URGENT HELP: How to measure thermal conductivity of gas-solid mixture?

[[email protected]]

Hi ,

I need your help and advice on how to measure thermal conductivity of a
gas (i.e. air, nitrogen) with micron sized particles in it flowing
through a pipe at a certain gas velocity or flow rate. I am quite new
to this and have been looking into possible techniques and theory but I
feel overwhelmed and confused. Hot-wire seemd like a right choice but
I am not so sure now. Looked into Constant Temperature Anemometers
(CTA), having an array of thermocouples etc. or I recently saw
thermistor based set up.

I need urgent help on a suitable technique or techniques,
practicalities, theory etc.

Any help and advice would be greatly appreciated.

Kind Regards,

John

 

[Donald]

Hi ,

I need your help and advice on how to measure thermal conductivity of a
gas (i.e. air, nitrogen) with micron sized particles in it flowing
through a pipe at a certain gas velocity or flow rate. I am quite new
to this and have been looking into possible techniques and theory but I
feel overwhelmed and confused. Hot-wire seemd like a right choice but
I am not so sure now. Looked into Constant Temperature Anemometers
(CTA), having an array of thermocouples etc. or I recently saw
thermistor based set up.

I need urgent help on a suitable technique or techniques,
practicalities, theory etc.

Any help and advice would be greatly appreciated.

Kind Regards,

John

You seem to be discribing "gas chromatography".

Google knows all about it.

http://www.google.com/search?hl=en&c2coff=1&q=gas+chromatography

 

[[email protected]]

I typed "thermal conductivity of gases" into google amd the first hit
has this:

"The thermal conductivity of gases is usually measured in a cylindrical
cell either containing an electrically heated wire coaxially mounted in
a tube of comparatively larger diameter as in the present apparatus or
consisting of two concentric cylinders with the gas occupying the
narrow gap between. With accurate knowledge of the apparatus dimensions
and electrical properties, absolute measurements of thermal
conductivity can be made."

Why don't you like hot wire? Are the particles sticking to it?

 

[[email protected]]

I typed "thermal conductivity of gases" into google amd the first hit
has this:

"The thermal conductivity of gases is usually measured in a cylindrical
cell either containing an electrically heated wire coaxially mounted in
a tube of comparatively larger diameter as in the present apparatus or
consisting of two concentric cylinders with the gas occupying the
narrow gap between. With accurate knowledge of the apparatus dimensions
and electrical properties, absolute measurements of thermal
conductivity can be made."

Why don't you like hot wire? Are the particles sticking to it?

The "hot wire" wouldn't have to get that hot - if it were platinum
wire, you could use it both as the heating element and the temperature
sensor, though the application pretty much demands that you mount it
under enough strain to keep it straight, making it les than perfect as
a temperature sensor. You would at least know the resistance of the
wire very precisely, and if you monitored the current through it with
equal precision you'd have a pretty exact idea of the heat beng
dissipated in the wire.

You'd want to be able to evacuate the tube, so you could measure heat
loss by radiation and conduction through the supports as a function of
wire temperature,, and subtract this from the heat loss neasured in the
presence of flowing gas.

 

[Phil Hobbs]

You seem to be discribing "gas chromatography".

Google knows all about it.

http://www.google.com/search?hl=en&c2coff=1&q=gas+chromatography

Thermal conductivity? You sound like you're measuring flow rate. What
are you trying to do?

Cheers,

Phil Hobbs

 

[Tony]

There are some questions associated with your problem. Gross thermal
characteristics of a flowing gas are a lot different than those of
stationary phase. "Thermal conductivity" would be a constant 'K" such
that

heat flow = K * delta temp * area / thickness.

Is that what you're trying to measure? Flowing gas phase has vastly
different boundary conditions than stationary phase gas because the
energy transport modes are different: moving gas carries some of the
energy away, it isn't conducted away.


If you describe exactly what you're trying to measure you'll likely get
good advice.

 

[[email protected]]

There are some questions associated with your problem. Gross thermal
characteristics of a flowing gas are a lot different than those of
stationary phase. "Thermal conductivity" would be a constant 'K" such
that

heat flow = K * delta temp * area / thickness.

Is that what you're trying to measure? Flowing gas phase has vastly
different boundary conditions than stationary phase gas because the
energy transport modes are different: moving gas carries some of the
energy away, it isn't conducted away.

The interesting distinction is between turbulent and laminar flow -
Reynolds number less than 2300 in a circular pipe gives laminar flow,
and higher than 2300 gives turbulent flow.

In the laminar flow region, the velocity profile in the pipe is
parabolic, peaking a the centre of the pipe and falling off steadily as
you get closer to the walls.You don't get any transverse fluid flow,

In the turbulent flow region the velocity profile is smooth and
regular, and laminar close to the walls - with the Prandtl bounndary
layer - and irregular over the centre of the pipe. You do get
transverse flows outside the boundary layer, giving more or less
homgenous "plug" flow at the centre of the pipe.

 

[[email protected]]

Hi Joe, Bill, Tony and Phil,

Thanks for all the useful comments. I think that heat transfer will be
dominated by convectivity and I do not know how to extract thermal
conductivity from it. All I am trying to do is to see whether
existence of these solid particles have any effect on the thermal
conductivity or may be heat transfer properties of this gas. I am
trying to determine this as well.
I got hold of a CTA system today as someone assured me that it would do
what I need i.e. thermal conductivity but I now realise that it
measures the velocity. I got hold of a book on hot-wire anemometry but
how one would extract thermal conductivity from those equations set, I
have no idea at present as there are so many parameters.

What is the best describing factor in the thermal properties of this
system? What property should I determine so that I could say "Look,
existence of these particles made this difference to so and so thermal
properties of this gas" and what is the easiest experimental method to
do it.

Thanks again for all your comments.
Kind Regards,
J

 

[[email protected]]

Hi ,

I need your help and advice on how to measure thermal conductivity of a
gas (i.e. air, nitrogen) with micron sized particles in it flowing
through a pipe at a certain gas velocity or flow rate. I am quite new
to this and have been looking into possible techniques and theory but I
feel overwhelmed and confused. Hot-wire seemd like a right choice but
I am not so sure now. Looked into Constant Temperature Anemometers
(CTA), having an array of thermocouples etc. or I recently saw
thermistor based set up.

I need urgent help on a suitable technique or techniques,
practicalities, theory etc.

Any help and advice would be greatly appreciated.

Kind Regards,

John

 

[[email protected]]

Hi John,

Our main business is development, production and sales of flow sensors,
for liquids and gases, based on the calorimetric principle. We are in
this business since more than 20 years.
With this technology the thermal conductivity of the medium is the main
factor for monitoring asnd measuring flow rates.

We are very interested in your project and we could see that we could
be of help for you in this matter.

Please let us have more details, specially more technical spec`s, to
become able to look after for a possible solution.

Best regards
Karsten Wippich
weber Sensors Ltd.

 

[[email protected]]

Hi Joe, Bill, Tony and Phil,

Thanks for all the useful comments. I think that heat transfer will be
dominated by convectivity and I do not know how to extract thermal
conductivity from it. All I am trying to do is to see whether
existence of these solid particles have any effect on the thermal
conductivity or may be heat transfer properties of this gas. I am
trying to determine this as well.
I got hold of a CTA system today as someone assured me that it would do
what I need i.e. thermal conductivity but I now realise that it
measures the velocity. I got hold of a book on hot-wire anemometry but
how one would extract thermal conductivity from those equations set, I
have no idea at present as there are so many parameters.

This is actually a pretty complicated problem, and depending on what
the particle is, it could drastically effect thermal conductivity. Do
you absolutely need k, or are you looking for some general measure of
heat transfer ability? Will the overall heat transfer coefficient be
good enough? To keep the particles suspended (how small are they?),
you will have to introduce convection in the form of flow to prevent
them from settling. This introduces coupled fluxes, which can truly
make things nasty if you don't know a lot about your system already.
And you also get into the mess mentioned by the previous poster about
laminar vs. turbulent flow, you'd probably witness a drastic increase
in heat transfer transitioning from laminar to turbulent, which may
allow you to decouple heat convection from conduction? This is why you
*always* run a heat exchanger in turbulent regime - convection
dominates but you also don't want to introduce a series resistance with
a low-conductivity material for the transfer surface walls.

Without thinking about this more unnecessarily, why don't you do a
literature search. This is exactly the sort of thing that fluid
dynamicists spend their entire lives making models for and correlating
back to real measurements. It's a shame to spend so much time
developing a model and then not use it when a real-world problem comes
along (this seems to be what usually happens). Heat transfer
experiments in "fluidized bed reactors" will be directly applicable to
what you're doing, a solid in gas analysis will be fundamentally the
same as solid in liquid if not quite a lot easier (inviscid), just with
different parameters - they're all fluids.

If all you need is a general measure of heat transfer, say comparing
the gas with and without added particles, the overall heat transfer
coefficient can easily be measured by designing a crude heat exchanger
with insulation and thermocouples. It could be measured for a wide
range of flow rates.

 

[tadchem]

Google "thermal conductivity gauge"

I have done a LOT of gas chromatography over a period of many years.

TC gauges are very simple, useful and robust.

They consist essentially of a wire inserted into a steadily flowing
stream of gas. A small DC current is allowed to pass through the wire
and the resistance is measured with a bridge circuit.

As the current flows, the wire heats up and the resistance changes. A
small wire will quickly equilibrate temperature as the gas flow carries
away heat by convection. The resistance of the wire will vary with
temperature. The equilibrium temperature will vary with the thermal
conductivity of the gas (and its contents). The higher the thermal
conductivity of the mixture, the lower the temperature (and resistance)
of the wire compared to a pure gas at the same pressure and flow rate.

We used it to measure composition of gas mixtures. The greater the
difference in thermal conductivity between the gas and its contenst the
more sensitive the detection is. We only had trouble measuring ppm
levels of hydrogen in helium. Air was a breeze ;-)

Tom Davidson
Richmond, VA