If you find this site useful, please support it by making a donation of $1 to help maintain and develop it. Click on the PAYPAL DONATE button to do this safely. But there is no obligation - please avail yourself of the information and facilities of the site at no charge.

An Improved Fortin Barometer

The barometer does have some features which make it an improvement on the Fortin Barometer, but it cannot be said strictly to be a Fortin barometer. This is because the Fortin barometer is characterised by having a leather cistern which can be compressed with a screw to adjust the lower level of Mercury. In the instrument described here, a similar adjustment is available to adjust the level of Mercury in the cistern to a fixed point indicated by an index, but this is achieved using a secondary reservoir, connected to the cistern via a rubber hose and which can be moved up and down by means of a rack & pinion. The instrument described here is also different from the Fortin Barometer in not relying on a vernier on the barometer housing, but using an external cathetometer to measure the mercury height.

But there is a further difference between the barometer described here and the Fortin instrument which overcomes the chief weakness of the latter. Even if a Fortin instrument is very carefully constructed and filled, there does remain a doubt over the quality of the Torricellian vacuum above the mercury and with no easy way of checking it. The instrument described here, overcomes this by adding a port to the upper end of the barometer tube which can be evacuated using a rotary pump. The residual pressure can be measured by a Pirani gauge and the barometer reading adjusted to take this into account. The over-all result is that measurements of the atmospheric pressure can be made with much greater confidence than previously and with similar precision.

Design and Construction of the Improved Barometer

The instrument is depicted in the photograph to the left and diagrammatically to the right. The photograph below shows the construction of the cistern and the mercury level adjustment mechanism. Referring to the diagram, the instrument is constructed on a 1100mm length of 50mm x 50mm box-section brass, B. A Class I satin chromed steel rule, S accurate to BS 4372: 1968 (see below) is mounted along the front face of the box-section and secured by 3 M3 screws.The rule was drilled using a carbide tipped 3mm drill. The barometric tube is supplied with a vacuum-tight stopcock T at its upper end, a section of 8mm internal bore tube R and a 25mm ID cistern C and a side-arm in the U bend with a flexible rubber hose connection to the reservoir P, the vertical position of which can be adjusted by the rack and pinion A. An index I is mounted on a brass block and secured by a thumbwheel W.

The index has a brass M3 shaft with a turned Teflon tip which is adjusted so that its tip is at the same vertical position as the lower edge of the rule S.
Clean Mercury is added through the reservoir P until at least a two centimetre depth is achieved in the cistern C. Vacuum is applied to the upper end of the barometric tube via stopcock T which is opened cautiously and partially to allow the mercury to rise slowly in the vertical section of the tube. When it has reached its barometric height, reservoir P is adjusted until the level of Mercury in cistern C just touches the tip of index I. The vacuum is maintained for several hours to remove any residual moisture and the barometric height is measured with a cathetometer. The reading obtained is corrected for the residual pressure in the Torricellian vacuum, by measuring the latter with a Pirani gauge.


It was found that the prototype could be read to 0.1mm Hg. The accuracy of the final reading depends largely on the accuracy of the height measurement. A Class I rule has a quoted accuracy of 0.2 mm across 1 metre length and is certified accurate at 20°C. The cathetometer used in the measurements had a similar specification (0.05mm across 500mm). The correction provided by the Pirani gauge was accurate to about 0.2 mB, which is equivalent to 0.15 mm Hg, so it would not be unreasonable to expect an overall precision of 0.2 mm Hg, or better, if multiple readings are made.

A note on the accuracy of steel rules

The accuracy of steel rules is subject to British Standard 4372:1968 and to DIRECTIVE 2004/22/EC OF THE EUROPEAN PARLIAMENT

These standards specify two criteria: (a) a limit to the departure from the nominal between any two graduations on a single scale and (b) a limit to the departure from the nominal between two adjacent graduations. Under BS 4372 the constraints are as follows:
Up to 300mmBetween 300 & 500mmBetween 500- 1000mm
Distance between any two graduation on a single scale0.10.20.25
Distance between any two adjacent graduation lines0.050.050.05
Position of the 10mm graduation line from its flat end datum0.080.080.08

The criteria specified by the European standard are similar:
The Maximum Permissible Error, positive or negative in mm, between two non-consecutive scale marks is (a + bL), where:
— L is the value of the length rounded up to the next whole metre; and
— a and b are given below.
Accuracy Class
a (mm) b (mm) c (mm)
I 0.1 0.1 0.1
II 0.3 0.2 0.2
III 0.6 0.4 0.3

The Maximum Permissible Error for the length between consecutive scale marks, and the maximum permissible difference between two consecutive intervals, are given below.
MPE or difference in millimetres according to accuracy class
Length i of the intervalIII III
i < 1 mm 0.1 0.2 0.3
1mm< i 0.2 0.4 0.6

The tables above show that the criteria are similar but the British Standard is more demanding in some respects, and less demanding in others, than the European equivalent. The barometer described above was built around a BS4372 1 metre rule manufactured by Stanley (formerly Rabone Chesterman). The photograph shows a comparison between this BS4372 rule (upper image)) and European Class II rule which was three times cheaper(lower image). The photograph shows that the Class II rule exhibits considerable departure from the British Standard rule and seems hardly to conform even to the former lower standard. Both rules (both 1000mm in length) were carefully aligned according to their 1cm graduation and the right hand section of the photograph shows the graduation near the far 1000mm graduation. The 980mm graduation mark appears 0.1 - 0.2 mm away frm that on the British Standard rule, but the difference seems to be substantially made up by the time the 1000mm graduations are reached.