1734-IE2C raw value drops

Hello people,
I have a 1734-IE2C module which reads a valve position. When the valve is 100% opened the channel data value is something about 17000, but sometimes it drops to 14000 for a half of second, then returns back to 17000. No problems for the process, it's just strange. Any suggestion - why could it happen?

Tag!
My go to for this one is noise.
Are you pulling contactors in with this same 24V power supply?
Long cable runs?
Improperly wired drives?
Is the valve sufficiently earthed? Fluid (or solid too) through a valve or pipework will build up static electricity if not earthed properly.

Chances are that the position feedback from a valve positioner is a loop powered AO 'option' card on the positioner with fair degree of isolation.

The 1734-IEC2 provides voltage for loop power. The manual has a note for the integrated 12/24V DC supply "Not protected, 0.30 A maximum"

I'm not sure what "not protected" means; lack of a crowbar if short-circuited? But 300mA is far more than even two loop powered outputs need, so the power supply should be stiff enough to not droop the loop current.

Plus most loop powered devices are fast responding enough to change their internal resistance to maintain the loop current even when the power supply droops, as long as the droop doesn't go below the voltage needed to drive the current through the loop resistance.

It could be noise, but if the cable is shielded twisted pair, then induced noise current in reverse polarity has to be pretty intense to drop a current signal 5%. And only for a half second? Of course on another forum the Modbus problem turned out to be non-twisted, non-shielded, parallel conductors so cabling does count.

I've never run into nor heard of an insufficiently earthed valve building up a static charge. But I could see that happening and if the discharge path is through the AO circuit, it could affect the current. The half second duration due to AO circuit filtering?

Is the positioner brand/model trusted enough to not have a firmware issue ? Outsourced firmware development in the past decade has lead to some process instrumentation errors that were unheard of in previous generations.

Without knowing anything about the sensor reporting the valves position;
If the valve position sensor is a sweeping contact type, and if the valve being fully open (100%) is common, then I'd look to see if the position sensor is getting worn or dirty. If you keep a sweeping contact in one position for long periods of time then vibration will cause that position to wear and corrode (I used to see this in older style throttle position sensors at idle all the time). When this happens, when in that spot, the values can jump around as vibration causes the resistance to change.
One possible way to deal with it (assuming it's not serviceable) is to increase the filter on the 1734-IE2C (if it has one) or create a filter in ladder logic.

Update: The valve is controlled by PSAP.3. And I found that indeed the value drops only for half of second when the valve is fully opened. But when it goes to 100% or begins to close - there is a lot of oscillation, I attached the picture of it.
Any suggestion?

It doesn't really tell us anything without knowing what we are looking at I.E. what is the oscilloscope attached to?

Green line is the setpoint from the analog output to X23 of PSAP.3

Blue line is the feedback, the value of analog input connected to X24.

Looks like one problem is solved, the oscillation is stopped. Now question is - why there is a jump between 14000 and 17000. I'll try to adjust P5 and P6 if I'll have a window in the process.

One of the pieces of the puzzle that's still missing is what is providing the analog feedback I.E. where does the analog signal that is connected to the IE2C come from? If it is a mechanical "sweep arm" valve positioner, (which the oscilloscope trace looks like) then I would say a dirty and/or worn out contact is the problem. A "sweep arm" sensor is basically a potentiometer that changes resistance value as it sweeps. It has a very small point of contact on a resistor and the point and the path it rides along wears and corrodes over time causing the resistance to change and become erratic rather than smooth and linear. As the water is flowing the valve vibrates and the sensors value starts "jumping around" with the vibration especially in the area(s) where it is worn the most (like the full open position). Both the card and the oscilloscope are showing a fluctuating value coming from the sensor. I'd be looking at the sensor.

The valve is operated by the PSR-E-C45 actuator. I made a little adjustment and now the valve is stable at 100%. I think the case is closed for now. Thanks for the help.