Author Topic: Thermocouople thermometers.  (Read 2157 times)

0 Members and 1 Guest are viewing this topic.

Offline 6PTsocketTopic starter

  • Regular Contributor
  • *
  • Posts: 212
Thermocouople thermometers.
« on: December 26, 2018, 07:03:27 am »
I have a K type thermocouple temperature function in my DMM. It is even ice bath calibrateable. Simple explanations I have found compare the temp of the reference  thermocouple to the measurement thermocouple. This explains why taking the dmm outside on a cold day produces drifting readings as the reference thermocouple in the dmm gets cold, too. I also have an instant read kithchen thermometer that does not seem to have this problem. They make expensive thermocouple thermometers for BBQ competitions where they are outside in all kinds of weather and are still accurate. I would doubt the reference is in some kind of temperature controlled oven. Anybody know how they do it? Thanks.

Sent from my SM-G900V using Tapatalk

 

Online IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Thermocouople thermometers.
« Reply #1 on: December 26, 2018, 07:08:51 am »
Other thermometers may use a variable resistance element (thermistor) rather than a thermocouple. The reference junction temperature measurement in a thermocouple meter will also use a thermistor.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Thermocouople thermometers.
« Reply #2 on: December 26, 2018, 07:26:25 am »
I have a couple of K type Thermocouples fitted into a Coffee Roaster that go through a pair of Multimeters for Logging data. I had to move the meters themselves away from the roaster as the ambient was killing the accuracy. So the key is to keep the electronics close to ambient temps to improve accuracy.

One more modern method https://www.maximintegrated.com/en/app-notes/index.mvp/id/4026
A more traditional approach https://www.omega.co.uk/techref/thermoref.html

The Omega Site is well worth having a good look around too  :)
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Thermocouople thermometers.
« Reply #3 on: December 26, 2018, 10:59:07 am »
A thermocouple thermometer has *two* thermocouples in series.  One is at the probe's measurement point and the other is at the "cold junction block" inside the multimeter where the thermcouple wires connect to the internal circuitry.  The temperature of the cold junction block is measured with another temperature sensor to compensate for its effect on the thermocouple's measurement.

If the thermocouple probe attaches to the multimeter banana jacks using an adapter, then it is part of the cold junction block also and it takes longer for the cold junction block temperature to settle.
 

Offline 6PTsocketTopic starter

  • Regular Contributor
  • *
  • Posts: 212
Re: Thermocouople thermometers.
« Reply #4 on: December 26, 2018, 02:27:23 pm »
A thermocouple thermometer has *two* thermocouples in series.  One is at the probe's measurement point and the other is at the "cold junction block" inside the multimeter where the thermcouple wires connect to the internal circuitry.  The temperature of the cold junction block is measured with another temperature sensor to compensate for its effect on the thermocouple's measurement.

If the thermocouple probe attaches to the multimeter banana jacks using an adapter, then it is part of the cold junction block also and it takes longer for the cold junction block temperature to settle.
Thanks. Look at the link from beanflying (above) it gets heavily into cold junction compensation methods from hardware, software and lookup tables. I understand the basic circuit. It was the compensation that I was ignorant about.

Sent from my SM-G900V using Tapatalk

 

Offline 6PTsocketTopic starter

  • Regular Contributor
  • *
  • Posts: 212
Re: Thermocouople thermometers.
« Reply #5 on: December 26, 2018, 02:28:02 pm »
I have a couple of K type Thermocouples fitted into a Coffee Roaster that go through a pair of Multimeters for Logging data. I had to move the meters themselves away from the roaster as the ambient was killing the accuracy. So the key is to keep the electronics close to ambient temps to improve accuracy.

One more modern method https://www.maximintegrated.com/en/app-notes/index.mvp/id/4026
A more traditional approach https://www.omega.co.uk/techref/thermoref.html

The Omega Site is well worth having a good look around too  :)
Thanks, those are very good links.

Sent from my SM-G900V using Tapatalk

 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 15151
  • Country: de
Re: Thermocouople thermometers.
« Reply #6 on: December 26, 2018, 03:11:20 pm »
The usual way is to use compensation with a different type of thermometer.

A special case is a type of Pt based thermocouple. This has very low thermal EMF around room temperature and thus does not care much about the cold junction temperature, but also only works from some 300 C up.  So if the cold junction is not too extreme it can get away without compensation.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Thermocouople thermometers.
« Reply #7 on: December 26, 2018, 03:30:01 pm »
Before I moved my meters on the roaster the ambient around them was about 50C so it could have been the cold junction or even other bits of the circuit not liking the temperatures.

I have a prototype controller running at present based on the MAX6675 https://datasheets.maximintegrated.com/en/ds/MAX6675.pdf and Arduino style board to get rid of the meters from the system all together. Walk up temperature consistency week to week is more important than ultimate accuracy in this case so far so good.  :)
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline graeme.c.payne

  • Newbie
  • Posts: 8
Re: Thermocouople thermometers.
« Reply #8 on: December 29, 2018, 03:35:40 pm »
First time posting here.

An interesting discussion of temperature measurement. I would like to make a couple of points more clear.

1. Thermocouples do not "measure temperature." A thermocouple produces a voltage proportional to the DIFFERENCE in temperature between the junction and the open end. The open end of the thermocouple should be connected to an isothermal junction block. That keeps the ends isolated from each other, and at a known and stable temperature. The temperature of the junction block is measured by a different method, like a thermistor or a platinum thermometer. That way the measuring circuit can determine the temperature difference.

The greater the difference between the measuring junction and the isothermal block, the better the measurement will be.

In a laboratory situation the isothermal block may be replaced by a thermocouple held at the triple point point of water (0.01 °C). This point is one of the 14 fixed point references of the International Temperature Scale, and certainly the easiest for most people to achieve.

(The triple point of water is the temperature at which water can exist as a gas, liquid and solid.)

2. A high quality (expensive) platinum resistance thermometer is the specified reference thermometer for the majority of the International Temperature Scale. Mass-produced  commercial platinum thermometers are not as good, but generally still better than thermocouples or thermistors.

My qualification: 30+ years in tthe maintenance, repair and calibration of electronic test and measuring equipment.


Sent using Tapatalk

 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Thermocouople thermometers.
« Reply #9 on: December 29, 2018, 04:23:24 pm »
The usual way is to use compensation with a different type of thermometer.

A special case is a type of Pt based thermocouple. This has very low thermal EMF around room temperature and thus does not care much about the cold junction temperature, but also only works from some 300 C up.  So if the cold junction is not too extreme it can get away without compensation.
A Pt is not a thermocouple.  It is a platinum resistor.  They do an excellent job. 
 

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2431
  • Country: mx
Re: Thermocouople thermometers.
« Reply #10 on: December 29, 2018, 06:37:26 pm »
I also suggest that you read the white papers available at Omega.com.
They are an excellent information source.


Like every technical decision one makes, one has to make tradeoffs and therefore depending on the particular application, performance requirements and cost, one can make an informed decision.
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Thermocouople thermometers.
« Reply #11 on: December 29, 2018, 08:04:34 pm »
The usual way is to use compensation with a different type of thermometer.

A special case is a type of Pt based thermocouple. This has very low thermal EMF around room temperature and thus does not care much about the cold junction temperature, but also only works from some 300 C up.  So if the cold junction is not too extreme it can get away without compensation.
A Pt is not a thermocouple.  It is a platinum resistor.  They do an excellent job.

Why did you disagree with Kleinstein, when they specifically used the word thermocouple in their post?

See the link below and look at "Type B" thermocouples:

http://www.capgo.com/Resources/Temperature/Thermocouple/Thermocouple.html

Quote
Good at high temperatures, no reference junction compensation required.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Thermocouople thermometers.
« Reply #12 on: December 29, 2018, 11:14:48 pm »
I think that the confusion here is that there is a PT100 platinum resistance temperature sensor. It has a resistance of 100 ohms at 0 degrees C and is very linear and accurate and usable to about 500 degree C. The PT thermocouple has two different alloys and does produce a voltage.   
 
The following users thanked this post: schmitt trigger

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Thermocouople thermometers.
« Reply #13 on: December 29, 2018, 11:31:12 pm »
Edit removed quotes as I had them jumbled

Kleinsteins "Pt Thermocouple" isn't helpful as it could give the impression to some that it was a Pt only device (so not actually a thermocouple) and the most common of those is the RTD. The Type B, R and S Thermocouples are highly specialized and not common by comparison.

As Arther stated these are Platinum and Rhodium based thermocouples in the link above and as such generate a voltage as distinct from PT100, PT1000 etc. which are just Platinum and vary resistance. Both have pros and cons but that is outside the scope of what the OP was chasing information on.

Link just for information on Resistance Thermometer types https://en.wikipedia.org/wiki/Resistance_thermometer or the Omega site will have more in depth information.
« Last Edit: December 30, 2018, 12:40:46 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Thermocouople thermometers.
« Reply #14 on: December 30, 2018, 05:34:41 am »
The usual way is to use compensation with a different type of thermometer.

A special case is a type of Pt based thermocouple. This has very low thermal EMF around room temperature and thus does not care much about the cold junction temperature, but also only works from some 300 C up.  So if the cold junction is not too extreme it can get away without compensation.
A Pt is not a thermocouple.  It is a platinum resistor.  They do an excellent job.


Why did you disagree with Kleinstein, when they specifically used the word thermocouple in their post?

See the link below and look at "Type B" thermocouples:

http://www.capgo.com/Resources/Temperature/Thermocouple/Thermocouple.html

Quote
Good at high temperatures, no reference junction compensation required.
Because he was talking about a Pt basted "thermocouple".  The Pt term is used for RTD type platinum temperature sensors. 
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Thermocouople thermometers.
« Reply #15 on: December 30, 2018, 05:46:03 am »
It could have been read either way was the point. A thermocouple by definition is a couple and not a single. You read it one way and he clearly meant it in another and Platinum is their major material of construction alloyed with a small percentage of Rhodium for one side for B, R & S types.

If anyone said to me Platinum temperature sensor or platinum thermometer I would generally think they were referring to RTD's and not the Three main Platinum based Thermocouples.

RTD's are never referred to as Thermocouples.

« Last Edit: December 30, 2018, 05:52:56 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 
The following users thanked this post: ahbushnell

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Thermocouople thermometers.
« Reply #16 on: December 31, 2018, 06:05:50 pm »
I just assumed from context that it was a Pt RTD.  There is no reason to use a Pt thermocouple for cold junction compensation but a Pt RTD is a great option.
 
The following users thanked this post: ahbushnell

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 15151
  • Country: de
Re: Thermocouople thermometers.
« Reply #17 on: December 31, 2018, 06:31:11 pm »
Sorry for confusing with the Pt base thermocouples.  :palm:

The Tpye B and S thermocouples  (2 platinum alloys) are still quite popular for high temperatures.

For cold junction compensation one usually has a large choice of sensors, as the temperature in moderate and the accuracy of thermocouples is usually not very high. So in most cases one can use rather cheap semiconductor based sensors for the cold junction (e.g. a diode). A platinum RTD would be kind of overkill in most cases.
 
The following users thanked this post: schmitt trigger


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf