I was interested to see Johan Liljancrants' article on leak measure-
ment. In spite of the large numbers of people doing pneumatic restor-
ation I am afraid there is there remarkably little interest in testing
and in leak measurement in particular. It however is nice to find
another soul who appears to understand the value of quantitative
measurement in pneumatic restoration.
I don't wish to be too critical but I am not sure that the concept
of using Flow Resistance is the most suitable one to use in this appli-
cation. It would only be accurate for pure laminar flow, i.e., where
the flow is directly proportional to the pressure differential. Even
in a leather pouch this is not the case since the leather stretches and
the porosity changes as the pressure across the pouch changes. You
would also need a different formula for turbulent flow, i.e., flow
through an orifice.
I have always felt that the best way of reporting leakage measure-
ments is as XX cubic inches per minute measured at XX inches of water
column vacuum. This is the method which I settled on in my leak test
apparatus described in 970205 MMDigest. [ See the article "Airflow
Measurement Equipment" in the MMD Archives]. This reporting method
is, I feel, more intuitive than the resistance concept and thus may
be better understood by many restorers. (This is not unimportant, for
there are many restorers out there who do not seem to even understand
the basic difference between pressure, pressure differential and flow.)
I know that Craig Smith has also built a leak test unit using the
orifice flow formulae which I have provided him. Although his design
is completely different to the one which I built he also has decided
to report leakage measurement results as cubic inches per minute at
some specific water column vacuum. (Perhaps Craig will finally get
around to the article he was planning for MMD.)
I would certainly welcome further discussion on the subject.
[ Editor's comments:
[ Let me begin the discussion by listing the uses of leak meters:
[ 1. Relative improvement of sealing. Any repeatable method
[ should work.
[ 2. Absolute calibration of any restriction. Since this is
[ most likely to be non-linear, only a chart of flow vs. pressure
[ differential will show its characteristics.
[ 3. Relative calibration of a restriction, e.g., comparing an
[ unknown bleed with a known (calibrated) flat-plate aperture bleed.
[ If you are rebuilding an organ pipe chest you will be concerned
[ with locating and reducing leaks, and maybe you will want to
[ assess the relative porosity of valve pouches. A player piano
[ action might receive similar attention.
[ If you are designing pneumatic valves, especially valves for
[ a reproducing piano which operates from 5 to 50 or more inches
[ water column, you will want a different method to characterize
[ airflow: a method which properly treats the non-linear flow of
[ air through an aperture or a port.
[ -- Robbie