Does there exists a sort of shrink tubing with good thermal conductivity?
Jan Panteltje
exchanger temperature. (expected temperature range between say 10 C and 70 C).
I found a small hole where I can put in a tiny thermocouple, say 2 mm opening.
But this thermocouple needs to be isolated from the metal of the heat exchanger.
So the question is, is there some sort of tubing I can
put over it with very good thermal conductivity?
Then I could just stick the thermocouple in it and glue it in place.
John Fields
---
Why go through the hassle when tiny glass bead thermistors are
available that you can epoxy into the hole?
http://www.ge-mcs.com/download/temperature/920_308a.pdf
---
JF
tm
"John Fields" <jfi...@austininstruments.com> wrote in message
news:euu6j69lpg2t86to4...@4ax.com...
Or dip the thermocouple in molten glass.
tm
Jan Panteltje
<jfi...@austininstruments.com> wrote in
<euu6j69lpg2t86to4...@4ax.com>:
Becaue there is already one in there?????? (see above text).
I want a THERMOCOUPLE, as I have a thermocouple interface.
lektr...@gmail.com
Omega Engineering (http://www.omega.com/toc_asp/subsectionSC.asp?
subsection=A07&book=Temperature) sells stainless sheathed
thermocouples. With a reasonable probe and lead length, they're not
too expensive. I think the last we bought were under $25. If you
want the fastest transition time, go with a grounded probe, otherwise
go with ungrounded (un/grounded means the gunction is in contact w/the
sheath).
Jan Panteltje
<the_ob...@whitehouse.gov> wrote in <igvume$kp$1...@speranza.aioe.org>:
Yes that would work.
Just have to go and melt some glass.
Anywhere between 570 and 1470 C IIRC.
Its a 'cool' idea, may just try it
John Fields
---
Problem with temperature coefficient of expansion between the metal
and the glass, I think, so the glass will crack off as it cools.
But, dipping the thermocouple into some beryllia loaded epoxy and then
centering it in a teflon tube until the epoxy cures might work.
---
JF
Jan Panteltje
"lektr...@gmail.com" <lektr...@gmail.com> wrote in
<349fe460-5a5b-49b8...@w2g2000yqb.googlegroups.com>:
>On Jan 16, 4:27 pm, Jan Panteltje <pNaonStpealm...@yahoo.com> wrote:
>> This cryocooler I am playing with has a NTC of unknown specs for the heat
>> exchanger temperature. (expected temperature range between say 10 C and 7=
>0 C).
>> I found a small hole where I can put in a tiny thermocouple, say 2 mm ope=
>ning.
>> But this thermocouple needs to be isolated from the metal of the heat exc=
>hanger.
>> So the question is, is there some sort of tubing I can
>> put over it with very good thermal conductivity?
>> Then I could just stick the thermocouple in it and glue it in place.
>
>Omega Engineering (http://www.omega.com/toc_asp/subsectionSC.asp?
>subsection=A07&book=Temperature) sells stainless sheathed
>thermocouples. With a reasonable probe and lead length, they're not
>too expensive. I think the last we bought were under $25. If you
>want the fastest transition time, go with a grounded probe, otherwise
>go with ungrounded (un/grounded means the gunction is in contact w/the
>sheath).
Yes, I know about omega.com, but I am making my own thermocouples
from transformer wire and resistor wire, works perfectly,
I twist the ends together and then solder those.
Costs nothing (wire comes from an old wire wound resistor,
and the transformer wire from an old transformer).
http://panteltje.com/panteltje/pic/th_pic/home_made_thermocouple_connector_img_2373.jpg
that one has some other multi-stranded copper, was a first try, latest ones are much thinner.
Seems to be within one degree C.
tm
The glass coating can be very thin and flexible.
Maybe even a glass frit and fire it with a torch.
tm
Dave M
Place a layer of Kapton film between the couple and what you're monitoring.
Then wrap it with heat shrink tubing or tape.
--
David
dgminala at mediacombb dot net
lektr...@gmail.com
> Yes, I know about omega.com, but I am making my own thermocouples
> from transformer wire and resistor wire, works perfectly,
> I twist the ends together and then solder those.
> Costs nothing (wire comes from an old wire wound resistor,
> and the transformer wire from an old transformer).
> http://panteltje.com/panteltje/pic/th_pic/home_made_thermocouple_conn...
> that one has some other multi-stranded copper, was a first try, latest ones are much thinner.
> Seems to be within one degree C.
Well, that is just too cool! I've made a lot of motors and coils from
wire salvaged from deflection coils out of old TVs, but can't remember
making my own scratch-built thermocouples! Now I see the tie-in from
where you were making your own instruments and getting calibration
curves (I only read the group occasionally...). A long time ago, I
worked at Texas Instruments, and we made our own TCs from TC wire, but
welded the junctions together using a special hydrogen/oxygen torch.
All the twisted pairs would fail when used in acid baths. I bet the
metal sheathing idea would still work, assuming the wires were
insulated except for the tip. Metal-filled epoxy might be another way
to go. Can you get JB Weld or similar?
GRe
"Jan Panteltje" <pNaonSt...@yahoo.com> wrote in message
>
[...]
>
> Yes, I know about omega.com, but I am making my own thermocouples
> from transformer wire and resistor wire, works perfectly,
> I twist the ends together and then solder those.
> Costs nothing (wire comes from an old wire wound resistor,
> and the transformer wire from an old transformer).
> http://panteltje.com/panteltje/pic/th_pic/home_made_thermocouple_connector_img_2373.jpg
> that one has some other multi-stranded copper, was a first try, latest
> ones are much thinner.
> Seems to be within one degree C.
What T-coefficient did you find for your DIY thermocouple?
With a quick and dirty test I found about 38猩/蚓 for a TC
using resistance wire (Monacor 62.4?/m 0.1mm) & roadrunner
prototyping wire (lacquered copper wire 0.2mm).
Regards, Gio
Jan Panteltje
<dgmina...@mediacombb.net> wrote in
<8eqdnXtTyc4mPK7Q...@giganews.com>:
Yes, that could work.
I will experiment a little later with the variuos methods suggested.
Jan Panteltje
"lektr...@gmail.com" <lektr...@gmail.com> wrote in
<65441d61-6809-4f08...@l17g2000yqe.googlegroups.com>:
>On Jan 16, 5:53 pm, Jan Panteltje <pNaonStpealm...@yahoo.com> wrote:
>
>> Yes, I know about omega.com, but I am making my own thermocouples
>> from transformer wire and resistor wire, works perfectly,
>> I twist the ends together and then solder those.
>> Costs nothing (wire comes from an old wire wound resistor,
>> and the transformer wire from an old transformer).
>> http://panteltje.com/panteltje/pic/th_pic/home_made_thermocouple_conn.=
>..
>> that one has some other multi-stranded copper, was a first try, latest on=
>es are much thinner.
>> Seems to be within one degree C.
>
>Well, that is just too cool! I've made a lot of motors and coils from
>wire salvaged from deflection coils out of old TVs, but can't remember
>making my own scratch-built thermocouples! Now I see the tie-in from
>where you were making your own instruments and getting calibration
>curves (I only read the group occasionally...). A long time ago, I
>worked at Texas Instruments, and we made our own TCs from TC wire, but
>welded the junctions together using a special hydrogen/oxygen torch.
>All the twisted pairs would fail when used in acid baths. I bet the
>metal sheathing idea would still work, assuming the wires were
>insulated except for the tip. Metal-filled epoxy might be another way
>to go. Can you get JB Weld or similar?
I could probably weld the ends, but this works just fine.
Yes all this stuff is part of a larger plan...
I am not using any acid or that, just very cold stuff, and up to maybe 70 C.
Jan Panteltje
<mynamespa...@xs4all.nl> wrote in
<4d3426e9$0$81476$e4fe...@news.xs4all.nl>:
>
>"Jan Panteltje" <pNaonSt...@yahoo.com> wrote in message
>>
>[...]
>>
>> Yes, I know about omega.com, but I am making my own thermocouples
>> from transformer wire and resistor wire, works perfectly,
>> I twist the ends together and then solder those.
>> Costs nothing (wire comes from an old wire wound resistor,
>> and the transformer wire from an old transformer).
>> http://panteltje.com/panteltje/pic/th_pic/home_made_thermocouple_connector_img_2373.jpg
>> that one has some other multi-stranded copper, was a first try, latest
>> ones are much thinner.
>> Seems to be within one degree C.
>
>What T-coefficient did you find for your DIY thermocouple?
>With a quick and dirty test I found about 38µV/°C for a TC
>using resistance wire (Monacor 62.4?/m 0.1mm) & roadrunner
>prototyping wire (lacquered copper wire 0.2mm).
>
>Regards, Gio
It is a type T,
I just compard it by puting it on my temperature controlled soldering iron,
it has a digital display, and it was tracking to +190 C within 1 degree C.
It is porbably better, but the biggest problem was keeping a good thermal contact.
I did not go any higher as I do not need that high, and the electronics
is designed to go lower, not higher:
http://panteltje.com/panteltje/pic/th_pic/
John Fields
<pNaonSt...@yahoo.com> wrote:
>On a sunny day (Sun, 16 Jan 2011 17:16:16 -0600) it happened John Fields
><jfi...@austininstruments.com> wrote in
><euu6j69lpg2t86to4...@4ax.com>:
>
>>On Sun, 16 Jan 2011 22:27:26 GMT, Jan Panteltje
>><pNaonSt...@yahoo.com> wrote:
>>
>>>This cryocooler I am playing with has a NTC of unknown specs for the heat
>>>exchanger temperature. (expected temperature range between say 10 C and 70 C).
>>>I found a small hole where I can put in a tiny thermocouple, say 2 mm opening.
>>>But this thermocouple needs to be isolated from the metal of the heat exchanger.
>>>So the question is, is there some sort of tubing I can
>>>put over it with very good thermal conductivity?
>>>Then I could just stick the thermocouple in it and glue it in place.
>>
>>---
>>Why go through the hassle when tiny glass bead thermistors are
>>available that you can epoxy into the hole?
>
>Becaue there is already one in there?????? (see above text).
>I want a THERMOCOUPLE, as I have a thermocouple interface.
---
Well, then, since the dimensions are so tiny, I don't think that
thermal conductivity is going to be a problem, particularly.
Try tiny polyimide (Kapton) tubing, like you can get here:
---
JF
J. Todd
says...
George Herold
Thermo couples leak a lot of heat via the leads. So you cna get some
funny readings if the leads are in air that is a much different
temperature from your piece of metal. (Wrap the leads onto the metal
piece if you have room.)
I found Sil pads have a much better thermal conductivity than thin
Kapton tape.
(For wahtever that is worth.)
George H.
Jan Panteltje
<jfi...@austininstruments.com> wrote in
<ddp8j6t5oa802rckc...@4ax.com>:
>On Sun, 16 Jan 2011 23:38:47 GMT, Jan Panteltje
><pNaonSt...@yahoo.com> wrote:
>
>>On a sunny day (Sun, 16 Jan 2011 17:16:16 -0600) it happened John Fields
>><jfi...@austininstruments.com> wrote in
>><euu6j69lpg2t86to4...@4ax.com>:
>>
>>>On Sun, 16 Jan 2011 22:27:26 GMT, Jan Panteltje
>>><pNaonSt...@yahoo.com> wrote:
>>>
>>>>This cryocooler I am playing with has a NTC of unknown specs for the heat
>>>>exchanger temperature. (expected temperature range between say 10 C and 70 C).
>>>>I found a small hole where I can put in a tiny thermocouple, say 2 mm opening.
>>>>But this thermocouple needs to be isolated from the metal of the heat exchanger.
>>>>So the question is, is there some sort of tubing I can
>>>>put over it with very good thermal conductivity?
>>>>Then I could just stick the thermocouple in it and glue it in place.
>>>
>>>---
>>>Why go through the hassle when tiny glass bead thermistors are
>>>available that you can epoxy into the hole?
>>
>>Becaue there is already one in there?????? (see above text).
>>I want a THERMOCOUPLE, as I have a thermocouple interface.
>
>---
>Well, then, since the dimensions are so tiny, I don't think that
>thermal conductivity is going to be a problem, particularly.
I found tha tthe slightes amount of air, or play between thermocuple
and object, will severely degrade the thermal contact.
>Try tiny polyimide (Kapton) tubing, like you can get here:
Yes thin walled would probably work, and then with heat sink paste.
I will try out some things later.
Fred Bartoli
"Heat leak" depends a lot on configuration.
One good trick is to have the thermocouple glued as deep as possible in
a low diameter bored hole :
| ____ Fill with glue
/---------------' /
< x============================== thermocouple wires
\---------------.
|
Doing so, the 'radial' thermal resistance of glue will be lower than the
longitudinal thermal resistance of the thermocouple wires and the solder
temperature will be what you want. IOW make sure the leaked heat doesn't
come from the solder environment.
Just use small bifilar thermo wires, not the heavy shielded one, small
heat shrink tubing, use an as small as practical hole diameter filled
with glue and you'll be OK.
--
Thanks,
Fred.
John Fields
<ghe...@teachspin.com> wrote:
>On Jan 16, 5:27 pm, Jan Panteltje <pNaonStpealm...@yahoo.com> wrote:
>> This cryocooler I am playing with has a NTC of unknown specs for the heat
>> exchanger temperature. (expected temperature range between say 10 C and 70 C).
>> I found a small hole where I can put in a tiny thermocouple, say 2 mm opening.
>> But this thermocouple needs to be isolated from the metal of the heat exchanger.
>> So the question is, is there some sort of tubing I can
>> put over it with very good thermal conductivity?
>> Then I could just stick the thermocouple in it and glue it in place.
>
>Thermo couples leak a lot of heat via the leads. So you cna get some
>funny readings if the leads are in air that is a much different
>temperature from your piece of metal. (Wrap the leads onto the metal
>piece if you have room.)
---
How, then, can you explain a thermocouple welded to a block of metal
at 500C and the meter reading 500C while being 100' away from the hot
block, with the leads in 27C air?
---
JF
George Herold
> On Mon, 17 Jan 2011 09:43:37 -0800 (PST), George Herold
>
Hi John, Jan is doing something where the thermocouple is not welded
to the piece of metal. The thermal voltage in generated right at the
junction. But if the junction is not in good thermal contact with the
DUT then heat can be conducted via the leads to the air.... and thus
the junction is at a different temperature than the DUT. The actual
temperature of the junction is a balance between the thermal
conductivity to the DUT and that of the leads to the air. 100'
doesn't matter it's the few inches right near the junction.
No matter what the thermometer it's always a good practice to think
about heat leaks via the leads. (Essential if you are doing low
temperature stuff in vacuum. A lessen learned years ago when I had to
redo most of the electrical wiring on a probe I was making.) Low
temperture stuff can be a PITA. A day to cool things down, a day to
see that something isn't working, A day to warm up. Fix and repeat.
The weeks pass pretty fast that way...
George H.
George Herold
wrote:
>
> - Show quoted text -
That's great Fred. We make little thermistor probes, (with 'big'
thermistor heads), part of the lead hole is filled with an aluminum
tube. Tube OD a bit bigger than thermistor head. We fill the gaps
with thermal grease. Any suggestions for the glue?
(silly question?)
Metals are such great thermal conductors are super conductors also
super-thermal conductors?
George H.
Rich Grise
>
> (silly question?)
> Metals are such great thermal conductors are super conductors also
> super-thermal conductors?
>
Helium-3, which is a heat superconductor - it doesn't boil, the atoms
just leave the surface since there are no hot spots.
Cheers!
Rich
G
I don't think thermal conductivity is an issue. It will
stabilize in a few seconds.
I often just dab enamel on it, then slip on shrink leaving the tip
exposed.
greg
George Herold
Hi Rich, Yeah super fluid helium is made with He4. Very cool. I did
research back in grad school with a glass helium dewar. When we
pumped the helium bath to go to lower temperatures I would always
watch the super fluid transistion. Lots of bubbling and then... it
suddenly stops.
Well it turns out that when something goes superconducting the thermal
conductivity drops! The electrons in the superconducting state don't
carry any entropy (heat) and you are left with thermal conductivity
only from those electrons *not* in the super conducting state. (As
well as the phonon contribution.)
George H.
whit3rd
>
> (silly question?)
> Metals are such great thermal conductors are super conductors also
> super-thermal conductors?
Deep question, actually. Superconductors have zero resistivity
because the electrons don't scatter from thermal (phonon)
fluctuations,
thus do not add energy to the heat in the conducting medium.
That decoupling makes the electrons completely ineffective
in moving heat inside the material, and superconductors are
thus TERRIBLE conductors of heat. Most superconducting
wiring, thus, has to be filaments of superconductor in a metal
matrix, or the first quench-while-carrying-current would
melt spots. The superconductor transition has been used
as a thermal 'switch', as well.
Spehro Pefhany
wrote:
AFAIUI, the heat capacity also drops dramatically below the critical
temperature (at in the case of Nb, it apparently increases right at
the critical temperature then drops exponentially towards zero as the
temperature is lowered).
George Herold
wrote:
> On Wed, 19 Jan 2011 12:06:44 -0800 (PST), whit3rd <whit...@gmail.com>
>
> - Show quoted text -
Yup that's what the data show in Kittel "Intro to Solid State Phys"
second edition. (Mind you this is an early version of Kittel and a
bit long in the tooth.)
George H.
whit3rd
<speffS...@interlogDOTyou.knowwhat> wrote:
> AFAIUI, the heat capacity also drops dramatically below the critical
> temperature (at in the case of Nb, it apparently increases right at
> the critical temperature then drops exponentially towards zero as the
> temperature is lowered).
By the third law of thermodynamics, we know that all materials'
heat capacities are zero at zero temperature.
Fred Bartoli
We used high thermal conductivity epoxy: STYCAST 2850 FT
Don't know if it was required or not (didn't do the calc) but we already
had it so it didn't cost anything.
--
Thanks,
Fred.