How do we know the SOV is failure 1

[hazel]

There are some critical interlock loops in our plant.
They were designed by 2 of 2 for DO TYPE point, that means
the loop have 2 solenoid valves for a device to prevent shutdown the plant if a solenoid failed.
My question is how do we know the solenoid valve is dead since the other one still work. Except to touch the solenoid to check the coil, what can I do?

thanks!!

 

[buzzp]

You may be able to get away with just feeling the solenoid but I imagine there are cases where the solenoid will be hot but still not function (sticky valve or something). If there was no heat then the coil is opened up (generally how they fail). I assume the 'liquid' is not too hot so that it won't heat up the solenoid. If this is real critical, I think I would want to detect a bad solenoid by measuring the flow from its output port rather than determining if the coil is in contact. You can tell electrically if the coil is bad fairly easily (monitor the current flow through the coil).
I would look at flow meters or some similar device to check for flow out the output port rather than checking solenoid temperature or coil current since both could check fine but the valve could still stick.

 

[ilan7]

Think the best way is to have a magnetic switch next to the solenoid. As long as the coil is on the magnetic field wil activate the solenoid and the switch.
There are switches like that using at Magnetic Level switches, and switches used as suppliment to magnetic level indicators.
The other option is to have a position switch for the actuator the solenoid is operating or pressure or flow swich to check the flow or pressure the solenoid is causind when activated.

 

[hacksaw]

This is a common problem. It is dealt with (usually) by regular testing ans/or by monitoring the consequence of the valve actuation.

with turbines you block the trip valve and test the ability of the solenoid to unload the throttle valve actuator. with large control valve you monitor valve position or an appropriate process variable.

There are some SOV's that are fail safe. Actually the fall in the category of motorized valves, but the inlude valve position switches etc. These are used in the valve trains for fired heaters and burners (Maxon Valves etc.).

 

[mbrooksuk]

Following on from the earlier reply which mentioned testing of the loop, what we often do these days, assuming the SOV is on a shut-down valve circuit, is to design the loop with a valve partial closure test facility. This allows a technician to start to close the pipeline valve, but the circuit only allows the valve to close by say 10% before it goes to full open again. This has de-energised the SOV, but has not allowed the pipeline valve to close completely, thus ensuring no upset to the process. The test circuit can either use three limit switches [open/closed/90% open] or can use the Fisher DVC smart positioner. MTBE calculations for the loop, together with a Safety Integrity Level [SIL] review will determine how often the test has to be conducted. Obviously if you have two SOVs in circuit, this test can determine if either of the SOVs has failed.

 

[hazel]

I think Buzzp is right.
These critical loops do not allow to do online test.
It means that you can not close or open pipe(process) valve
during operation. That is why we design 2 of 2 logic solenoid valve interlock loops to minimize false shutdown.
So we have to monitor the 2 solenoid valves, make sure their functions are OK.
I will install the flow and pressure measurement between the
SOV. Flow one is to detect master SOV,and pressure is monitor redundant one.
Does anyone have such installation experience, please give me advices. Are these devices avaliable in instrument line?

 

[Scotsinst]

Asco and Triconex have testable systems with 2oo2D and 2oo3 solenoid valves respectively.
They are not cheap but are excellent if online testing is required.
The Asco system can be supplied with its own PLC, pressure switches and extra testing solenoid. I would check Asco's website for info.

 

[jsummerfield]

Consider the ASCO RCS testable solenoids.

John