Double block & bleed: understanding a barrier – part 6
As illustrated in this series of articles, there are many ways of creating double block & bleed or double isolation and bleed. But in this name you need to ha a depressurized volume between the two barriers and the name itself makes a problem. If taking a double piston ball valve like the one illustrated in figure 27, with 100 % inlet pressure 50 % pressure in the cavity and fully depressurized downstream side, you do have a valve with two barrier, but you don’t have a depressurized volume between the two seats.
Looking at figure 28 it´s even worse. You do have two pressure-activated seats where the seal force is increased compared to the seal force in the figure 27, but you do certainly not have a bleed in-between the seats. As explained earlier, when taking the cavity pressure away you are taking the seal force away from the down stream seat, and you can end up with a leaky valve.
To explain: If taking a medium large valve like a 16” class 900 as an example, to explain what can happen if the seats are locked in the seat pockets due to high friction. The areas exposed to a differential pressure on a 16” valve are around 1250 cm2. Looking at figure 29 the valve is in closed position with equal pressure on inlet, cavity and outlet, the ball is in total balance with equal force acting on all sides as indicated with the red arrows. Let us use 120 bar as the system pressure in this class 900 valve.
What happens when we depressurized the outlet of the valve, shown in figure 30? When taking away the outlet pressure, the pressure in the cavity of 120 bar will act on the right side of the inside of the ball with a force equal to 150 tons (1250 cm2 x 120 bar = 150 000 kp) this will make the outlet part of the ball oval and force the outlet seat slightly downwards, that’s ok as the seats are floating. When depressurize the cavity as indicated in figure 31, the force towards the outlet side of the ball are taken away and the ball gets back to its round shape. The floating seat are preloaded with coil springs, but the force of the springs are only fraction of the force from the differential pressure and if the friction between the seat and the seat pocket are to high, the seat are stuck and in bad contact with the ball. You have created a small leak path on the outlet seat, but you still don´t know that.
Now, what happens on the inlet side of the ball? The inlet side was in perfect balance until the cavity was depressurized, but when taking away the support pressure from the inside of the ball bore as illustrated in figure 31, the inlet pressure will now flatten the ball and force it slightly towards the down stream side away from the inlet seat. As long as the seats floats it´s ok the inlet seat seals. But if the seat is stuck due to friction in the seat pocket you have created a small leak path. The inlet seat starts to leak and as long as there are lack of contact between the outlet seat and the ball the outlet seat will leak too. You do now have a leaky valve and a problem!
This can be a difficult problem to solve, and if you are without lubrication fitting to the seats you may have to disassemble the valve. As everybody working with valve design knows it is challenging to design a valve. There are a lot of aspects to consider; dimension – material quality – the flexibility of the material – soft seals (O-ring or lip seal) – tolerances between moving surfaces – hard facing – lubrication to mention some of the reasons for headaches. What happens with the valve when it is put into service? If having a 16” valve with lip seal as the radial seals on the seats, having a 0,3 mm clearance between the seat and the seat pocket you don´t need much fantasy to understand that it is easy to get problem if you get a high temperature dry gas with sediments clogging up between the seat and the seat pocket or if the round seat are getting slightly oval. One thing is to produce the valves, another thing is to use them in the correct way.
I will end this series with a short story when everything was done in accordance with the company policy of double block and bleed.
It all started with a small hydrocarbon leak out the spindle on a 2” class 600 wedge gate valve. I was phoned and asked what to do? My answer: Put the valve in back seat position, test it and if the valve seals tighten or change the stem packings. The valve was placed in back seat position and sealed perfectly. The solution to change the packings under pressure was presented the platform management and the answer was; no you don’t have a double block and bleed. To change the spindle packings on the valve there had to be double block & bleed in all direction. A plan was set up, and all the valves to be closed and drains to be drained was marked up on the P&ID. How many valves that was involved with this job I don´t know, but I was told there were a lot, and they had to shut down the production on the platform. The job preparing to change spindle packings on this 2” valve took more then 12 hours with total shut down, and the time changing the spindle packings took 30 minutes. In my head this is total madness and lack of common sense.
The solution of this total lack of knowledge is training, valve training and valve understanding and maintenance procedures made up to fit the valves construction and needs. Lack of knowledge is not SAFE.
Good luck with your valves.
Vil du laste ned denne artikkelen som PDF?
Klikk her for å laste ned