Would this work to confirm a solenoid valve is not burned out?

[daba]

Sensing that a solenoid valve's coil is passing current seems a sensible way to prove that the coil is not "burnt out".

However, just because it is passing current, it doesn't mean that the solenoid valve is opening, which I'm guessing is the real need....

If it is critical to know that the valve has opened, then you need to provide feedback in some form. There really is no other way.

You can use pressure, temperature, or flow, as appropriate in the application.

 

[The Plc Kid]

Measuring the current when the coil should be on would be best. A Rogowski coil would be the most accurate at that low current. Lots of companies make these for small loads.

http://taehwatransus.com/products/rogowski-coil/rogowski-coils-for-smart-meters/

 

[Saffa]

Yes, I hadn't read the original post properly, my apologies.

And yes, this may actually kinda work. But it would definitely not be standard practice as far as I'm concerned.

 

[OkiePC]

And if you want to nip the problem in the bud, Mac Valve solenoids don't burn out.

Position sensing at the end of the drive train as suggested closes the whole mechanical loop and you can program an alarm such as "motion failure: extend CYL123", I am sure you are familiar with this, but if the problem caused by the failed coil is significant enough to warrant an added controls solution, for the same effort you can cast a bigger net.

 

[Bit_Bucket_07]

I had a request to try to add something to a system that would check if the solenoid on a valve is burned out. I suppose one way would be to add a current sensor to the line with a discrete output and monitor that.

Another thought I had was what is in the attached image. Any thoughts on whether this would work or not? I know the coil is 22watts cold so that takes about 1.9Amps. If I put a small relay in there, will it get burned up by this much current?

Your solution assumes that solenoid failure = open circuit. It is more common, in my experience, for a solenoid to fail with a shorted turn in the coil, resulting in higher current in the circuit.

 

[ndzied1]

Thanks everyone for your replies.

These are hydraulic valves so MAC wouldn't apply. And, even if it was impossible for a MAC coil to burn out (yes they are legendary for their ruggedness but nothing is 100%) there is always the possibility of the cable getting cut or wires or connectors falling off.

The best solution is a valve spool position sensor when the valve is fully open which I am looking into.

The system has two rams lifting a platform from a flow divider. We have end position limit switches and even a tilt sensor. The customer is very cautious and was trying to see if there was a way to sense the condition where a blocking valve failed (closed when we told it to open). This would make the platform go out of level and should be caught by the tilt sensor. But, if we could easily sense that the coil failed or valve failed then we could shut it down before the tilt sensor kicks in and save the maintenance guys some hassle of getting everything lined back up.

 

[JerryH]

This would work if using a 12v sol and a 12v relay with the same resistance as the sol and a 24v power source.

But really no good indication the valve is working properly. Just indicates the coil is not open and is drawing current.

 

[milldrone]

Presuming that Norm is using AB gear there are diagnostic output modules that check for open and shorted outputs. An example would be 1756-OB16D.

 

[keithkyll]

If you use sinking outputs, just connect an input to the output. The coil will act as a conductor for 24DC, sending a high to the input module. When you energize the coil, it will go low. This is how your car computer knows when you leave a plug disconnected, etc.

If you're using sourcing outputs, you'll need a pull-up resistor. 2400-2700 Ohms would be good.

 

[ndzied1]

If you use sinking outputs, just connect an input to the output. The coil will act as a conductor for 24DC, sending a high to the input module. When you energize the coil, it will go low. This is how your car computer knows when you leave a plug disconnected, etc.

If you're using sourcing outputs, you'll need a pull-up resistor. 2400-2700 Ohms would be good.

I did think about that but as I think is clear now, sensing the spool is the best solution. I'm worried about the valve not being open and the spool sensor will do it whether the coil is burned out or shorted out or wire falls off etc.

 

[daba]

I did think about that but as I think is clear now, sensing the spool is the best solution. I'm worried about the valve not being open and the spool sensor will do it whether the coil is burned out or shorted out or wire falls off etc.

That's not strictly correct... the valve solenoid spool may be moving, but it doesn't mean the spool is opening the valve. I don't know what type of valve it is, butterfly, gate, or whatever, but there must be a mechanical linkage between the actuator and the valve itself.

I once saw a dosing line that hadn't done anything for an unknown period of time because the actuator had "fallen off" the valve body, and was just actuating fresh air. The feedbacks were always correct, so no-one knew that the "valve", as an assembly, had effectively failed.

As I stated in post #16, if it is critical to know that the valve is opening, you have to measure something that happens when the valve is open. You should not assume anything in this game...

 

[ndzied1]

This is the type of valve.

http://www.sunhydraulics.com/model/LKHCZ

If the pin is all the way over (and sensed) then the valve is open. Not linkage. I suppose that the spool could break but then again, if you keep putting checks on checks on checks you are open to more nuisance trips than anything else.

I also proposed measuring each ram and if motion stops on one side when we expect motion we can shut it down but they aren't going to pay for that.

This is just a "would be nice if" type of thing.

 

[daba]

This is the type of valve.

http://www.sunhydraulics.com/model/LKHCZ

If the pin is all the way over (and sensed) then the valve is open. Not linkage. I suppose that the spool could break but then again, if you keep putting checks on checks on checks you are open to more nuisance trips than anything else.

I also proposed measuring each ram and if motion stops on one side when we expect motion we can shut it down but they aren't going to pay for that.

This is just a "would be nice if" type of thing.

If the solenoid "armature" and the valve stem (that opens/closes ports in the valve body) are one piece, then I agree, measurement that the solenoid armature has moved is adequate. And there is a built-in limit switch to signal that.

Just a thought though, what if the valve ports become blocked, and the valve doesn't pass anything, even though the stem is moving ???

 

[osmanmom]

install PT

 

[ndzied1]

Just a thought though, what if the valve ports become blocked, and the valve doesn't pass anything, even though the stem is moving ???

Well, there is really a never ending chain of "what if" scenarios. We already have a tilt sensor on the machine that will sense if things get out of kilter and shut things down.

This is an attempt at secondary protection. And, we are pretty focused on filtration on our hydraulic systems so total blockage is not something that happens unless there is a catastrophic failure of something else in the system that sends chunks into the oil.