Ultrafast Temperature Measurement

[Gerd Roethig]

Hello all,


for medical research purposes we are desperately looking for a good
method to measure temperature very quickly.
The most important thing is the time constant of the measuring system;
it should be as low as possible.
Microbead thermistors appear to be one possible solution but they have
some problems, mostly mechanical ones.

We want to measure temperature in breath air, for example in
ventilated patients. So there are flow and water vapour that will be
dangerous for microbead sensors.
Encapsulated ones are not suitable since their time constant is too
high due to the encapsulation.

So there are two questions:

Is there an alternative to microbead thermistors for an ultrafast
temperature measurement?

How can microbeads be protected against mechanical and chemical
influences without prolonging the time constant by some encapsulation
or things like that?

Any hint would be greatly appreciated.

Thank you very much for your efforts in advance and best regards

Gerd

 

[Gerd Roethig]

Hello,

Can you define "too high"?

The time constant should be as low as possible. There are
Thermometrics microbead sensors that are supposed to have a time
constant of 10 ms in still air. Seems that this is the lowest possible
value.
They achieve this time constant by using a very low amount of material
for the thermistor itself which makes that microbeads somewhat
sensitive to environmental influences.

I suspect you'r not actually wanting to measure the temperature of
the patients exhalation, but some second order effect.

What is this?

You're partially right - we really want to measure temperature changes
caused by certain factors. But the system must react very fast to
these changes - that's the problem.

Cheers,

Gerd

 

[Zak]

You're partially right - we really want to measure temperature changes
caused by certain factors. But the system must react very fast to
these changes - that's the problem.

Wouldn't a thin platina wire be useful? Corrosion resistant and easy to
warm up if there is any appreciable flow.

It is big, but you have the mass to be measured - and it will react
quickly. How thin platina wire can one get? How wind resistant is it?


Thomas

 

[Tony Williams]

We want to measure temperature in breath air, for example in
ventilated patients.

[snip]

In one of the Omega temperature measuring books there is
a reference to: "The Measurement of Respiratory Air
Temperature", by P Webb. Rev Sci Instrumentation, Vol 23,
pp 232-234, May 1952.

He appears to have constructed a fine copper-constantan
micro-thermocouple for use in nasal passages. Claimed
results were 0.1C resolution, with the response time
being 50% down at 200 milliseconds.

The Omega book is "Temperature Measurement in Engineering",
Vol 2. Might be worth obtaining, because it describes
various resistive, thermoelectric, and velocity of sound
techniques that have claimed responses down to 1uS.

 

[Ian Stirling]

Hello,



The time constant should be as low as possible. There are
Thermometrics microbead sensors that are supposed to have a time
constant of 10 ms in still air. Seems that this is the lowest possible
value.

They achieve this time constant by using a very low amount of material
for the thermistor itself which makes that microbeads somewhat
sensitive to environmental influences.



You're partially right - we really want to measure temperature changes
caused by certain factors. But the system must react very fast to
these changes - that's the problem.

I suspect it could be done with IR spectroscopy to work out the water/CO2
content, then measuring the speed of sound.
You would need also to measure the Oxygen in the incoming air.
Measuring Argon and Nitrogen would be nice, but you could probably skip
it, and go with just pressure.

This could have response times in the microseconds, but may not be
very accurate.

The patient can't change temperature in 10ms, or even 1s significantly,
unless they are on fire.

Are you trying to measure peak exhalation temperature, or pick up
faults in the equipment?

Very fine platinum wires can be obtained, and you can accurately
measure the resistance of these.
This is somewhat complicated as you can't have significant current
flowing through them due to self-heating causing large errors.

Condensation on the sensing element may also be a problem, depending
on where it is.

--
http://inquisitor.i.am/ | mailto:[email protected] | Ian Stirling.
---------------------------+-------------------------+--------------------------
"The device every conquerer, yes, every altruistic liberator should be required
to wear on his shield... is a little girl and her kitten, at ground zero"
- Sir Dominic Flandry in Poul Andersons 'A Knight of Ghosts and Shadows'

 

[Zak]

Very fine platinum wires can be obtained, and you can accurately
measure the resistance of these.
This is somewhat complicated as you can't have significant current
flowing through them due to self-heating causing large errors.

You could use two, and extrapolate the temperature of a currentless one
from one with I and the other with 2I current.

Condensation on the sensing element may also be a problem, depending
on where it is.

Condensation... hmm. If the wire heats up a bit as above...


Thomas