Hi, Forrest
I'm
trying to build an electronic balance that was featured
in an October 2000 issue of Scientific American
magazine in "The Amateur Scientist"
column.
Greg
Schmidt wrote the code for an Atmel microcontroller
used in the balance. I want the code but haven't been
able to find it on the web.
I
have owned the "The
Amateur Scientist" CD for a few years, but
I have never done anything with it.
Now
I have an STK-500, an old ammeter, an IC with matched
transistors, and the hope that I might finally be
able to build a project.
Can you give me the code for the STK200 mentioned
in the article?
Sincerely
Yours,
Lorincz
Huff
Please see Shawn Carlson's Classics
column in the 12 November 2004 issue of TCS
(The
Amateur Scientist Classics: Measuring Micrograms)
for the answer to your question. Let us know how the
project works. Editor.
Shawn,
I enjoyed your balances, except for the
sad part, which is indeed a tragedy.
For sample masses, here's a thought: For
calibration, you don't need a weight that's exactly
the same every time you use it, but rather a way of
knowing precisely what its mass is when you do.
With that in mind, a single sheet of 20-pound
paper weighs about 5 grams. It's easy to weigh 100
(or 500) sheets to +/-1 gram, and it's easy to divide
a single sheet into 8 equal-area pieces with sub-1%
error using a paper cutter. That sample will weigh
1/800 as much as the stack of 100, and could serve
handily as an accurate weight of around 600 mg that's
accurate within one percent or so.
Of course, it won't have the same weight
from day to day, because paper absorbs water vapor
from the air. But if you simply keep the 100 sheets
in the same environment as the 1/8-sheet sample, you
can easily recalibrate by re-weighing the stack when
needed.
There is probably some unit-to-unit variation
in paper sheets (although I bet not much, given the
precision required to ensure reliable service in high-speed
printers). You could compensate for
That
by making smaller samples and combining them. For
example, cut 1/128-size samples (which are more than
1.25 cm on a side, and still fairly easy to handle)
from different positions on 16 different sheets.
With paper weights, all the cautions you
mention about handling are especially important. Don't
touch them, ever, and don't write anything on them
(not even guide lines for cutting). They also have
to be protected from dust. I imagine that storing
them between full sheets would be a reasonable way
to do that.
There are undoubtedly other mass-produced
items that are very close to identical and available
in large quantity (tiny glass or plastic beads, maybe?
or aluminum foil?). The same approach of weighing
a bunch and taking several samples to average out
unit-to-unit variation ought to be effective there.
I haven't actually tried this, but it seems
like it ought to work. It would be interesting
to measure the reliability of this approach by making
multiple weights that should be identical and testing
them.
Olin Sibert
Dear Olin,
An interesting idea. But it troubles
me a bit.
Measuring the weights of many sheets produces
a double average. The stack provides an average
measurement of the individual sheets, and each sheet
gives an average of the small parcels that make it
up. This method is an excellent way to find the average
weight of the tiny parcels. But one cannot assign
that weight to any randomly selected parcel until
one knows the width of the distribution that describes
parcel weights.
Consider if one were to divide the many
sheets of paper into parcel-sized units, and then
weigh each one and plot the distribution, one would
almost certainly find that they formed a Gaussian
distribution. The fractional width of that distribution
is going to be larger (and it may be much larger)
than the fractional width obtained by the average
obtained by measuring many sheets of paper. This is
because of errors in determining the paper size, as
well as non-uniformities in structure and materials
within the paper itself. Knowing the width of the
average tells us nothing about the width of the distribution
of parcel masses, except that the fractional width
of the parcel mass distribution must be wider. However,
we must know the width of the parcel mass distribution
to know what error must be assigned to the estimate
obtained from extrapolating an individual parcel mass
from the average parcel mass. Until one can have high
confidence that the mass of a randomly selected parcel
has a high likelihood of being within an acceptable
error of the average value, one cannot have high confidence
in one's results.
You may be right. Paper may be so uniformly
constructed that this extrapolation does not introduce
a significant error. But my point is that has to be
determined experimentally before such a method can
be relied on.
Your thoughts?
Shawn Carlson
Mr. Mims,
I'm
interested in looking for evidence of global atmospheric
trends. I remember that you suggested using data that's
already been gathered and is available online. While
surfing around and looking for a good source of data,
I began to ponder the concept of global warming.
One
EPA website for kids, "Global
Warming: What it is", says the earth has
warmed about 1 degree F over the last 100 years.
Temperature
alone doesn't provide the energy content of the atmosphere.
If I remember my thermodynamics, you must know the
specific heat of a medium in order to determine its
heat energy content (or heat energy density) at a
given temperature.
Do
you think scientists only mention temperature changes
because the public can relate to temperature? Do atmospheric
pressure and humidity become a significant part of
the calculation of specific heat for the atmosphere?
Thanks,
Jeff
Bledsoe
SAS
member
P.S.
Maybe we (my daughter and I) could present a display
(Proposed topic: Long-term Atmospheric Energy Density
Changes at Various Elevations) at the SAS conference
in January.
Temperature is the key parameter, but water
vapor certainly plays a role. Temperature determines
the amount of water vapor that a given volume of air
will hold. Warm air can hold more water vapor than
cool air. Water vapor is the most potent of the greenhouse
gases. Increasing the water vapor in the atmosphere
can trap more heat, thus increasing temperature. This
is positive feedback. On the other hand, increasing
the water vapor can also increase cloud cover, which
reduces the solar irradiance that increases the temperature.
This is negative feedback. The interactions of these
feedback mechanisms contribute to the uncertainty
of global warming models. So does the presence of
both absorbing and reflecting aerosols, including,
respectively, black carbon and sulfate haze. Recent
findings reported in Nature that the sun is more active
than at any time in the past 8,000 years may also
be very significant. The melting of glaciers is powerful
evidence that the earth is indeed warming. However,
urban heat island effects taint the historical temperature
record. For example, I have compared the temperature
record kept by Thomas Jefferson at Monticello, Virginia,
from 1810 to 1816 with the temperature measured nearby
by the Southeast Regional Climate Center from 1982
to 1994. This is a rural site, and the mean of the
modern measurements (13.2 C) is only 0.1 degree C
warmer than the mean of Jefferson's measurements (13.1
C). It's interesting that Jefferson believed that
the temperature in his day was warmer than during
the Roman era. Editor.
Forrest,
I saw the most amazing phenomenon today, but didn't
have quick access to a camera so I couldn't document
it.
A passenger jet, on its way out of the Providence,
Rhode Island, airport, passed overhead and left a
contrail. The individual traces from each engine remained
thin and distinct. They did not discernibly expand.
After about half a minute, the tracks each started
to become wavy, in an undulating pattern, like a sine
wave. Only each track was 180 degrees out of
phase with the other. The tracks separated and
came together and looked like a growing length of
fuzzy chain in the sky. I was quite surprised
to see such a strange pattern of relative motion appear
between the two tracks. But then I was astonished
to see that at the points where the tracks touched,
they almost instantly vanished. What's more,
the disappearance propagated in both directions along
both tracks from that point, until only small arcs
at the sine wave extremea, and then they too finally
disappeared. It looked as if some agent propagated
along the trails, rendering them invisible as it went.
I have never seen anything like this, and if someone
had described this to me without showing me photographic
evidence I would have been quite skeptical. But
I did observe this strange phenomenon. I have
no good explanation for this effect, except I believe
that one should start
looking at how the plane interacts with the atmosphere. Perhaps
the output from its engines was not regular, but pulsated
at some frequency that corresponded to the growth
of this strange pattern.
Anyway, you know much more about contrails than I
do. Have you ever seen or heard of this phenomenon
before? Any guesses as to the mechanism behind
it?
Shawn
P.S. I should tell you that there were
two witnesses. My 7-year old daughter, Katherine,
was with me.
Shawn,
Your
contrail report is quite fascinating.
I
have been a very active contrail observer since they
used to cool the tiny metal shed in which I worked
back in Albuquerque back in the 1970's. The cooling
was highly undesirable on cold winter days.
The
phenomenon you describe is totally new to me. I've
seen many examples of waves in contrails, but nothing
resembling what you describe. May I send your report
to Dr.
Lin Chambers, a NASA contrail expert? She will
be attending the SAS annual meeting.
I
previously covered contrails in TCS as a Citizen Scientist
Challenge. Let's run this in Backscatter as the
first a response to the Contrail Challenge.
Forrest
Hi
Forrest,
As to the contrail report, yes, please
do send it on to Dr. Chambers, and, of course, I am
happy to share it with our membership. Perhaps TCS
can run an article
on contrail observation for our members to get
involved with.
Yes I remembered the contrail
article you wrote. That article focused
on climate effects as I recall. My interest
at the moment is the contrails themselves.
For instance, if one knows the size of
commercial aircraft, visually identifies the plane
that creates a given contrail, and measures the angular
displacement, wing-tip to wing-tip and nose to tail,
of the plane against the sky, then the distance to
the plane can be estimated. If the time required for
it to move between two points of known angular separation
(simple hand-held instrument) is then measured, the
plane's speed can be found. The rate of drift of the
contrail in the sky tells
wind
speed. Non-uniformity in the drift over the track's
length reveals wind sheer along the surface defined
by the contrail. One can also measure the rate of
spreading, which would also be an interesting diagnostic.
BTW, your article states that spreading
is due to a transverse wind effect. I'll confess to
being confused as to why that should create spreading.
Why wouldn't it simply cause the whole line to drift?
To get spreading by such a mechanism, doesn't one
need sheer? A cirrus cloud spreads because it is distributed
over a fairly large region of space (right?) and the
wind speed is different at different locations, especially
altitudes. But that effect can't explain contrail
spreading because the height of the contrail cloud
starts out at roughly the width of the engine's exhaust
port. On the distance scale over which wind speed
changes, the contrail is very narrow (right?). What
am I missing here?
I have seen contrails expand to such a
degree that I have doubted whether I was watching
simple spreading. Rather, it looked like the presence
of the droplets, and possibly other particles as well
as turbulence and thermal energy in the plane's exhaust,
stimulated additional cloud formation that propagated
outward from the trail. Perhaps the air is sometimes
supersaturated with moisture and the disturbance created
by the plane stimulates a rapid phase transition that
propagates outward from the plane, like dropping a
tea bag into a microwaved cup of super-heated water.
It would be very interesting to see if one could find
a way to measure how much of the widening of the cloud
is due to simple spreading, and how much is due to
propagation of cloud growth. Could some of the spreading
effect be due to the cloud falling towards earth,
and thereby appearing to get larger as it approaches?
It seems to me that contrails could be
excellent probes of the atmosphere itself.
At any rate, I still wonder with amazement
at the phenomenon I witnessed. Like I said, without
photographic evidence to support this, I would not
have believed such eyewitness testimony.
Shawn
Shawn,
Thanks
for sending your contrail notes, all of which I would
very much like to use in "Backscatter."
The
TCS
Contrail Challenge was only an introduction.
The literature has many formal contrail studies that
discuss most of the points you raised, including spreading
of contrails. Several years ago there was a major
contrail campaign in which research aircraft flew
inside contrails at various distances from an aircraft
generating them.
I
have photographed spreading contrails that clearly
show the ice descending as it blows across the sky.
I showed one such slide at the Earth Explorers Institute
last week in Baltimore. The main contrail can also
be blown a considerable distance across the sky.
As
for measurements of aircraft speed and height from
contrails, based on my photointelligence experience,
I assume that was first done during WWII. Digital
cameras and webcams certainly make this a viable research
topic today.
Contrail
observations are among the projects in the draft research
proposal for the Earth Explorers Institute that I
prepared at the airport yesterday. I've been working
on this as a proposed extension for the Earth Observatory web site, and the
EEI offers an unexpected possibility. Dr.
Lin Chambers is the Principal Investigator for
GLOBE's contrail program.
Forrest
Send e-mails to "Backscatter"
at editor@sas.org. Place
"Backscatter" in the subject line. Letters
are subject to light editing for punctuation, grammar
and style. 
