Thermocouple Switching

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
Lenp
Posts: 1464
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Thermocouple Switching

Post by Lenp » Wed Jan 19, 2011 9:28 am

Project time again!
Let's see if my thinking is in the ballpark

5 thermocouple probes, (type J or K) sequentially switched to one display. I can do the logic for the switching and hope to use relays to select the T/C probes.

I know dissimilar junctions will cause an error in thermocouple circuits. If I use T/C rated wire for the connections and a 2 pole relay with both the + and - T/C leads through the contacts wouldn't the errors cancel since both contact junctions would be at the same temp?

Humm...
Len
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Thermocouple Switching

Post by MrAl » Wed Jan 19, 2011 12:23 pm

Hi there Len,

Just guessing here, but i would think that a multimeter made for use with thermocouple probes would have a temperature sensing device inside the meter to measure the 'cold' junction temperature. This could be right at the connector or on a circuit board i guess.
Using ordinary wire would couple any electrical signal, but would not properly couple the heat to the inside of the meter. This i think would cause a problem.

There's nothing stopping you from trying it out though, using a very simple experiment. Measure some distilled ice water with distilled water ice in it with the meter in the usual way, noting the temperature. Next, connect two wires, one to each probe connector pin and connect them to the meter. This gives you a two wire jumper from the probe to the meter. All you have to do now is measure the ice water again and see if the reading is the same.
Note that it may be the same if the inside of the meter is at the same temperature as the probe connector, but you'll really have to try this out.

Let us know how it works if you do decide to try it.

As a side note, you can make your own meter and that would give you the ability to access several probes all with their own cold junction compensators.
LEDs vs Bulbs, LEDs are winning.

jimmy101
Posts: 82
Joined: Tue Mar 24, 2009 11:08 am
Contact:

Re: Thermocouple Switching

Post by jimmy101 » Thu Jan 20, 2011 10:01 am

If you are doing logic switching I would think you might as well use an analog mux/demux instead of a relay. Use a pair of MUXs, one for the reading thermocouples and another for the cold junctions. That should largely cancel any EMFs generated by the wire junctions.

User avatar
jwax
Posts: 2205
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Re: Thermocouple Switching

Post by jwax » Sat Jan 22, 2011 6:06 am

I've switched T/C's through standard wafer switches with no problem.
Once the millivolts are generated at the junction, they travel through switches and relays just fine.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Thermocouple Switching

Post by MrAl » Wed Jan 26, 2011 2:10 am

Hi there,


Actually though, the voltages are not generated across junctions, the voltages are generated across the lengths of wire.

If we had a way to measure the voltage generated across one single strand of wire with one end hot and the other end cold without incidentally also heating the wire probe we use to make the measurement, we could use a single wire to make temperature measurements. Unfortunately, any attempt to measure the voltage across a wire heated at one end with any type of probe that also conducts electrical current (so we can make the voltage measurement) also means we end up heating the end of the wire used to make the measurement, which means that wire too has one hot end and one cold end which means another voltage drop occurs which ruins the measurement. This is why we end up using two different metals of known and different temperature vs voltage characteristics so that we can subtract the two effects and still end up with a usable voltage to measure at the cold ends.

What this means is that any meter that is going to use two different metal wires to measure temperature must have a means to measure the cold ends of the two wires, and the two cold ends can not touch but must have good thermal coupling so that they stay both at the same temperature. This also means that if we stick a two foot piece of two conductor copper wire in between the meter and the ends of the thermocouple wire (connector) the meter will see the cold ends of the two copper wires, not the two ends of the thermocouple wires. Thus, if there are any significant temperature differences between the meter input terminals and the ends of the thermocouple wires (connector) there will be an error before we even start. If we get lucky and the ends of the copper wires are the same temperature as the ends of the thermocouple wires (connector) then we wont get an error, but of course that's hard to be sure of in practice, and we have to remember that the wire will have a thermal gradient across it from end to end but only up to the point where the copper wire begins, and then it will have a different thermal gradient that the meter doesnt know about.
LEDs vs Bulbs, LEDs are winning.

User avatar
jwax
Posts: 2205
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Re: Thermocouple Switching

Post by jwax » Wed Jan 26, 2011 8:48 am

Do let us know how you do it, and how it works, Lenp!

User avatar
reloadron
Posts: 519
Joined: Sat Jun 28, 2008 8:57 am
Location: Cleveland, Ohio
Contact:

Re: Thermocouple Switching

Post by reloadron » Wed Jan 26, 2011 3:53 pm

Hi Ya Lemp. You may want to give this a read. When you get through it there is a good example of multiplexing thermocouples using one of these as the mux. Looks like it should work. I would not use relays to switch the TC inputs. I have used switches designed for thermocouples with good results but given a choice today I might look to what I linked to. Looks promising anyway.

Ron

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: Thermocouple Switching

Post by Bob Scott » Thu Jan 27, 2011 7:24 am

Omega: is a company. All they do is temerature sensing and equipment sales of all types of sensors. They have a complete library with anthing you would ever want to know about themocouples. Check out the technical reference section.

http://www.omega.com/
-=VA7KOR=- My solar system includes Pluto.

User avatar
Lenp
Posts: 1464
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Re: Thermocouple Switching UPDATE

Post by Lenp » Tue Feb 08, 2011 1:54 am

Thanks to all, here's an update...

All things considered I've decided that thermocouples are probably not the way to go. Too much hassle with compensations and switching issues. Over a wide span thermocouples are pretty good, especially at high temperatures, but in a tight span at lower temperature there are wider errors. The PT100 sensors are resistive, more accurate and a lot more linear so would likely be better in this application. One of the main reasons for switching was cost savings by not using separate displays. I have found a remarkable PID controller that accepts multiple types of sensors, and is under $40! (More accolades possible after testing) The control features are unneeded but at that price it is less than a comparable panel display. I may still consider switching if the PT100 errors are reasonably insignificant but because of the controller's small size, I may decide to use individual units.

Len

BTW: Omega is a full line supplier of process controls and instrumentation, in addition to temperature!
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
reloadron
Posts: 519
Joined: Sat Jun 28, 2008 8:57 am
Location: Cleveland, Ohio
Contact:

Re: Thermocouple Switching

Post by reloadron » Tue Feb 08, 2011 5:37 am

Hi Ya Lemp

If you want to reconsider the use of thermocouples I may have something that can help you out. I do quite a bit of work with temperature. Several years ago we converted a few large systems from Type J to Type K. The systems were using Temperature Transmitters similar to those carried by Omega and a dozen other suppliers. I refer to them as "Hockey Pucks". They have a thermocouple input (these are all type J) and provide a nice compensated linear current output of 4 to 20 mA. These were early units and fixed for Type J with a range of 35 to 700 degrees F. They are the TT221 series that can be found here. From that link there is a link to the manual for these units. They were heading for the dumpster when I grabbed them.

Units like this are powered using a 24 volt supply. They are nice because all the CJC is done and they provide a nice linear output. The 4 to 20 mA can then easily be run through a 250 or 500 ohm resistor to provide 1 to 5 volts or 2 to 10 volts. So in this case using these temperature transmitters you end up with 35 deg F. to 700 deg F. = 4 to 20 mA or 1 to 5 volts or 2 to 10 volts. The 1 to 5 being very useful for the A/D of most micro controllers.

A few years ago I did some work helping a computer hardware site involving hardware. The results of some of that experiment can be found here. I set things up for computer temperature data acquisition using one of these starter kits which also include some nice charting software. I wrote my own using VB but either way not bad.

Actually the accuracy is very good as they can easily be calibrated. The only real limiting factor is the 10 bit resolution of the A/D.

You mentioned 5 thermocouple channels. If you want to go this route I will send you five of these things. I have a pile of them and during my lifetime will never use them all. I don't want anything for them as in Free. I only ask that you use and enjoy them. If you have questions just ask.

Ron

User avatar
reloadron
Posts: 519
Joined: Sat Jun 28, 2008 8:57 am
Location: Cleveland, Ohio
Contact:

Re: Thermocouple Switching

Post by reloadron » Sun Feb 13, 2011 5:55 am

I replied to your PM, did you get it?

Ron

User avatar
Lenp
Posts: 1464
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Re: Thermocouple Switching

Post by Lenp » Mon Feb 14, 2011 3:56 am

Ron,
Last message from you was Feb 8. which I responded to. Check for new mail.
Len
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
reloadron
Posts: 519
Joined: Sat Jun 28, 2008 8:57 am
Location: Cleveland, Ohio
Contact:

Re: Thermocouple Switching

Post by reloadron » Mon Feb 14, 2011 2:23 pm

Got it, give me a few days as my shipping department (wife) had an argument with the ice and lost. She sends things from work and should be back there in a few days.

Ron

User avatar
Lenp
Posts: 1464
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Re: Thermocouple Switching

Post by Lenp » Mon Feb 14, 2011 2:36 pm

Ron,

Absolutely no rush, wait till the thaw!

Len
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
Lenp
Posts: 1464
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Re: Thermocouple Switching

Post by Lenp » Thu Feb 24, 2011 12:26 pm

Project update:

Thermocouples lost, PT100 sensors won.
I located a supplier Auberins.com that supplied a great quality PID controller and very small PT100 sensors at at a decent price. In fact the price of the controllers, $36 made it possible to use a separate controller for each channel rather than the switching. And, as a benefit, the controller can operate a status indicator and they are tracking at at .1°C!
Once the project finishes I may post a pic or two..

Len
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

Post Reply

Who is online

Users browsing this forum: No registered users and 44 guests