[wxqc] wxqc Digest, Vol 50, Issue 19

Dr. Dana-Renee Lee PhD drlee6530 at drlee.com
Thu Dec 25 04:34:58 CST 2008


The Bellows at the bottom of the unit seals the bottom chamber and has a
screw in the middle to allow

adjustment of the column to calibrate the mercury column to the proper high.

 

-----Original Message-----
From: wxqc-bounces at lists.gladstonefamily.net
[mailto:wxqc-bounces at lists.gladstonefamily.net] On Behalf Of Victor Engel
Sent: Tuesday, December 23, 2008 8:57 AM
To: Discussion of weather data quality issues
Subject: Re: [wxqc] wxqc Digest, Vol 50, Issue 19

 

This is new to me. Can you go into more detail about the function of the
bellows?

Victor

On 12/23/08, Dr. Dana-Renee Lee PhD <drlee6530 at drlee.com> wrote:

It is a little more work to cal a mercury unit but far easer than an
electronic sensor such as a strain gauge.

A mercury unit is a long tube closed on one end and filled with mercury.
The open end is in a pool of mercury and has a bellows at the bottom to set
the reading to the correct value on the built in measurement fixture.  You
must first before you can set it is to have a known accurate reading and
calculate your reading from the difference in height above the accurate
reading to get your reading.  Remember, barometers are set to sea level
pressure.  If you are at sea level just find another at sea level and
duplicate the reading but if you are say 5000 feet above the sea then you
have to calculate the difference in reading to set yours.



Dr. Renee Lee PhD



-----Original Message-----
From: wxqc-bounces at lists.gladstonefamily.net
[mailto:wxqc-bounces at lists.gladstonefamily.net] On Behalf Of A.J deLange
Sent: Sunday, December 21, 2008 9:49 AM
To: wxqc at lists.gladstonefamily.net
Subject: Re: [wxqc] wxqc Digest, Vol 50, Issue 19


RE:

> But one could argue that the fact you really can't precisely
> calibrate a mercury barometer makes it inherently less useful.

A mercury barometer is actually one of the simplest instruments to
calibrate because what you are calibrating is just length. All you
need is a good "ruler". Actually, I imagine (I have never worked in or
even been in a barometer factory) some sort of brass jig is used. In
this way the calibration (of the barometer)  could be done without
regard for temperature as the brass jig would expand as much as the
barometer body (also of brass) does . Obviously this jig would have to
be compared to some standard of length traceable to NIST and that
would have to be done with regard to temperature depending on the
primary standard (osmiridium and brass would have different
coefficients of expansion).  Supposing that the jig is 30" long at 62F
it is then a matter of adjusting and fixing the scale on the body of
the instrument in the position that places the 30.00 inch mark at the
end of the jig when the tip of the mercury surface pointer is at the
other (irrespective of temperature). The metric scale would require
another jig calibrated against a length standard at 0C but again it
could be used at any temperature. More probably a single jig would be
used  for either a metric or inch level and the corresponding scale
marking in the other system specified. Note that all this can be done
before the tube is assembled to the cistern or the device filled with
mercury.  Once the scales are fixed at the proper distance from the
pointer the unit is "calibrated". This does not mean that an
additional check of the system may not be done in a controlled
pressure chamber. This would require a pressure standard (as opposed
to just the length standard) which would have to be tied to an
acceptable (acceptable would be defined by the weather service, the
military or whoever has ordered the calibration). Any difference
between the reading and the pressure chamber test can then be recorded
as a "certificate correction" (that's what Princo called them) and
added to the other corrections when the barometer is read.

An instrument calibrated as above is ready to be used but its readings
must be corrected for temperature (the mercury and brass both expand,
but by different amounts, for each degree increase in temperature) and
the local gravity field if all that is wanted is local pressure (QFE)
to, for example, calibrate a laboratory instrument such as the digital
density meters used by breweries and other (less important)
industries. Temperature is measured by a thermometer in intimate
contact with the barometer's brass tube. An adequate gravity
correction comes from knowledge of latitude. On high plains additional
opposing corrections for increased distance from the center of the
earth and for the extra mass between the barometer and the center of
the earth are applied. The sum of these  is quite small relative to
the latitude correction.

For aviation sea level pressure in the ICAO standard atmosphere (QNH)
is wanted. This is obtained from a simple formula applied to the
corrected station pressure and this  requires the station elevation to
be known quite precisely. Each foot change in elevation results in a
change in QNH of about 0.001" (depends on altitude) so it is very
important to know what your station elevation is. This is, in my
opinion, is the main difficulty with mercury barometers as the
altitude is not easy to determine. At my station 5800 GPS readings
gave a mean elevation of 291 feet  but the SEM (Standard Error in the
Mean) was 27.8 feet (in other words the GPS readings were all over the
place). A month later 11830 readings gave a mean of 280 ft with an SEM
of 6.2 feet (these are consistent results). Lately I have taken the
approach of finding the elevation value which minimizes the weighted
(by the reciprocal of the distances) rmse (root mean square error)
between my barometer's QNH and KDCA's (9.3 miles) and between my
barometer's and KIAD's (16.5 mi) [KIAD and KDCA differ by an average
of 0.4 mb (DCA higher) with SD of 0.5 mb.] Using this approach my
current estimate of my station's elevation is 276.6 ft. I've got my
Davis pressure sensor tweaked to agree with the mercury barometer to
an average difference of .08mb (SD 0.32 mb). Average difference with
KDCA is 0.04mb SD (.22 mb) and with KIAD is 0.32 mb (SD 0.40 mb). The
mercury vs KDCA difference is, on average 0.13 mb (SD 0.4 mb) and with
KIAD is 0.23 mb (SD 0.38 mb). Note that rms quantizing error (rounding
reported QNH to the nearest 0.01") amounts to an SD of  about 0.14 mb
so I consider myself pretty well calibrated and don't take the QC
reported errors too seriously. But I am still "surveying in" my
station with respect to its elevation.

Another big question is as to whether the airports are reporting QNH
(altimeter setting) which assumes the atmosphere is at 15 C at sea
level with the normal (0.0065 C/m) lapse rate up to the station  or
QFF (mean sea level pressure) which assumes it is at the average
station temperature all the way from sea level to the station. The
difference between QNH and QFF. There is little difference at modest
elevation when the average temp is near 59 but for my elevation
(84.3m) when it gets down to about 30 F the difference is more like
0.6mb (0.017 in Hg). If I pull up the NWS page for an airport it says
"Pressure (altimeter)" which leads one to believe that it is indeed
QNH that is being reported. However, in the RMK field of the METAR one
finds the SLP (QFF) field filled with a value which matches the QNH
value (e.g. "Pressure (altimeter)"  reported as 30.07 inches
corresponding to 1018.3 mb and "SLP 183"). Problem is that when it is
this cold (average temp as defined by  the Federal manual was 37F for
this particular reading) SLP should be more like 1018.8. So now I'm
confused.


Further to the discussion of aneroids: It occurred to me that an
altimeter is an example of an accurate aneroid device. As I recall the
calibration requirement for them was +/- 100 ft corresponding to a
pressure accuracy of +/- 0.11 in Hg. As I recall this was achieved
using a portable box which contained a vacuum pump and another
altimeter which has to have been calibrated against a precision
pressure standard in a lab. For IFR flight the aircraft's pitot static
system had to be checked every other year (again, if I recall
correctly). To use an altimeter as a barometer one turned the
adjustment knob until the hands read the field elevation and then read
the setting in the Kollsman window (for in-flight use the pilot sets
the reported QNH in the Kollsman window). At small airports this was
the source of the "altimeter setting" (QNH) radioed to approaching
pilots but it was not "official" and not to be used for IFR.

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.

_______________________________________________
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/20081225/23b6bde3/attachment.html>


More information about the wxqc mailing list