Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: A weighty subject - a reply to the fixed weighters...



Bravo, Leigh, Michael and Brian your clear and insightful comments and
discussions this evening have provided generous food for thought.

Cliff Parker

brian whatcott wrote:

At 17:49 10/14/99 -0400, Michael Edmiston wrote:
...
Brian, what does a thermocouple measure? Won't virtually everyone say
it measures temperature? Are you going to insist that it measures the
Seebeck Voltage?

If by thermocouple, you mean the junction of two dissimilar metals, as
I expect you do, then I need to say no temperature related voltage is
to be found there.
The electrical effect which acts as a temperature proxy acts along the
length of the wires.

What does a mercury thermometer measure? Won't virtually everyone say
it measures temperature? Are you going to say it measures the
expansion/contraction of mercury?

I most certainly will *not* say that a capillary mercury thermometer
measures the thermally driven volumetric change in mercury with
temperature. That is quite wrong. Volumetric temperature methods
measure the difference in volume of two materials with temperature:
in this case, of glass and mercury.

What does an analog voltmeter measure? Won't virtually everyone say it
measures potential difference? Are you going to say that it actually
measures current? Or maybe you're going to say it measures magnetic
force?

There are two common forms of analog voltmeter:
the D'arsonval movement uses current as a proxy for volts:
the moving vane electrostatic voltmeter uses electrostatic force
against a spring as a proxy for voltage and of course the usual FET
input digital device is electrostatic field sensing in nature.

I can sit here and think of these kinds of examples all afternoon.
Almost everything we measure is measured by way of some change that is
easily observed, but this observed change is in something proportional
to the actual thing being measured. We don't measure temperature
directly, we measure the effect of temperature on something else. But
people don't get uptight about saying we're measuring temperature when
we're really measuring the volume and pressure of a gas thermometer.
So why are you getting picky about mass measurement and holding it to
a different standard than virtually everything else we do?

If you don't keep the actual modality of the indicator in mind, then
you are apt to chose the wrong instrument for a particular circumstance:
the equal arm balance for mass determinations in deep space was my
example.
I could sit here all evening and think of more germane examples of
misapplied instruments. (Or so I can claim as a goading riposte :-)

Balances are designed to measure mass just like voltmeters are designed
to measure potential difference. Saying that balances measure mass is
no more wrong than saying voltmeters measure potential difference.

I agree that it is no more wrong. But why be wrong at all?

Unless the manufacturer errs, or the user errs, an instrument measures
what it is designed to measure.

Of course this is the case, if the user will only stay within operational
boundaries. (But it is a great nuisance to purchase a kilovoltmeter
at auction, to find on delivery, a 50 micramp movement.
It is unwise to place too much face in the ostensibly measured quantity.)

The reason I started this whole
discussion was to point out that the user might be erring if he thinks
an "electronic scale" is measuring force or weight, because it might
actually be measuring mass. And I intend this last sentence with the
same semantics as I am using in this and the previous several
paragraphs.

Although the various scales (or balances, if you will) in my possession
all promote the fact they measure mass, *I* at least know what they use
as a proxy, so I will do as NASA does when I stray outside the bounds
within which they do not lie: I will choose an alternative force to
measure which persists in these 'forbidden' regions.

Brian says that "balance" need not apply to equal-arm balances. I say
it depends upon who you ask. The "Academic American Encyclopedia" (one
that I have found extremely reliable) says about the word balance: "The
term is properly applied to an instrument that opposes equal weights in
two pans suspended from the ends of a lever that has its fulcrum
precisely in the middle." But the "McGraw Hill Encyclopedia of Science
and Technology" (also a reliable source) says that the word balance
historically applied to equal arm balances, but "In contemporary usage,
the word 'balance' may be used, but the method is quite different from
the classical two-pan design." It also says, "There are many types of
balances: spring type scales, single-lever deflection types, precise
two-lever-arm designs, torsion designs, substitution designs, and
complete electronic weighing systems."

Perhaps if you take a wider view of the word "to balance" it will be
easier to see that any scale in fact balances forces, so it may
reasonably, without any strain be called a balance in that
sense.
Whereas, in the broader view, any balance without a scale does not
really count as a scale. The equal arm balance was not originally
a scale: the electronic balance invariably is - a scale, in this sense.

By the way, all the language in the McGraw hill source uses language
that says we are measuring mass.

I don't wish to be churlish: I keep a set of McGraw Hill Enc. of
Science & Technology at home after all, but what it gains in breadth,
it somewhat loses in depth as compared for example with the single
volume Enc. of Physics: Besancon: Van Nostrand Reinhold.
I expect you would agree. The Mass & Inertia article is a modest
and clear exposition, in my view.

At one time I would have said that balances measure mass and scales
measure force. But I am quite sure that I cannot make that distinction
any more. The electronic age, and public usage of the words, have
completely muddied this distinction. That's why I asked, in a recent
post, if we need to invent some new names.

If you would accept the understanding of the words 'balance' and 'scale'
which is familiar to students even before they have received science
instruction, you would not find it inimicable to the most advanced usage,
I fancy.

* * * *
Brian says: "... a
quite popular current design (Ohaus) has the load pan balanced
coaxially by a magnetic balance coil. (or is he also agreeing with this
concept? - it is not clear from his text.)"

I have actually designed and built electronic balances. And we have an
Ohaus "balance" using the design Brian mentions. In our labs I just
counted 15 electronic "balances" that measure mass in one way or
another. I believe these use roughly 5 different methods of operation
ranging from strain gauges on a load cell to electronically nulled
balance beams, and including the coaxial design mentioned by Brian. I
didn't think it was necessary to describe all possible designs, and I
specifically stated that "...the exact mechanism is not so
important..."

I hold out that there is one detail of a scale's design which is
important: is it capable of working in a gravity free or a free-fall
environment?
If you had even *one* of these, I dare say you would know - their
action is distinctive and recognizable.

* * * *
When I said that the instrument would-be/could-be re-calibrated on the
moon. Brian said probably not, because the engineers might not have
included that much latitude (6:1) in the operation of the calibration
or capability of the balance. I actually thought about that, and I
don't know the answer. ...

I am afraid that mentioning details of engineering constraints is
likely to dissuade any hardy spectators from following along with
this dialog.
Technical debate is invariably more enthralling to the protagonists
than to bystanders, I find.

So I will suggest only that you could call a scale
supplier's technical support desk and ask if their product will work
reliably in free fall. (There ARE suppliers who will say 'yes'.)

What sort of product is that, I wonder.... :-)

Thank you for giving me this opportunity to debate the merits.

Sincerely

brian whatcott <inet@intellisys.net>
Altus OK