Recommendation? PID Calibration

T

Tarkus

Joined
Jun 10, 2020
Messages
36
Is there a thread for this yet?

My current PID (NAIS KT4) will hold temperature for days, but it is reading out by about 50°F using the salt melt test. (Pure NaCl is melting at "1426"°F) I recently replaced my thermocouple which is how I got it this close (previous TC was sharply bent and out by 130°). Any advice or redirection to info would be very appreciated!
 
I zeroed in the exact temperature since posting this. Salt melts at 1474°F but my oven reads 1409°F.
That means the oven is 65°F hotter than what it's reading.
 
Folks ... I’ve been working offline with tarsus on this, and the fact that the oven temp is reading low (versus high), but keeping steady, seems to me to point to a sensor/thermocouple issue ... but I am really weak on what can cause a thermocouple to read off by this much. He has swapped out thermocouples, and continues to have the problem. Can anyone help him?
 
Wrong type wires
Incorrectly connected polarity
Wrong type plugs, sockets, connectors
Wrong type TC or improper installation
TC in a sheath?

I would suggest checking against a different TC and readout. It may be his salt test that is inaccurate.
 
My guess would be a cold junction error. The 65° degrees difference should correspond to the room/enclosure temperature your controller is in. One some controllers, there's a sensor that automatically corrects this. One the one you have, there is a 'sensor correction setting' menu in which you can enter a fixed temperature value to offset the measured value. Note that if the room/enclosure temperature differs from the correction setting value, an error will reappear...
 
Last edited:
Stacy - in what way can the salt test be in error? The melting point of salt is pretty well defined, and predictable..... ??
 
Stacy E. Apelt - Bladesmith said:
Wrong type wires
Incorrectly connected polarity
Wrong type plugs, sockets, connectors
Wrong type TC or improper installation
TC in a sheath?

I would suggest checking against a different TC and readout. It may be his salt test that is inaccurate.
Click to expand...

This is interesting! It's an 8g type K thermocouple in a ceramic sleeve (bare tip).
The polarity is correct. (When wired opposite, the temperature rapidly drops into the negatives while the coils are fully powered)
As for the connections, though... The TC is straight wired into the PID using standard 14g electrical wire, except for the ceramic mounting block. I'll try attaching the wires directly to the TC leads.

As for the salt test, I'm using pickling salt that lists the only ingredient as NaCl (table salt has some other stuff thrown in for flavour, apparently). I put a little pile on a piece of 304 stainless and let it sit for 10 minutes at temperature before checking.

Edit: I have used two different PIDs (can't remember names) and three other TCs (one 14g with stainless sleeve and two other 8g ceramic sleeve, but those I had bent for a better shape), all of which had FAR worse results than what I'm getting now.
 
Last edited:
Gab-e said:
My guess would be a cold junction error. The 65° degrees difference should correspond to the room/enclosure temperature your controller is in. One some controllers, there's a sensor that automatically corrects this. One the one you have, there is a 'sensor correction setting' menu in which you can enter a fixed temperature value to offset the measured value. Note that if the room/enclosure temperature differs from the correction setting value, an error will reappear...
Click to expand...

I'm not entirely sure what you mean about the enclosure... The controller box only gets up to about 100°F at any time.

I appreciate the offset advice, though! Just seems absurd for the temperature to be so far out... And if there's an issue now, I'd like to correct it so thing stay predictable in the future.
 
Tarkus said:
This is interesting! It's an 8g type K thermocouple in a ceramic sleeve (bare tip).
The polarity is correct. (When wired opposite, the temperature rapidly drops into the negatives while the coils are fully powered)
As for the connections, though... The TC is straight wired into the PID using standard 14g electrical wire, except for the ceramic mounting block. I'll try attaching the wires directly to the TC leads.

As for the salt test, I'm using pickling salt that lists the only ingredient as NaCl (table salt has some other stuff thrown in for flavour, apparently). I put a little pile on a piece of 304 stainless and let it sit for 10 minutes at temperature before checking.

Edit: I have used two different PIDs (can't remember names) and three other TCs (one 14g with stainless sleeve and two other 8g ceramic sleeve, but those I had bent for a better shape), all of which had FAR worse results than what I'm getting now.
Click to expand...
I think you need to use TC wire to connect to the PID, not copper.
 
...The TC is straight wired into the PID using standard 14g electrical wire, ..
This is the first place to start. You need type K TC wire. regular wire will give an inaccurate reading. If you need some, email me and I'll send you a piece. The TC block is also polarized and should be marked + and -.

In an electric HT oven, the TC is normally exposed, not in a ceramic sheath.

The melting point of things is a technical point. The actual salt you use may not be 100% lab grade NaCl. Even though the container is marked "sodium chloride" most is still a Nacl/Kcl mix unless it is a technical grade chemical. There is oxidation and other things that can greatly affect the melting point as it reaches liquidus. Additionally, it doesn't necessarily melt all at once. Take an ice cube and set it on the kitchen counter and see if it melts instantly. This is even more of an issue with solids tat take high temperatures to melt. There is the slump stage first and then finally the flow stage. It can be quite a range between them with some materials.



I would get some Templi-cones or kiln cones and use them to test the oven. They slump from a cone to a blob at a rather tight temperature. Templi-stix work similarly once you get the hang of them.
 
Tarkus said:
I'm not entirely sure what you mean about the enclosure... The controller box only gets up to about 100°F at any time.

I appreciate the offset advice, though! Just seems absurd for the temperature to be so far out... And if there's an issue now, I'd like to correct it so thing stay predictable in the future.
Click to expand...

A thermocouple is essentially a voltage generating device made out of 2 dissimilar metals. When you connect the thermocouple to the PID controller terminals, the screws/terminal are made of another metal which creates another voltage generating junction (the cold junction). This additional junction being at room temperature will influence the measure you are getting from your salt pot.

Edit: missed the part about the tc being wired with standard copper wire...this will cause the same issue (basically creating a second tc which is measuring the air temperature around the junction between tc leads and copper wire) .
 
Last edited:
Tarkus said:
Is there a thread for this yet?

My current PID (NAIS KT4) will hold temperature for days, but it is reading out by about 50°F using the salt melt test. (Pure NaCl is melting at "1426"°F) I recently replaced my thermocouple which is how I got it this close (previous TC was sharply bent and out by 130°). Any advice or redirection to info would be very appreciated!
Click to expand...
What brand is oven ? How/where are HT element ,where and how deep inside is positioned TC ? You put salt on the floor , right ?
 
Stacy E. Apelt - Bladesmith said:
...The TC is straight wired into the PID using standard 14g electrical wire, ..
This is the first place to start. You need type K TC wire. regular wire will give an inaccurate reading. If you need some, email me and I'll send you a piece. The TC block is also polarized and should be marked + and -.

In an electric HT oven, the TC is normally exposed, not in a ceramic sheath.

The melting point of things is a technical point. The actual salt you use may not be 100% lab grade NaCl. Even though the container is marked "sodium chloride" most is still a Nacl/Kcl mix unless it is a technical grade chemical. There is oxidation and other things that can greatly affect the melting point as it reaches liquidus. Additionally, it doesn't necessarily melt all at once. Take an ice cube and set it on the kitchen counter and see if it melts instantly. This is even more of an issue with solids tat take high temperatures to melt. There is the slump stage first and then finally the flow stage. It can be quite a range between them with some materials.



I would get some Templi-cones or kiln cones and use them to test the oven. They slump from a cone to a blob at a rather tight temperature. Templi-stix work similarly once you get the hang of them.
Click to expand...

Ohhhhhh ok! That seems like such an obvious answer, thank you so much! Will Kanthal A1 wire work? I have a bunch of that left, otherwise yeah I'll email you. Thank you!!!

The fuse point of the TC is exposed, the sheath only covers the leads. And the block has no markings, just ceramic with two metallic tubes and screw connectors on the ends.

I gotta look into these cones, sound pretty useful!
 
Gab-e said:
A thermocouple is essentially a voltage generating device made out of 2 dissimilar metals. When you connect the thermocouple to the PID controller terminals, the screws/terminal are made of another metal which creates another voltage generating junction (the cold junction). This additional junction being at room temperature will influence the measure you are getting from your salt pot.

Edit: missed the part about the tc being wired with standard copper wire...this will cause the same issue (basically creating a second tc which is measuring the air temperature around the junction between tc leads and copper wire) .
Click to expand...

Oh ok, that makes a lot more sense! Thank you, hopefully I can find some type K TC wire!
 
Natlek said:
What brand is oven ? How/where are HT element ,where and how deep inside is positioned TC ? You put salt on the floor , right ?
Click to expand...

Like I said, it's home made with a NAIS KT-4 PID and the salt is piled on a piece of 304 stainless. Elements on the sides and back of the 4x7x22.5" oven, TC tip is about 1" down from the top, dead center
 
You must log in or register to reply here.
Back
Top