Re: Viscosity demo.?

[brian whatcott <inet@INTELLISYS.NET>]

At 00:43 4/1/00 -0800, Bernard G. Cleyet & Nancy Ann Seese wrote:
> People!
>
> A few years ago some one at UCSC showed me a demo that was
> supposed to demonstrate the
> increased (yes) viscosity of a gas with temperature.
> It consists of two copper tubing
> spirals connected to a "T".  This way natural gas or other may
> be supplied at equal
> pressure.  The flames at the other ends indicates the volume of
> gas flowing through the
> tubes, initially equal.  If one heats one of the spirals its
> flame height , counter
> intuitively, rather noticeably decreases, while the other
> is unchanged.  Conversely,
> cooling one with liquid nitrogen causes its to more than double.
> All well and good until
> one realizes that  heating a gas will reduce it's density
> and therefore,  also cause
> decreased flame height, at least to a first simple minded
> approximation.  What is it?  Any
> of you heard of this demo., and what does it "really" demonstrate?
>
> Panzers, in  advance, I pray.
>
> bc

This is indeed an amusing puzzle - or rather two puzzles.
(What does "Armor-plates in advance please" mean?)

The first misdirection is the assertion that flame height
indicates gas volume combusted.  This is possibly
oversimplified. One could propose supply pressure, calorific value
and speed at the nozzle as possible contributors to flame height.
    I fancy that temperature difference between flame and ambient
could also play a part.

The second misdirection is the question, 'What does the effect
"really" demonstrate?'.
   A cynic pigeon-holes this stance as physicist-talk; the supposition
that one probably knows better than the other fellow - even where
 a physics department is in question.
 Hehe...let us suppose that those folks really knew what
they were doing, when they came up with this demo.

   They used a differential effect - always a useful way to exclude
extraneous factors - so we can reasonably assume a common gas pressure
at the tee point.
 The viscosity of gases (as commonly offered in databooks like the Rubber
Company's) almost invariably increases with temperature (yes).

 A feature of the 'ragbag' style of data acquisition one sometimes
finds in databooks like the one mentioned, places datapoints from
 different sources in counterpoint.
Case in point; if one looks hard enough, one finds this:

methyl alcohol

degC       visc uP
 66.8      135.0    ref 1
111.3      125.9    ref 2
217.5      162.0        2
311.5      192.1        2

Ref 1 Steudel 1882
Ref 2 Titani  1933

To reassure oneself as to the properties of natural
gas I refer to propane as a reasonable surrogate:

 17.9       79.5   Ref 3
100.4      100.9       3
199.3      125.1       3

Ref 3  Trautz, Sorg 1931

All in all, one can reasonably conclude that the gas
directed to the burners exhibits increased viscosity
with temperature.

One next inquires: what happens to the heated gas?
Q1) Does its pressure increase?
Q2) Does its volume increase?
Q3) Does its flow speed increase?
Q4) Does its (bulk) kinetic energy decrease?

One concludes that the answer to Q1) is no - the control flame would
otherwise be affected.
Q2) must inevitably call for a Yes answer
and thereby Q3) = yes
Finally, with higher flow speed and higher viscosity one
reasonably answers Q4) as yes - with another heating contribution
from viscous losses.

I tentatively conclude that when this 'hot' gas cools to its previous
temperature, the effect is similar to reducing the bore of
one side of the tee (with no excess heating) less gas issues forth.

Viscosity modulation has changed the heating value of gas delivered.

Brian



brian whatcott <inet@intellisys.net>
Altus OK