Interesting phenomenon. Both temperatures jump at the exact same time, but in different directions. You replaced the PLC thermocouple input module, so assume not the PLC module. By "made sure it is routed correctly away from high frq. and grounded correctly", are they "ungrounded" TC probes? Are the TC common wires at the PLC module (CH1,CH2,CH3,CH4) connected to module common (0V)? Are the TC cables shielded? Or running in metal conduit? Is something happening (something turning on or turning off) at the time the temperature readings jump? It could be voltage induced somehow. Are the (2) TC cables running parallel in the same direction? or opposite directions during any section of the cable run? As a test, can you install an ungrounded TC probe, having a short TC wire as possible, connect it to a TC-to-4-20ma transmitter, and monitor the transmitter temperature for any jumping. This could prove/disprove induced voltage theory.

Is the PLC 24VDC or 120VAC powered? Can you monitor the voltages (L-N,L-G,N-G) to the PLC (or +to-, +toG,-toG), and include it on your trend graph of temperatures? Also monitor the voltages to the power supply for the TC module 24VDC power. This could verify that the PLC and module is receiving clean power, or not, when the jumping occurs, again made by something happening in the plant.

Your wiring, connections, and installation sounds good.

Your testing with the TC calibrator sounds good. This will test the PLC and module, as the furnace rep believes that is where the problem lies. You changed the TC module, but you did not change the PLC, correct? I am guessing this has worked fine for years, and now the problem presented itself, so something changed.

You mention (2) TCs connected to this TC module. Are there any other TC connected? If so, do they exhibit faulty behavior? If no other TCs, could these TCs be moved to the spare inputs to rule out the input channels? Will you need to replace the PLC to keep the furnace rep happy?

You mention "the issue is more prevalent at higher temps (when current to the heaters is higher)." Is this electric heat? with Solid State Relays (SSR) controlling the power to the heater elements? Are the SSRs zero-crossing firing? or phase-angle firing? I am guessing phase-angle firing. Phase-angle firing creates more noise, but you probably need the precise control. Does this higher heat effect the temperature of your PLC? How old is the PLC? An older PLC may be more prone to problems with excess heat.

What I suggested was to connect a TC-to-4/20ma transducer to the TC probe in question, or better yet, install a separate TC probe in the same location in the furnace, run the 4-20ma cable to a 4/20ma input module, and monitor that temperature reading. If installing a separate TC probe, this test would eliminate everything except the PLC. (But I don't know if what I am suggesting is feasible) As far as monitoring the voltage to the PLC and the TC module, you could use voltage transducers and connect to the 4/20ma input module. You can view and save the voltage trend with what you used for your temperature trend graph. What did you use to trend your temperatures? DirectSOFT Trend View?

Long time no post... but this issue has come back after a Control and Overtemp thermocouple replacement. The problem is very sporatic... come and goes (has been going on for over a year). But, after this replacement it certainly came back.

To answer your question, no we have not replaced the PLC.

Recently, to test I removed the Overtemp wiring from the TC input module and hooked up a thermoucouple calibrator right into it. That way I could put in an artificial value and see if the noise still comes through. If it does, then the PLC/inut module is to blame. If it doesn't and the control TC has noise, then it is outside the PLC. However no noise came through on both the Control or the Overtemp readings.

I wasn't expecting the PV readings to clear up as well.

After this we tried running new TC wire from the Overtemp TC right into the input module. Did a test run, and the noise/fluctuations came right back.

Then I hooked up the calibrator to the furnace side end of the Overtemp TC wiring. That way we're still using all the normal wiring while also putting in an artificial temp. This run was nice and smooth.

Does this mean the overtemp thermocouple or connector has to be to blame? And why does putting in an artificial overtemp reading also smooth out the control TC data? Like I said before, the fluctuations are opposite of each other. As in, the exact second that the control TC jumps down 30F, the Overtemp will jump up somewhere around 30F. And when the calibrator is on to smooth it out, the control TC smooths out as well.

"has been going on for over a year"! An ongoing head ache!

Your test of disconnecting the furnace over-temp T/C, and injecting a T/C voltage on the T/C extension cable at the furnace end - and both the over-temp and control temperatures became stable - shows the T/C extension cables, PLC T/C module, and the PLC are all ok. This also shows the problem is at the furnace end. (I know the furnace rep would like to blame the PLC)

An explanation is that a momentary voltage occurs on the over-temp T/C probe during the temperature fluctuation, that is different than the common ground voltage at the PLC and module. This voltage could be induced, or conducted, on the probe. When you disconnect the over-temp T/C to connect your calibrator, you no longer have this electrical connection.

Is one, or both, T/C junctions grounded to the T/C probes? Which ones? What happens if you inject the T/C voltage on the control T/C? Does the over-temp become stable? Or exhibit the fluctuation? Are the T/C probes metal (stainless steel)? Are the probe(s) in contact with the furnace metal? If so, can you insulate the probes from the furnace metal using ceramic tubes? Does insulating the probes from the furnace metal help?

We replaced the pigtail coming off the overtemp TC because that was one of the only pieces between the calibrator (when it was hooked up to the TC cable) and the thermocouple itself that wasn't tested. A test run after still showed the noise.

To answer your questions:

-The control TC wiring is grounded at the PLC input module (pic)

-The overtemp TC pigtail and wiring is grounded at the terminal block, where the TC wire is plugged in (pic)

-TC probes have a ceramic sheathing but do have an exposed tip at the junction (pic)

-I am reluctant to use the calibrator as the control TC because this will directly affect the power and I will need to keep the temp right on the setpoint the whole time. But I can give it a shot if you think it will give me important information

Right, I understand about not wanting to disconnect the controlling T/C. So it appears, from your T/C junction photo, that the T/C junction is not connected to a metal probe sheath, or connected to any furnace metal. So possible furnace stray voltage is not conducted to the T/C junction. Are the control T/C and the over-temp T/C installed in the same manner? Are both T/C junctions not contacting furnace metal?

I am suspect of the cable shield drain wires. It appears from photos you are treating each cable shield differently. How are the T/C wires connected to the pigtails? How are the pigtail shield drain wires handled? Insulated? Grounded? Try testing by insulating all drain wires, no grounding of drain wires, insulate all exposed drain wires with electrical tape. Perhaps a drain wire is connected to, or touching, a ground at multiple locations, causing ground loops.

When connecting your calibrator to the over-temp extension cable, at the furnace, both temps are stable! So it's got to be the something with the over-temp T/C installation (including drain wire), and/or the furnace operation effecting the over-temp T/C, right? (you already replaced the over-temp T/C) If you haven't already, contact tech support. They can help get to the bottom of this.

We recently tried a thermocouple that is fully sheathed in ceramic and metal. Just like that the noise is gone. So, it sounds like there is some kind of interference specifically around the overtemp port hole into the furnace.

But anyway, we would like to continue using this fully sheathed TC. However, our Control TC is an exposed junction and sees heat much quicker. When I start a cycle the Control will read over 250F before the Overtemp hits 130F (TC reading do not show until they reach 130F). This means that the furnace thinks the OT is not working and labels it as "OPEN" and activates a fault. When the Control gets to 250-300, the OT will finally turn on at 130F. Then I can clear the alarm and continue. Once the temps get up closer to 2,000F the TC readings start merging together. But we cannot have operators clearing OT OPEN faults every run. Is there a way to get the PLC to read the TCs at a lower temp? Or wait longer until it sends an alarm for the OT being open, because it eventually warms up.

TIA