[wxqc] Fortin Barometer

Victor Engel brillig at gmail.com
Tue Dec 23 16:32:49 CST 2008

There is a third way:

The scale could be on a material that floats on the surface of the mercury
in the cistern. As the mercury level rises and falls, so does the scale.

I think our barometer in grade school used a 1/2 inch tube (same diameter as
normal test tubes) and a normal sized beaker (size similar to a Coke can).
This was in Guatemala, by the way.


On 12/23/08, A.J deLange <ajdel at cox.net> wrote:
> If one mounts a cup to a support, fills the cup with mercury, fills a
> scrupulously clean tube a bit longer than 800 mm with triple distilled
> virgin mercury, seals the mouth, inverts the tube, places the end beneath
> the surface of the mercury and then removes the seal he has just made a
> mercurial barometer (Kew design). The height of he mercury in the tube will
> be determined by the atmospheric pressure acting on the surface of the
> mercury, the "capillarity" of the tube and the pressure of the mercury vapor
> at the top of the tube. If one fixes the tube and a length scale to the
> support with the scale marked in inches from the surface of the mercury in
> the "cistern" he will have correctly calibrated the barometer as described
> in my last post and this should be fine as long as the tube is more than
> 1/2" in diameter in which case the capillarity error is small enough to be
> disregarded. But a 760 mm column of mercury 1/2 inch in diameter contains 96
> mL of mercury which, with its density of 13.5 g mL amounts to 1.3 kg (over
> three pounds if you include the mercury in the cistern). Over a pound is
> "reportable" and mercury is very expensive so not many barometers are made
> with half inch tubes. For smaller tubes capillarity correction must be
> included and such barometers are calibrated by comparison with standard
> barometers whose calibrations are traceable to NIST.
> Assuming the calibration has been done at standard pressure (i.e. the scale
> is properly marked at 760 mmHg) consider what happens when the pressure
> rises. The additional pressure pushes mercury up into the tube which means
> the level of the mercury in the cistern goes down and the "offset"
> determined from the calibration at 760 no longer applies. The calibration by
> comparison has to be repeated at 780 mm (and several other pressures)
> because the 760 mm mark is not simply 20 mm higher than the mark at 760 mm.
> The scale is not linear. There are 2 ways to solve this problem. One is to
> make the diameter of the cistern very large compared to the tube. This means
> more mercury, more $ and more regulatory hassle. The other way is to adjust
> the height of the mercury in the cistern to the same level (fiducial point)
> for every measurement. This is the function of the "bellows" which is really
> more like a bag than a bellows in the Fortin design. The cistern's floor is
> a piece of leather into the center of which a screw protrudes. Before each
> reading one taps the cistern (to deal with any surface tension issues in the
> cistern) and then adjusts the screw, thus raising or lowering the floor of
> the cistern and the surface of the mercury, until the tip of a pointer
> affixed to the bottom of the brass tube which carries the scale just touches
> the tip of its image in the mercury. There is a decent diagram at
> http://www.npl.co.uk/server.php?show=ConWebDoc.1231. One then taps the top
> of the tube and reads the height from the scale which is fixed relative to
> the surface of the mercury for each reading. The 780 mm mark can now be 20
> mm above the 760 mm mark (the scale is linear) and it is only necessary to
> compare to a NIST traceable instrument at one pressure.
> All this, and lots more, is, presumably, explained in the Manual of
> Barometry published by the USG in 1963 and sold by the Superintendent of
> Documents and then NOAA but I can't find a copy anywhere. Does any one know
> where these can be had?
> A.J.
> _______________________________________________
> wxqc mailing list
> Post messages to wxqc at lists.gladstonefamily.net
> To unsubcribe or change delivery options, please go to:
> http://server.gladstonefamily.net/mailman/listinfo/wxqc
> To search the archives:
> http://www.google.com/coop/cse?cx=008314629403309390388%3Aknlfnptih9u
> The contents of this message are the responsibility of the author.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://server.gladstonefamily.net/pipermail/wxqc/attachments/20081223/c9917016/attachment.html>

More information about the wxqc mailing list