety function with possibly solutions at todays technology, for a better future solution.

Good luck!

 

[Danlap]

Good day All,

Thank you gearhardl. bcd, LittleInch for the input. It has provided my team with extra views on how to assess this case. Again excuse for some sloppy remarks mentioned earlier (height and hardness), a bit of the summary below.
As much as we (engineering) team would like a new hydraulic and valve system, unfortunately we should make the best of current situation.

Subject of failure: spindle of 12”-1500# Gate valve (blue circle below)
General protocol:
If there is an indication shall vessel (V1104) will be over-pressurized>it will trigger the alarm>control room personnel push the button manually to open UZ456 and UZ457 (motion of operator from hearing the alarm to touch the button whether it is 5 seconds or 30 seconds is somehow bias)>the control valve in between, its function is to control how much excess pressure should be passed through the flare
Not clear: opening/closing time requirement

Measured and witnessed (by several accounts):
-Hydraulic actuator pressure 180 bar
-Opening/Closing time: 3-4 seconds (video is too big to be uploaded)
-Refer to 1990 drawing, stroke length of valve is 275 mm and stroke length of actuator is 290 mm. Opening’ stopper exceeds by 15 mm thus valve backseated??
-After control room personnel push the button, it took operation several minutes to hours before sending the operator using the stairs 110 m above to manually “lock” the 180 bar pressure within red circle compartment and preventing the valve from “accidentally” closed
-Commissioned 4 years ago, the opening/closing time was 12 seconds.
-Failed spindle is 17-4PH material with hardness 33HRC. Now we apply 17-4PH with 29HRC which is less prone for sulfide stress corrosion

Summary:
-This hydraulic system and valve is indeed old configuration
-There is reliability issue with the regulating valves for the hydraulic system. 12 seconds vs 3-4 seconds??
-Open = backseated
-Remain pulled towards backseat position for xx minutes or hours (until operator lock the actuator by turning the block valve whilst at the same time lock the stem movement)

Hypothesis:
Crack was initiated due to impact upon opening (and maybe also closing)>location of crack located in the narrowest and “sharp” area of the spindle>33HRC is more prone shall compared to 29HRC for sulfide corrosion cracking>again one more time valve opening> due to constant pulling force for xx minutes the stress crack grew bigger and spindle snap> closing the valve>one more time valve should be open, however now the wedge remains in close position due to already dis-engaged with the spindle

Thank you all for sharing your experiences and fine tuning the theory.
Kind regards,
MR

https://nosuchvalve.com

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected

http://www.eng-tips.com/faqs.cfm?&rat1=2&fid=376