4-20mA signal unstable voltage and operation on Rotork actuator

I'm fairly new to the world of PLC's. So far I've managed to resolve the problems I've crossed but this one has me stumped.

I've got a Rotork actuator (IQT) thats fed a 4-20mA modulation signal that sets the position of the valve its attached to. This was working correctly for some months however I have now found that it continuously moves open and closed quite randomly. The first couple of times it failed it worked again for a time after before failing again but now it just moves randomly all the time the analogue drive signal is attached.

With a multimeter I measured the current entering the actuator and its totally solid however I did notice that the voltage driving the current is all over the place. Anywhere from 1-10v quite randomly. I suspected the output card of the PLC as that is driving the current so I disconnected the actuator signal wires and placed a dummy 250 omh resistor in its place. Both the current and voltage over the resistor where solid so I have no reason to think its the PLC card at fault. Interestingly I checked the resistance across the actuators input terminals and found that it was varying alot, quite randomly. Generally around 220 ohm but anywhere from 200-500ohm. My multimeter is cheap and I suspect that its missing alot of the variation. A scope would probably show the full extent of the problem.

Any ideas? It looks to me as if the analog input of the actuator is at fault but it seems unlikely that that would break in any way. I suspected that it might have been set up to take 1-5v not 4-20ma but that doesn't explain how it worked previously or why it has a resistance.

My next step is to get hold of the remote (supplier of the skid the actuator is part of removed it from the actuator) and check how the inputs have been setup.

Thoughts? Similar experiences?

I have had a similar experience with a Rotork IQ Mark 1, which was fitted with a folomatic card (the card that accepts the 4-20 mA and sets the position for the actuator). Similar problem to what you are facing, valve opening and closing randomly, and it turned out to be the folomatic card. New one from Rotork fitted and all worked fine again.

Thanks for the reply, so it could be the foliomatic card, interesting. I only started have a problem with the actuator a couple of weeks back during the first stint of hot weather. It was in the hottest part of the day it failed but once the day had cooled down it was fine. The same happened again the next day. I rote it off as been a coincidence but maybe the heat found a weakness in the card. The actuator is also mounted onto a valve with hot gases passing through it and is also mounted below a flare which had also been running for some time. Pretty warm where the actuator is.

What was the procedure like replacing the card? Can it been done without breaking the warrantee void stickers? Can removal be done in situ or will it need to be taken away?

Cheers

Just to follow up. I haven't replaced the foliomatic card but I did find a short term workaround.

I set the Rotork to accept a 1-5v signal instead of 4-20ma and stuck a 250ohm resistor in parallel with the terminals to give it a voltage to work with. The actuator now behaves perfectly. I suspect that I may still have problems with the foliomatic card later however this does allow me to operate the equipment in the short term while I get a replacement card sent.

Nomad Engineer said:

Just to follow up. I haven't replaced the foliomatic card but I did find a short term workaround.

I set the Rotork to accept a 1-5v signal instead of 4-20ma and stuck a 250ohm resistor in parallel with the terminals to give it a voltage to work with. The actuator now behaves perfectly. I suspect that I may still have problems with the foliomatic card later however this does allow me to operate the equipment in the short term while I get a replacement card sent.

Click to expand...

Good idea!!

Have you tried forcing a signal with a process meter? I've used a Fluke 787 in the past but bought a Fluke 773 clamp on millAmp meter. Its amazing. No more unwiring to read amps. Its about a grand but well worth the money.

Saturn_Europa1 said:

Good idea!!

Have you tried forcing a signal with a process meter? I've used a Fluke 787 in the past but bought a Fluke 773 clamp on millAmp meter. Its amazing. No more unwiring to read amps. Its about a grand but well worth the money.

Click to expand...

Dont have a signal injector unfortunately and I cant see my company forking out for a 1k clamp on process meter as much as I'd like one. The best I can do to force a signal is via the PLC output card, does the trick for the time been.

Nomad Engineer said:

I suspected that it might have been set up to take 1-5v not 4-20ma but that doesn't explain how it worked previously or why it has a resistance.

Click to expand...

Inputs designef to handle a DC current 0-20mA pr 4-20mA control signal need a dropping resistor to create the IR voltage drop needed for electronic input circuit to 'see' the current signal. The nominal 200 odd ohms that you measured at the input is typical.

Inputs designed for 1-5 volts, with a 'live' zero, are assumed to be 4-20mA current signals driven through a 250 ohm resistor.

True voltage outputs, less common on industrial products, typically are zero based, 0-1V, 0-5V, 0-10v. They expect a high impedance input, typically 1Mohm or higher, so that the output is not excessively "loaded" by the parallel resistance of the input circuit. Voltage output resistance/impedance is typically about 100 ohms or less.

A 100 ohm output in parallel with a 250 ohm input would load the output down - resulting in a huge error.

100 ohm output in parallel with 1Mohm has negligible error.