RTD- failure

Hello,
We have simple "cooking" process which includes two probes (RTD's).
One immersed directly in water while other is "stabbed" into the product.
The process is heating by steam, holding and then cooling using chilled water. ( full cycle time may reach 8 Hrs)

Temperature range is 80 Celsius to 10
I noticed repeated failure of the product probe and it always happen during the cooling cycle.
when it fails, i get like open circuit reading
-32,768
Is there any specific reason(s) might cause that?
thank you
 
The open circuit reading of -32 deg indicates a broken wire or element. The internal wires are very small gauge wires.

The rough physical useage, the stabbing action, breaks either a wire-wound RTD element or the wires connected to a thin film solid state element or a wire-wound element.

If replacement cost of stab style elements is not considered the "cost of doing business", then you might consider doing a trial with other brands to see whether those are constructed in a more robust fashion that the ones you're using. You might one one that is, or you might find that they're all about the same. Only a trial will tell. The salesman's claim that theirs will last longer than what you're using is called "puffery" and has no evidence to back it up, but a trial with documentation (how many stabs) will tell you whether one is better than another.
 
The open circuit reading of -32 deg indicates a broken wire or element. The internal wires are very small gauge wires.

The rough physical useage, the stabbing action, breaks either a wire-wound RTD element or the wires connected to a thin film solid state element or a wire-wound element.

If replacement cost of stab style elements is not considered the "cost of doing business", then you might consider doing a trial with other brands to see whether those are constructed in a more robust fashion that the ones you're using. You might one one that is, or you might find that they're all about the same. Only a trial will tell. The salesman's claim that theirs will last longer than what you're using is called "puffery" and has no evidence to back it up, but a trial with documentation (how many stabs) will tell you whether one is better than another.
thank you so much David for your response and help.
I can fully understand the open circuit situation but, if I try same probe again
it is going work perfectly in heating and holding however, again fail shortly once starting cooling phase.
That is what confuses me, i understand open circuit must be permanent
 
There are many types of RTDs (Pt100; DIN100; etc.). They each have different ranges of temperatures they work best in. Are you using the right RTD for the range you are measuring?

But, David_2 brings up a good point: If the RTDs are physically the same the only difference is the “stabbing” that one is being subjected to. And if you’re only getting one heat/hold/cool cycle per RTD, then are you sure the RTD you’re using is correct for the “stabbing” application?
 
Heat makes the components expand and in the expanded state, the broken wires touch and there is continuity so the wires conduct the current from the analog input source.
The wires shrink when cooled, the wires no longer touch and the circuit is “open”.

historical anecdote
At a trade show in the late 1970’s I was shown a prototype of a penetrating RTD that used a heavy knife, like a US Marine Corps Ka-Bar combat knife that had a slot milled along the upper part of the blade and hole drilled through the handle and hilt for the signal cable. The RTD sheath lay in the milled slot and its wiring out came out the back of the handle. The knife blade took the brunt of the force of penetrating the object (slaughterhouse carcasses if I recall), not the RTD sheath. The mass of the knife blade was many times that of the RTD sheath which would mean a long time constant for thermal response but the combat knife aspect amused the vendor guys in booths. I have no idea whether it ever went into production or not.
 
I saw this problem quite often in cooking furnaces for butchers.

We solved these problems by switching from the mechanically sensitive Pt100s to mechanically more rugged thermocouples, i.e. a type K thermocouple. But your measurement input must support that of course.

I can assist you with the temperature sensors, temperature controller or power electronics for driving the heating/cooling.
See an overview here: https://cascade.net/en/all-products/

Temperature control in all kinds of processes and equipment is my speciality. I also made complete control systems for cooking kettles heating with steam (in double wall of kettle) and cooling with ice- or normal water.
Attached a brochure and picture. Unfortunately in Dutch language, but copy+paste in google translate will do wonders.
Feel free to contact me if you are interested. I can switch the language automatically to English in the software of this system.
Can also be operated remotely via smartphone/tablet/windows PC application, if you connect the touchpanel to the internet.
100_3261.JPG
 

Attachments

I also wonder about using RTD’s for the “stabbing” application….

Sometimes (too often) economics drive engineering design decisions and sometimes the other reasons—once understood—make sense but not really. I believe many T/Cs are not so good at temperatures this low, but depending on the tolerances of the process they might be fine, or they might not.
 
Type T thermocouples are alloys of copper and not only work well at ambient, below ambient and cooking temperatures, have half the error of Types K and J, but are widely used in food and pharmaceutical applications specifically for the reasons cited. Specifying limit-of-error wire halves the error again. But I always wonder how much the wire error really matters in comparison to the cold junction sensor in a handheld thermocouple meter. Moving from a warm area to a cold area (or vice versa) can have multiple degrees of error if the plug cools or heats more rapidly than the internal CJ sensor, until the CJ attains the same temperature as the plug.

Light gauge lead wires can break when they take a pounding whether they're copper RTD lead wires or alloy thermocouple lead wires. My gut feeling is that a well constructed Type T thermocouple is likely more robust than an RTD.

One cannot examine the internals of an MgO packed thermocouple or an MgO packed RTD, so trying ones from different manufacturers can reveal the sturdiness of a given design.
 
You are right. A type T thermocouple is also what I would advise. It is more accurate (and stable) than a type K.
We also have them calibrated so they come with a calibration certificate. Can of course be calibrated at the temperatures to your specification.
A customer of mine uses them for testing sprinklers in buildings and the measurements of the type T thermcouple are registered and certified. Every 6 months the type T thermocouple goes into calibration.
They actually have 2 type T thermocouples. 1 is in calibration and after that on stock and the 2nd one is in the sprinkler calibration system. So they always have 1 spare in case of trouble.
Type T thermcouples are well suited for these kinds of applications, where a mechanical more rugged sensor is needed than a Pt100.
 
Top