Thermocouple plug

Hi. Building a RIMS system and I dont want to permenantly hard wire my thermocouple probe to the back of the PID. I know thermocouple wire is kind of special, but I have some bad thermocouples with wires still on them i can use as pigtails. My question is, if this wire is so special, would adding a male/female disconnect between the TC and the PID cause an issue? I was thinking a typical headphone panel mount jack on the PID side and a headphone jack solder plug on the TC side.

Thinking this
http://www.amazon.com/dp/B0008JFHHO/?tag=skimlinks_replacement-20

and this

http://www.amazon.com/dp/B000ML4A2Q/?tag=skimlinks_replacement-20

You need a special connector when using a thermocouple. The headphone Jack most likely won't work

you should be very careful when extending/changing the wiring on a thermocouple.. it is very critical that both leads are of the exact same material, and of a known resistance (ie: if you add a switch or connection in between, you are changing the impedance).

RTD's are far more robust when it comes to what I think you woudl like to do.. your pid should support a PT100, and they are not that expensive.

Same question here. I will be ordering an Auber SYL-2352 and planned on using a K-type thermocouple but this would be a better option with the deluxe cable so it's removable?

http://www.auberins.com/index.php?main_page=product_info&cPath=20_15&products_id=246

You don't have to worry about using a different material for splicing thermocouple wire provided the material being used is the same on both sides (for example, using a copper barrel splice) and the transition from thermocoiple wire to splice material and back again happens at the same temperature.

A thermocouple is actually just 2 dissimilar metals joined together. Any time you join 2 dissimilar metals, they will create a small electric current that is proportional to the temperature of the coupling.

What this means is when you transition the thermocouple wire to copper, you are creating one thermocouple and another on the transition back. If the same material is used , and the same temperature occurs at both junctions, they cancel each other out.

They make something called a uniform temperature reference which basically does this, only the transition back to thermocouple wire doesn't happen. This allows cheaper copper wire to be used instead of thermocouple wire when you have several temperature measurements in a single remote location.

Hope this helps

DougK said:

You don't have to worry about using a different material for splicing thermocouple wire provided the material being used is the same on both sides (for example, using a copper barrel splice) and the transition from thermocoiple wire to splice material and back again happens at the same temperature.

A thermocouple is actually just 2 dissimilar metals joined together. Any time you join 2 dissimilar metals, they will create a small electric current that is proportional to the temperature of the coupling.

What this means is when you transition the thermocouple wire to copper, you are creating one thermocouple and another on the transition back. If the same material is used , and the same temperature occurs at both junctions, they cancel each other out.

They make something called a uniform temperature reference which basically does this, only the transition back to thermocouple wire doesn't happen. This allows cheaper copper wire to be used instead of thermocouple wire when you have several temperature measurements in a single remote location.

Hope this helps

Click to expand...

Yes, it helps a lot. So based on what you have said I shouldn't have an issue removing the ends of the thermocouple, soldering onto a 3.5MM headphone jack, mating that with a female headphone jack, then running thermocouple wire from the solder side of the female jack back to the PID?

Is there any issue soldering the connections as it introduces another type of metal into the mix? I noticed that the lug connections on my thermocouples now are not soldered, but that could be because they are cheap chinese ones.

pentiumone133 said:

Yes, it helps a lot. So based on what you have said I shouldn't have an issue removing the ends of the thermocouple, soldering onto a 3.5MM headphone jack, mating that with a female headphone jack, then running thermocouple wire from the solder side of the female jack back to the PID?

Is there any issue soldering the connections as it introduces another type of metal into the mix? I noticed that the lug connections on my thermocouples now are not soldered, but that could be because they are cheap chinese ones.

Click to expand...

That is correct, provided you don't use a gold plated plug in a standard jack. The key here is the splice goes from thermocouple wire to splice wire and back to thermocouple.


Solder may introduce a bit of inaccuracy into the system, I'm actually not totally sure on that. When connecting wires at work, we always use either crimp connections or screw terminals. If you were soldering thermocouple wires together, it wouldn't affect it at all. Commercial thermocouple wire is often soldered or brazed at the junction to keep the wires together, but I am not sure how it would affect soldering thermocouple wire on to another metal. The study that I saw and in which I used to base my original post on was specifically looking at quick-connect pins and sockets. Is a special (much more expensive) pin/socket needed on a cannon connector, or will the standard work? It was found that provided the junctions happen at the same temperature, and most of the time they are when making a quick-connect, the temperature reading wasn't affected.


I'll have to look into this and get back to you.

I used a headphone type disconnect for my RTD, cheap and effective.

Use the auberins or other RTD with the panel disconnect or use a TC without any disconnects for an accurate temp reading.

I was thinking about this more, and if the issue is different types of metals creating a reaction and throwing off the temperature then how is it possible to connect the thermocouple to the PID at all? The wires crimp onto lug connectors. These lugs, I assume, are standard lugs not made out of specific material. Then, the lugs are screwed onto metal screws on the PID, which, again I assume are not made out of specifica materials. THEN....inside the PID there is traces and connections on the boards before it actually gets to its final destination in the logic, right?

How does this work? Its the same concept as cutting the cable in the middle and soldering in a copper wire. The metal type changes two or three times already between the sensor itself and the inside of the PID.

pentiumone133 said:

I was thinking about this more, and if the issue is different types of metals creating a reaction and throwing off the temperature then how is it possible to connect the thermocouple to the PID at all? The wires crimp onto lug connectors. These lugs, I assume, are standard lugs not made out of specific material. Then, the lugs are screwed onto metal screws on the PID, which, again I assume are not made out of specifica materials. THEN....inside the PID there is traces and connections on the boards before it actually gets to its final destination in the logic, right?

How does this work? Its the same concept as cutting the cable in the middle and soldering in a copper wire. The metal type changes two or three times already between the sensor itself and the inside of the PID.

Click to expand...

I had this thought too. What I am understanding from previous posts is that as long as all these junctions occur at the same temp then it works, its when you have a temp delta across the junctions that will cause inaccuracies?

Go to omega.com they sell t/c diconnects for panel mount for just a few dollars. These are made for this purpose. I have always had good luck with their products.

pentiumone133 said:

I was thinking about this more, and if the issue is different types of metals creating a reaction and throwing off the temperature then how is it possible to connect the thermocouple to the PID at all? The wires crimp onto lug connectors. These lugs, I assume, are standard lugs not made out of specific material. Then, the lugs are screwed onto metal screws on the PID, which, again I assume are not made out of specifica materials. THEN....inside the PID there is traces and connections on the boards before it actually gets to its final destination in the logic, right?

How does this work? Its the same concept as cutting the cable in the middle and soldering in a copper wire. The metal type changes two or three times already between the sensor itself and the inside of the PID.

Click to expand...

Depending on the unit, it will either have an internal temperature probe that they use to correct for the cold junction, or it will apply a constant correction and assume the controller will always be the same temperature. They may also ignore this transition, and assume the voltage is correct.

The cold junction is where you normally transition from thermocouple to copper. Normally 0 degrees c is used for this reference. If the cold junction occurs at 70 degrees c and is not corrected, the reading is going to be of by about one degree. For some applications, this amount of inaccuracy may be acceptable.

I'm an engineer in the aero industry, and our industry accuracy requirements are just a little more strict. I'm used to dealing with systems that have a 0.25 degrees of accuracy or less. I doubt you really need greater accuracy for brewing, however my experience in brewing is still pretty low.

Might just be a case of analysis paralysis. I'm thinking about it too hard and approaching it from the thought that any inaccuracy is unacceptable.

MagooBrew said:

Go to omega.com they sell t/c diconnects for panel mount for just a few dollars. These are made for this purpose. I have always had good luck with their products.

Click to expand...

Plus you get free Dilbert comics with every order.

I also get a kick out of their packaging tape.

I run type K TCs and often use extensions to get from various parts of machines to a data acquistion unit. Most of my supplies are ordered from Omega, including the TC mini-plug connectors and the TC extension wire. There seems to be effectively no change when running these extensions over lengths of 20-25 feet, give or take. Have not tried longer as there is no point for my use, but I suppose I could put a jack and plug on the ends of a spool of TC wire and see what happens.

Can you guys show me what you mean by these plugs and jacks that you are using? everything on this site is completely different looking to me, having a hard time finding what I should be using that isnt super overkill.

mpj-k-f for a mini panel jack for use with type k thermocouple.

Nmp-k-m for a mini plug for type K thermocouple

Then you don't have to worry about any extra cold junctions.

DougK said:

Depending on the unit, it will either have an internal temperature probe that they use to correct for the cold junction, or it will apply a constant correction and assume the controller will always be the same temperature. They may also ignore this transition, and assume the voltage is correct.

The cold junction is where you normally transition from thermocouple to copper. Normally 0 degrees c is used for this reference. If the cold junction occurs at 70 degrees c and is not corrected, the reading is going to be of by about one degree. For some applications, this amount of inaccuracy may be acceptable.

I'm an engineer in the aero industry, and our industry accuracy requirements are just a little more strict. I'm used to dealing with systems that have a 0.25 degrees of accuracy or less. I doubt you really need greater accuracy for brewing, however my experience in brewing is still pretty low.

Might just be a case of analysis paralysis. I'm thinking about it too hard and approaching it from the thought that any inaccuracy is unacceptable.

Click to expand...

The problem is bigger than you might think. The Auber PIDs have a reference diode to attempt to correct for the temp of the cold junction. The cold junction is assumed to be at the back of the PID, so the reference is placed near the vent holes (you can actually see a leaded diode there). This works fine unless the temp at the reference diode is much different than the temp at the cold junction.

You can alleviate the problem by using thermocouple wire from the correct thermocoulple socket back to the PID. In my case, I also have a switch to select which TC my PID is looking at, so the switch became the cold junction even though I had the correct wire.

So, if you move the cold junction to, say, the outside of the control panel, then you might have a problem when the inside of the control panel starts to heat up.

So, the moral of the story is, if you can keep the back of the PID about the same temperature as your thermocouple plug, wherever that is, you'll be OK. If not, i.e., if you the back of your PID is near a hot SSR, you'll see several degrees of inaccuracy as the box heats up. Ask me how I know this.

BTW, I got a dual yellow TC plug on mcmaster carr. Here http://www.mcmaster.com/#thermocouple-connectors/=ld62bi