On Dec 3, 1:24*pm, hamilton wrote:
On 12/3/2010 10:59 AM, George Herold wrote:



On Dec 3, 10:08 am, Glen *wrote:
On Wed, 01 Dec 2010 20:21:21 -0500, Phil Hobbs wrote:
Glen Walpert wrote:
On Tue, 30 Nov 2010 14:15:28 -0500, Rich Webb wrote:
Thermocouples are really really horrible temperature sensors--almost
as bad as ICs. *Their advantages are small size, relatively low cost,
and (potentially) high speed, but their disadvantage is that it's
really hard to get good measurements.

One reason for this is thermal conduction down the leads (which is a
big problem for most temperature sensors). *There's a worse one,
though: due to their very low sensitivity, thermocouples are
extremely vulnerable to errors caused by offset drift in the
circuitry. *See Figure 20.3 on P. 803 at
http://electrooptical.net/www/book/draftthermal.pdf

"Thermocouples generate a voltage related to the temperature
difference between two junctions of dissimilar metal wires, and are a
pain in the neck. ... For less specialized applications, avoid
thermocouples like fleas."

Okay, now I'm going to *have* to buy that book!

Yes, that is a good one, and I can't argue with Phil's assessment of TC
accuracy, but "You can learn all you ever wanted to know about them
from the Omega Engineering catalogue." is true only if you have no
interest in how they actually work (as opposed to how to use them) or
how to make a really good low drift TC measurement system with accurate
CJC and high immunity to EMI, for instance.

There are many trade-offs in sensor selection, and thermocouples excel
in temperature range and durability as well as (sometimes) speed. *A TC
can be smashed flat with a hammer and suffer no loss in accuracy until
the wires break, or welded to metal parts for excellent thermal contact
with the part being measured, for instance. *Try that with a thermistor
:-).

You can solder RTDs down, which is about the same thing. *And the
circuit details aren't really information about the _sensor_--any
sufficiently poor sensor will have those problems.

I've used thermocouples reasonably often, generally running inside an
evaporator or someplace like that, and attached to a Fluke thermocouple
thermometer with built-in cold junction compensation. *That was probably
good to a couple of degrees, which was all I really needed, especially
since I didn't have to replace the TCs, so the measurements correlated
pretty well over time.

*From the perspective of your book - temperature stabilization of electro-
optical components - I agree completely that thermocouples should
generally be avoided like fleas. *From my perspective, with a background
in power plant control, thermocouples and RTD's are the rule and
thermistors the rare exception.

What really irks me about Omega is that their "Thermocouple Introduction
and Theory" section is actually a "thermocouple introduction and
completely bogus theory". *Correct theory can be found at:
http://www.electronics-cooling.com/R...rticles/JAN97/
jan97_01.htm

And better yet is another article which I can no longer find on the web,
which I will post on ABSE with thread title "thermocouple theory article"
in case anyone is interested.

Glen- Hide quoted text -

- Show quoted text -

Hmm, that link sends me to "Microthermal imaging in the infrared"
Nothing about thermal couples?

George H.

Did you fix the wrap ??- Hide quoted text -

- Show quoted text -

That didn't help, but I searched the site for "thermocouple" and
found it near the end of the list.

George H.