BX-04THM issues on a heat treating oven.

[alrobot]

Hi:
I have a BRX CPU with a BX-04THM module for my thermocouples. 3 K type, 1 T type.
When I heat the oven up, I get to around 1250 Deg. F. and my input falls off to around 1000 Deg. F. this is an issue since I am using this input to control my Oven PID loop for the main heater.
The input is daisy chained with a Eurotherm Electronic chart recorder, the temperature reading on it does not falter when this happens.
I also had a second K type tc hooked to another input on my BX-04THM, and it too, fell off on temperature input readings(Raw Data).
This is driving us nuts, we have replaced the BX-04THM card with a brand new one, no change...
I have ordered new thermocouples which will arrive Monday/Tuesday of next week (1/10/22).
The product we heat treat is very expensive and failure like this is not an option. The boss man is making noise about throwing out the BRX and starting fresh with something else. I would like to try to avoid that.
I have since removed the secondary tc from the BX-04THM and run it thru a single loop controller as a safety(Overheat) protection.
Any ideas???

[ADC Product Engineer]

The input is daisy chained with a Eurotherm Electronic chart recorder, the temperature reading on it does not falter when this happens.

Don't do that.  Thermocouples have a very small signal and they don't like sharing.  If you need it to go to something else, then you need to either put in another thermocouple or retransmit the signal as 4-20ma or voltage.

[MikeS]

"I get to around 1250 Deg. F. and my input falls off to around 1000 Deg. F. "

Does the input value stick at 1000? or does it lose 250 degrees but keep changing?

You're probably aware that the #1 problem with THMs is noise. The THM input signal range is 156mv so it doesn't take much stray noise to make an impact on the signal. Does something happen in your process at 1250 deg. (like a motor coming on or shutting off) that could be a source of noise that's not present before 1250 deg.?

I agree with the ADC post that daisy-chaining the THM input is bad. Using separate THMs is the preferred method. This is the way.

[alrobot]

I am looking at possibly going to a TC transmitter that with split the signal, mV/4-20mA. I was trying to keep away from the scaling issues associated with that.
Answering a few questions in Replies:
There is nothing that happens at 1250, that is just where it seems to happen.
The input value at the Eurotherm CR holds at 1250 and continues to climb as heat is applied. There are also 9 other TC's hooked to the Chart recorder in zones inside.
At one point, I got the Fluke control meter, and plugged in the TC I unhooked from PLC, it read accurately, and I dialed in an output on the Fluke and the PLC read correctly. This really baffled me... Its almost like the BX-04thn went into some shutdown mode...
Agree to Agree that daisy chained TC's are just asking for trouble. This thing has been a PITA since it was delivered. I'm coming in 4 years into the nightmare that is this furnace. I figured I would wave my magic PLC wand, and everything would be fixed. I must have forgot to renew my PLC magic wand policy.
The weird part is, I had a TC directly coupled to PLC that exhibited same/similar issues. that one is now on a single loop controller as my overtemp safety.
This morning after it cooled down from yesterdays test, I found the - TC wire up on top was loose, I fixed it to the junction block good and tight and restarted my cycle.
One of the biggest problems is, i have a 4 hour heat up, if anything goes awry, I cant work on a 1200+ degree oven until it cools, which is the next day. Of course, this is the hard part to explain to the boss...

[alrobot]

Quick update:
I tried running a cycle. It heated up to 1231 Deg F. Then it fell down to 195 Deg F. I put the PLC into program mode, and back into Run Mode. The temp then creaped back up to 800 Deg F. over about 20 minutes.
This is so aggravating.

Alan...

[jrayb]

Are the thermocouples grounded or not grounded? I had issues like this & my problem was that I was using grounded thermocouples. You will have to use ungrounded thermocouples.

[alrobot]

Jrayb:
I have had the grounded/ungrounded TC issue already. as far as I know these are ungrounded TC's that we make in house in the maintenance shop. I ordered some high temp ready made from Omega, they are due Monday or Tuesday. Those are ungrounded TC's.

[Controls Guy]

An inline 4-20 transmitter is just a superior way to go, even if you weren't splitting the signal to go to two different instruments.   TC signals are crap, just run 4-20, always.   Same with RTDs or thermistors.    Long distance, splitting, even if nothing special, just run 4-20.

The head-mount transmitters are nice if your application will tolerate them.   I like the non-programmable ones because if one fails, they replace it with the part number printed on the side and everything works.   Get a programmable one, and they may not know they need to, or may not be able to find the programming hardware, the software, or the parameters.

I'm old, cranky and opinionated, but at least I'm an expert on my own opinion.