Ramping/Pausing Temperature input (w/ video)

[phuz]

There was no great way to describe the behavior this temperature input is demonstrating, so I attached a video. It ramps, then hesitates, then ramps, etc. Obviously, this causes the loop to respond poorly. On an identical process, I have a much more linear temperature input and the loop is near perfect. The input is a Pyromation RTD with hockey puck (0-300F) going into a 5069-IF8. I have the card set for 20ms with 500ms averaging. There is a chart recorder in series with the 4-20mA loop and its digital display behavior mirrors that of the HMI, which tells me its in the signal and not the PLC. Is this a bad transmitter or what could cause this?

https://youtu.be/4JAxfW6PaYY

 

[Pandiani]

I have seen something like tihs before. What kind of thermocouple you're using?
Please read this.
https://blog.wika.us/products/tempe...hermocouple-selection-grounded-vs-ungrounded/

And also have a look at this: http://www.plctalk.net/qanda/showthread.php?t=83196&highlight=Grounded+thermocouple

 

[phuz]

I have seen something like tihs before. What kind of thermocouple you're using?
Please read this.
https://blog.wika.us/products/tempe...hermocouple-selection-grounded-vs-ungrounded/

And also have a look at this: http://www.plctalk.net/qanda/showthread.php?t=83196&highlight=Grounded+thermocouple

Actually stated above, but I'll share again
Pyromation RTD with transmitter. (100 ohm Pt)

Thanks for the links, but they don't apply, unfortunately.

 

[Bullzi]

Not really knowing your system I will make a few suggestions:
1. Is it possible that the temps shown are accurate? In other words has your process gone out of wack so bad that it is running away?
2. Are both loops programmed and wired exactly the same? Do the both have a chart recorder in both loops?
3. If no to #1 and yes to #2, then it really boils down to 4 things. The RTD, Transmitter, chart recorder or the Analog Input Channel. Can you easily swap our the RTD/Transmitter combo between the 2 processes? If you do and the problem follows the RTD/Transmitter then you have your answer. If the problem still exist then you could have a issue with the chart recorder or you could have a input issue.

 

[phuz]

Not really knowing your system I will make a few suggestions:
1. Is it possible that the temps shown are accurate? In other words has your process gone out of wack so bad that it is running away?
2. Are both loops programmed and wired exactly the same? Do the both have a chart recorder in both loops?
3. If no to #1 and yes to #2, then it really boils down to 4 things. The RTD, Transmitter, chart recorder or the Analog Input Channel. Can you easily swap our the RTD/Transmitter combo between the 2 processes? If you do and the problem follows the RTD/Transmitter then you have your answer. If the problem still exist then you could have a issue with the chart recorder or you could have a input issue.

Brand new process, identical to one I did a few months ago (which runs flawlessly).

1) Temps are pretty accurate (not counting slight lag of a thermowell on an RTD), but behavior is not consistent with what is actually happening. On the running system, the temperature increases smoothly as you would expect with a wide-open steam valve.
2) Both loops have identical programming. Chart recorder (same model, same input) exists on both and is in series with the controlling RTD/transmitter.
3) Electricians swapped RTD with another we had in the system and saw similar behavior, so it's most likely not the component, unless Pyromation does something with the transmitter that I'm not aware of. I picked up an IFM RTD today and have it being installed first thing in the AM.

Two more things I plan to do: bypass the chart recorder completely, and swap input channels.

Just to be clear, do you understand the issue I am showing in the video? The fast ramp, then it slows, then fast ramp, etc.

 

[Bullzi]

I guess I didn't notice that the first time. Have you tried taking the averaging out of the card? I wonder if that is having some issues and giving you the problem.

 

[phuz]

I did, as well as change the polling rate of the module, but it had little-to-no effect.
The chart recorder exhibits the same pattern, so it's in the analog signal.

 

[g.mccormick]

Can you get an rtd signal generator and connect in place of rtd probe? This will allow you to simulate temps and verify if real or not.

 

[rupej]

I've found that the analog OUTput of most chart recorders behaves exactly as you're experiencing, where the output follows the pen exactly, not the input to the chart recorder.

But since you're measuring in series with the input of the chart recorder, I'm not sure what that would happen. Still, chart recorders do strange things with analogs and for sure I would separate the signals regardless. That's probably the issue... somehow.

 

[phuz]

I've found that the analog OUTput of most chart recorders behaves exactly as you're experiencing, where the output follows the pen exactly, not the input to the chart recorder.

But since you're measuring in series with the input of the chart recorder, I'm not sure what that would happen. Still, chart recorders do strange things with analogs and for sure I would separate the signals regardless. That's probably the issue... somehow.

As I stated, the chart recorder is in series with the analog signal. It is not a separate output.

All this being said, I got a text from the electrician this morning. I told him to go ahead and swap the RTD out and fire up the cooker if he wanted before I got there. He sent me this picture. They did swap RTDs yesterday afternoon with an identical pyromation unit and had the same problem, so this is a Pyromation thing. We have three others in the system, but they are not as critical and will work fine for what we need. I thought Pyromation was a good company and I don't recall seeing this behavior before. I'll be anxious to hear what they say when I send them the video and results from changing to another brand RTD. I can't make sense of it.

 

[BryanG]

You need to do some fault finding.
Connect the temperature sensor to a tester, does the output vary the same. If you don't have a tester then an isolated PLC analogue input will do.
Do you get the same fluctuation if you remove the chart recorder from the loop.

What supplies the 4-20mA loop, check the voltage level and stability.
Check the loop total resistance, the transmitter will have a limit.

You can't blame the temperature transmitter until you have excluded everything else. 4-20mA loops can be tricky little b*ggers.

 

[Pandiani]

Actually stated above, but I'll share again
Pyromation RTD with transmitter. (100 ohm Pt)

Thanks for the links, but they don't apply, unfortunately.

Sorry, when I saw 268, I thought it is probably 268°C and thermocouple instead of Pt100.
For similar applications, we always used mA signal splitter instead of connecting devices in series.

 

[phuz]

You need to do some fault finding.
Connect the temperature sensor to a tester, does the output vary the same. If you don't have a tester then an isolated PLC analogue input will do.
Do you get the same fluctuation if you remove the chart recorder from the loop.

What supplies the 4-20mA loop, check the voltage level and stability.
Check the loop total resistance, the transmitter will have a limit.

You can't blame the temperature transmitter until you have excluded everything else. 4-20mA loops can be tricky little b*ggers.

Sure, you could check all those, but two *brand new* Pyromation RTDs behaved the same way. Swapped out the RTD and changed NOTHING else, and it behaves perfectly. I'm content with saying the Pyromation RTDs are to blame.

 

[Dirt]

Could there actually be temp fluctuations in your process causing this? I'd test the whole loop in an oven or bath with a known good temp meter. If possible I'd adjust rtd placement in process to see if it changes.

 

[phuz]

Could there actually be temp fluctuations in your process causing this? I'd test the whole loop in an oven or bath with a known good temp meter. If possible I'd adjust rtd placement in process to see if it changes.

Nope
All good, though, with the new RTD