On 2010-01-19, Frnak McKenney wrote:
Sorry, guys... I couldn't find any clean way to trim this. FAMcK

I put a lot of thought into my response, so I take this as a
complement. :-)

I'll trim out some of these where I don't see the need to
comment farther.

On 18 Jan 2010 21:43:20 GMT, DoN. Nichols wrote:
On 2010-01-18, Frnak McKenney wrote:

[ ... ]

Which of the following exercises in electronics, for example, should
be considered "assembling" and which would be "building":

O.K. I'll give my personal opinion of each of these as applied
to me.

o Plugging Arduino "shield" boards onto the processor board and
writing software for the package? (Think VME or S-100 if you're
not familiar with the Arduino grin!)

Having just looked up the "shield" boards, The only part of this
which *I* would consider "building" would be the writing of the
software.

Fair enough, DoN, since I think we're both assuming "boughten"
boards.

Right.

[ ... ]

o Etching your own PC board?

To your own design -- or using a layout from a magazine or other
publication?

I can see three sub-cases he etching a board pre-coated with
resist,

So -- there are sources of boards coated with resist to a
pattern, but not etched? Or do you mean "pre-coated with photoresist", in
which case you have to expose and develop it before etching? I've done
a lot of the latter, usually with my own layout on the boards. I even
set up a facility in the lab at work, using an 8x10" view camera and a
nice trans-illumination box on the wall to do this work. The camera was
mounted on a slide for a lathe-bed style optical bench.

At home, I did similar things, but with less professional
equipment. A 4x5" Crown Graphic camera and a opal Plexiglass backing
clamped to a piano bench type table. :-)

applying resist from a provided pattern,

Yes -- either from a silkscreen, or using photoresist and a
copier or a copier-based resist in today's world.

and applying one's
own circuit pattern generated by (say) Eagle.

Eagle is a computer program for PC layout I presume? I've now
got an open-source one on my unix systems for the same purpose.

Not sure what that
breakdown contributes to the discussion, though.

Well -- if you generate the pattern (other than by typing in a
board descriptor file from the actual source of the design) then I would
call that "building".

[ ... ]

o Screwing wires down on an old bread board instead of soldering
then together?

Almost certainly your own design of circuit, so "building".

But if I were following instructions from (say) the ARRL manual, or
a Boy Scout Merit Badge booklet, you would consider it to be
"assembling", yes?

How close will it look to what is in the book? If the layout is
taken as verbatim as you can manage, then "assembling". Otherwise, a
hint of "building" comes in -- especially if you change things around to
work better for your needs. (E.G. move the RF circuits closer to where
your antenna would come into the room, and the audio circuits closer to
your chair. :-)

o Using only discrete components, e.g. transistors and no ICs?

Duplicating someone else's published design, or your own design?

What confuses the issue is that "designs" come in different forms
and levels of detail. If we assume that a schematic is supplied,
but I select and purchase the component values specified and wire
them together, am I "building" or "assembling"?

Do you feel free to change the components a bit to make it work
better for your needs? An example would be higher power for the audio
output stage (assuming a radio receiver), or increasing the quality (and
type) of the coil and capacitors for higher "Q" for sharper tuning.

For a digital circuit -- things like changing the chip family
from say TTL to CMOS to allow lower power operation gets well into the
"building" side of things. (This would, among other things, translate
to different pin numbers, working purely from the logic diagram to
select the components, and data books to find which pins you need.

If I get a schematic and parts in a kit, but wire the device up on a
"proto" board with my own layout (and _lots_ of jumpers grin!),
does that change the answer/categorization?

It becomes an experimenter's build -- since the prototype board
suggests that you will be wanting to try different modifications to the
circuit. I don't think that you would use the prototype board for
something intended to be used unmodified without at least *trying*
changes. :-)

o Winding your own toroids (transformers), rubbing pencils on
paper to make resistors, and using tinfoil and wax paper for
capacitors? (These all work, by the way.)

Aside from the winding of torroids (or air-core or ferrite core
transformers), these are more a learning experience than a project to
make something for long-term use. But I would still class these as
"building" not "assembling".

Plus being a lot of work. grin!

Indeed so. Note that some circuits were made with resistor
values a bit too high, since the actual value depended on other
components and the luck of the draw, and using a pencil to selectively
lower the resistance value until it worked as desired. :-)

o Drawing your own wire and personally vacuuming your tubes?

*Serious* building. Probably weaving and waxing your own
insulation for the wires as well. :-)

Actually, that sounds like a lot of fun to do. Once or twice,
anyway. grin!

:-)

[ ... ]

I think that my distinction between assembling and building is
whether two or more people doing the same project would result in a
nearly identical final product. When personal creativity comes into
it, so things look different and behave differently, I would consider it
"building". If just following detailed instructions, then it is simply
"assembling".

It has taken me two cups of tea, some aspirin, and a couple of Diet
Cokes, but I think that I'm starting to follow you. (I can't
_believe_ that our current Congress is considering "health care
reform" legislation that doesn't provide funding for critical
treatments like sinus transplants!)

Hmm ... sinus transplants -- that sounds like a "building"
project -- at least for the first few tries. :-)

I think I see what you're getting at, and that means I have to
re-examine my initial comments. My original focus was as much on
what the person doing the building or assembling was accomplishing
for themselves -- the learning aspect -- as it was on the visible
result, what someone else could or would see. If I throw in the
term "construction" to cover both "assembly" and "building", then
perhaps it does make more sense to separate the construction process
from skill acquisition, even though I see it as vital to the
process. (And I guess I owe sparky an apology. grin!)

The skill acquisition may have happened at an earlier stage, and
you simply need this to work -- perhaps to help in another project. :-)

So... if we agree on the distinction between assembling and
building (or perhaps between "following instructions" and "creating
or modifying instructions"), can we use that to address the original
poster's query about inspiring the next generation?

O.K.

You mention building a number of kits. What was it that inspired
you to build your first one? If you saw "listening to the radio" as
something desirable, what made you think of building (okay,
"assembling") the AR-15

Hmm ... what I assembled was an AJ-15 (the stand-alone tuner),
having already designed and made my own amplifier and preamp. The other
two in the family we

AA-15 -- same chassis -- but only amplifier and preamp in the
chassis.

AR-15 -- same chassis but tuner, amplifier and preamp all in the
same chassis.

as a way to accomplish this rather than
buying one already built for you?

I liked assembling, as well as designing and building. And it
got me a higher quality tuner than I could have otherwise afforded.

For that matter, what allowed you
to see yourself as someone who _could_ build a kit, as a person who
could accomplish the leap from a catalog description to the finished
receiver?

Well ... I had assembled quite a few kits some years before --
mostly test equipment kits (multimeter and RF signal generator as
examples). The purpose was to use in assisting designing, building, and
troubleshooting projects of my own.

Examples of projects of my own were transistor testers and
temperature controllers. One of the latter was used to control the
flat heater under my (now long defunct) waterbed to maintain a proper
temperature through variations in environment. (It was in an
concrete-floored apartment on linoleum tiles, so otherwise it would have
changed with the weather.) A second temperature controller cycled Rotron
muffin fans blowing air into the base of the shrouded radiator
assemblies which were mostly blocked by the frame of the waterbed. This
allowed me to control the room temperature as well. Later, after moving
to my own house (and joining with a wife) one of those was repurposed to
control the mix of hot and cold water into a sink to stabilize the
temperature of color slide processing.

Note that one of the characteristics of things I "built" instead
of "assembled" is that they were designed around the used and cheap
parts which I could find at various places, instead of parts selected by
someone else -- probably with the same criteria, but with a different
"junk box" at hand. :-) Also, someone designing a project for
production, as a kit or as a magazine article, is rather constrained to
select parts which are readily available -- and no two junk boxes are
alike. :-)

[ ... ]

Quite a few Heathkits (and other brands, such as Paco for test
equipment) prior to and after that. I had no desire for a TV, which is
probably why I did not build that kit. I still have the Paco RF signal
generator, FWIW -- one of my first kits.

I repaired a few TV sets during my high school years, but I have to
admit that I've never built a TV from the ground up, kit or
otherwise. The Mohican shortwave receiver my father and I built so
my mother could pick up her father's amateur transmitter in New
Jersey is still running back in my "computer lab", albeit with a few
new (socketed!) audio transistors courtesy of NTE (Ge transistors are
_expensive_!).

Now they are. (Hmm ... for that matter, back then they were
too, at the start. I remember my first two transistors (Raytheon
"CK-722"s) cost $7.50 each -- back about 1955 or 1956 I think.) The
prices fell drastically for a while as production ramped up, and then
starting climbing again as they were replaced by Silicon transistors.

[ ... ]

Yes -- but most of the kits were simply a means to an end.
Sometimes that end was building things of my own design.

And these qualities seem to build on one another. The satisfaction
of completing a project makes one more eager to take on a new one,
and learning (say) how to cast metal adds the concept of creating a
new part rather than searching catalogs for it a part of one's
problem-solving repertoire.

Indeed.

Perhaps that's a key, and one I overlooked when I spoke about the
Heath TV set: some people enjoy the feedback they get from others
about what they have accomplished, and some enjoy the feeling they
get from accomplishing something, but if one gets _no_ enjoyment
from a project it's hard to sustain an interest in it.

Understood.

BTW My first acquisition of machine tools was to make nicer looking
projects. Things like a Unimat SL-1000 lathe/mill for cutting
out nice rectangular holes for meters in panels.

And seeing the wonderful sparks one can
create with simple household items such as a screwdriver and a lamp
socket often leads one to a deeper understanding of electricity (and
fuses and circuit breakers grin!).

Reminds me of some fellows in one of the dorms which had thermal
only circuit breakers. As a result, they could arc weld directly from
the line for about fifteen seconds, then let it cool down for a couple
of minutes before the next weld cycle. :-)

Should I ask _what_ they were welding in a dorm room? grin!

I think that they were welding two steel trash cans together. :-)
Base to base, I think, but I did not actually see it, just heard from
others who saw it.

All that said, here's a paradox for you: much, perhaps all, of the
"creativity" I'm describing depends on at least two things: the
"rote assembly" effort required to _provide_ whatever basic building
blocks one happens to be using at any given time, and the time and
other resources needed to study the problem at hand and devise --
and hopefully implement -- a solution.
--snip--

Actually -- my early PC work started with hand painted resist to
draw the circuit -- followed by etching in nitric acid. :-)

Resist pen? Or an actual paintbrush?

Actual paintbrush. The resist came in a wide-mounted small jar,
from Lafayette, IIRC.

Later, the photo-resist allowed me to lay out with tape on
paper, then make a photographic negative (Kodalith Ortho), and expose,
develop and etch the board (using FeCl) -- none of the heat transfer
starting from computer images and photocopies onto special films.

I'm looking at making the move from point-to-point wiring on "proto"
boards to etching my own; too many useful and interesting parts are
only available in SMT packages. Problem is that I need better
eyesight or (and?) a good assembly/inspection scope to begin to
handle the little buggers people use as discretes these days.

Look for one of the B&L or American Optical stereo zoom
microscopes. A 0.7 - 3.0 zoom range, with 10X eyepieces gives you a
7-30X zoom range which is quite nice for the purpose. Get one with a
boom mount.

And
sneezing... _definitely_ have to give up sneezing. grin!

Wear a surgical mask. Or perhaps a dust mask and have plenty of
replacement filters for it. :-0

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

On 23 Jan 2010 04:23:17 GMT, DoN. Nichols wrote:
On 2010-01-19, Frnak McKenney wrote:
Sorry, guys... I couldn't find any clean way to trim this. FAMcK

I put a lot of thought into my response, so I take this as a
complement. :-)

I'll trim out some of these where I don't see the need to
comment farther.

"'Tis a consummation devoutly to be wished." grin!

On 18 Jan 2010 21:43:20 GMT, DoN. Nichols wrote:
On 2010-01-18, Frnak McKenney wrote:

[ ... ]

Which of the following exercises in electronics, for example, should
be considered "assembling" and which would be "building":

O.K. I'll give my personal opinion of each of these as applied
to me.

o Plugging Arduino "shield" boards onto the processor board and
writing software for the package? (Think VME or S-100 if you're
not familiar with the Arduino grin!)

Having just looked up the "shield" boards, The only part of this
which *I* would consider "building" would be the writing of the
software.

Fair enough, DoN, since I think we're both assuming "boughten"
boards.

Right.

[ ... ]

o Etching your own PC board?

To your own design -- or using a layout from a magazine or other
publication?

I can see three sub-cases he etching a board pre-coated with
resist,

So -- there are sources of boards coated with resist to a
pattern, but not etched? Or do you mean "pre-coated with photoresist", in
which case you have to expose and develop it before etching? I've done
a lot of the latter, usually with my own layout on the boards. I even
set up a facility in the lab at work, using an 8x10" view camera and a
nice trans-illumination box on the wall to do this work. The camera was
mounted on a slide for a lathe-bed style optical bench.

While I don't know of any source for un-etched boards with resist
pre-applied, that was the distinction I was trying to make. I could
see a case being made for including one or two such in, say, an
"Introduction to PCB Etching" kit/course so the student could have a
sample of "professional" quality, I don't see them as a high-volume
item. grin!

At home, I did similar things, but with less professional
equipment. A 4x5" Crown Graphic camera and a opal Plexiglass backing
clamped to a piano bench type table. :-)

applying resist from a provided pattern,

Yes -- either from a silkscreen, or using photoresist and a
copier or a copier-based resist in today's world.

and applying one's
own circuit pattern generated by (say) Eagle.

Eagle is a computer program for PC layout I presume? I've now
got an open-source one on my unix systems for the same purpose.

CadSoft Online -- Eagle 5.7
Schematic Capture and PCB Design Package
Linux, Windows, Mac
http://www.cadsoft.de/

I've used the "freeware" version several times as an electronic
sketchpad. If/when I get to the point of laying out and etching my
own printed circuit boards I'll probably spring for the "Light"
($50) version.

Not sure what that
breakdown contributes to the discussion, though.

Well -- if you generate the pattern (other than by typing in a
board descriptor file from the actual source of the design) then I would
call that "building".

[ ... ]

o Screwing wires down on an old bread board instead of soldering
then together?

Almost certainly your own design of circuit, so "building".

But if I were following instructions from (say) the ARRL manual, or
a Boy Scout Merit Badge booklet, you would consider it to be
"assembling", yes?

How close will it look to what is in the book? If the layout is
taken as verbatim as you can manage, then "assembling". Otherwise, a
hint of "building" comes in -- especially if you change things around to
work better for your needs. (E.G. move the RF circuits closer to where
your antenna would come into the room, and the audio circuits closer to
your chair. :-)

o Using only discrete components, e.g. transistors and no ICs?

Duplicating someone else's published design, or your own design?

What confuses the issue is that "designs" come in different forms
and levels of detail. If we assume that a schematic is supplied,
but I select and purchase the component values specified and wire
them together, am I "building" or "assembling"?

Do you feel free to change the components a bit to make it work
better for your needs? An example would be higher power for the audio
output stage (assuming a radio receiver), or increasing the quality (and
type) of the coil and capacitors for higher "Q" for sharper tuning.

For a digital circuit -- things like changing the chip family
from say TTL to CMOS to allow lower power operation gets well into the
"building" side of things. (This would, among other things, translate
to different pin numbers, working purely from the logic diagram to
select the components, and data books to find which pins you need.

Plus having different rules about unused inputs and outputs, and
adapting the surrounding discretes such as pull-up resistors.

If I get a schematic and parts in a kit, but wire the device up on a
"proto" board with my own layout (and _lots_ of jumpers grin!),
does that change the answer/categorization?

It becomes an experimenter's build -- since the prototype board
suggests that you will be wanting to try different modifications to the
circuit. I don't think that you would use the prototype board for
something intended to be used unmodified without at least *trying*
changes. :-)

"There _must_ be a 3.3k (3k3) resistor around here somewhere..."

On 26 Jan 2010 06:31:04 GMT, DoN. Nichols wrote:
On 2010-01-25, Frnak McKenney wrote:
On 25 Jan 2010 05:28:54 GMT, DoN. Nichols wrote:

[ ... ]
[...]
That is, what kind of projects led you to
think that you needed -- or simply had a use for -- that test
equipment?

I had been experimenting with components form stripped down
discarded equipment for a long time, and would occasionally try to build
something -- and needed a way of assuring that the components were as
marked, and the voltages were near what was expected.

You know, I never thought about that aspect of it. I could always
use my father's Simpson analog multimeter (as long as he was at work
or on a railfan trip). I just had to be very careful to put things
away exactly where I found them... and with three brothers,
"plausible deniability" was fairly easy.grin!

O.K. My parents did have any test equipment at all -- and only
a couple of tiny claw hammers with one claw broken off each. :-)

(I'll assume that was a "didn't".)

So you had to carry all your projects to the local TV repair shop,
six miles through the snow and uphill both ways? grin!


[ ... ]
[...]
All wiyht. Rho sritched mg kegtops awound?

Hmm ... a reminder of another useful habit of the geek/scrounger
type. When dealing with rebooting everything, I had to lift the
keyboard for the firewall down from the top of the rack, and it disloged
a Cenco "Lab Jack" which fell corner first on the Sun keyboard which
switches between the Mac Mini and the Sun Fire 280R (Rack mount server
version of the Sun Blade 1000/2000).

Ouch!

I know the feeling. I think I need to build a robot that is smart
enough to move my cup of tea a second or two _before_ my forearm
knocks it over onto my desk. (Did you notice the 10-minute gap in
my typing? grin!)

... It totally destroyed the keycap
for "Page Down" -- so I went to the peanut can full of keycaps from
another keyboard of the same construction, and a bit of digging through
found the proper replacement key cap. Just a bit lighter in color. :-)
Luckily, the keyswitch below it survived, and it was just the key cap.
I've had to disassemble some of these keyboards and take the flexible
printed circuits and other parts down to the shop and use compressed air
to remove the accumulation of cat hair which had worked between the
layers.

So you were able to finally justify the twenty-odd keyboards in
various states of disrepair that had been cluttering up your
basement for the past ten years? (Sorry -- thinking about my own
basement, not yours. grin!)

But seriously... if you're interested in an overpriced and
generally impractical solution to your Damaged Keytop Problem, a
solution that would suck up _hours_ of your time, check out the
latest issue of "Make" magazine (v21, January 2010); its theme is
"Your Desktop Factory: 3D Manufacturing at Home".

Here's what you would do:

1) Build the SplineScan "DIY 3D Scanner" from p.54
2) Find an indentically-shaped keytop. If it's a specially-shaped
Enter key, glue and sand the broken one; if you can't find all
the pieces, patch it with wax or plaster.
3) Scan the keytop.
4) Use the scanned model to create a 3D model.
5) Create a replacement keytop using a 3D plotter like the
mis-named "CupCake CNC" device from MakerBot described on p.46,
or one of a number of similar gadgets described on assorted 'web
sites.
6) Sand, drill, letter, and paint the result.

But here's the "leverage" part: if you post your 3D model on a
publicly accessible site, the next person with a broken Sun keyboard
can skip steps 1-4, since he only has to download your model.

If you think about it, this is a marvelous way for manufacturers to
save on parts inventory, the ultimate JIT method for keeping a huge
number of plastic parts readily available for decades (or even
centuries -- "You, Too Can Build Your Own 2059 Edsel!"). The only
ongoing expenses are the 3D printer, "toner", a stack of read-only
32Gb SD chips, and someone to do the lookup and kick off the machine
as required.

Oh, and someone to collect the money. grin!

Seriously, I realise that you can't cost-justify such a setup to
replace one broken keytop, any more than you can justify building
even a 78RPM phonograph disc cutter and a playback unit simply to
make a permanent copy of your child's first word, even if it's
something as world-shattering as "photosynthesis". But... record a
thousand songs and make it possible for others to listen to music
whenever they want, and you've created an entire industry.

You should be able to pick up a copy of Make v21 at Borders or
Barnes&Noble, or check their 'web site at www.makezine.com.

* * *

Jumping back to long-snipped digression in this thread, a discussion
which might even have been with you -- pencil-drawn resistors -- the
other day I stumbled on a DIY project called "DrawDIO!" that
actually _uses_ this effect as a part of yet another re-invention of
the Theremin:

http://www.ladyada.net/make/drawdio/

The circuit is, like most really good ideas, simple enough once you
know the secret: a 555 multivibrator controlled by pencil-drawn
lines. What I found most impressive was the number of different ways
people had found to _use_ the device. I think someone earlier in
this thread felt that creativity required manual dexterity; see if
you agree after viewing the video clips showing the DrawDIO off.


Frank
--
I can explain it to you, but I can't understand it for you.
--
Frank McKenney, McKenney Associates
Richmond, Virginia / (804) 320-4887
Munged E-mail: frank uscore mckenney ayut mined spring dawt cahm (y'all)

On 2010-01-27, Frnak McKenney wrote:
On 26 Jan 2010 06:31:04 GMT, DoN. Nichols wrote:

[ ... ]

O.K. My parents did have any test equipment at all -- and only
a couple of tiny claw hammers with one claw broken off each. :-)

(I'll assume that was a "didn't".)

You are correct. My fingers were apparently lazy. (And, of
course, the spelling checker could not detect that as an error. :-)

So you had to carry all your projects to the local TV repair shop,
six miles through the snow and uphill both ways? grin!

Yep! Or assemble my own stuff from kits.


[ ... ]
[...]
All wiyht. Rho sritched mg kegtops awound?

Hmm ... a reminder of another useful habit of the geek/scrounger
type. When dealing with rebooting everything, I had to lift the
keyboard for the firewall down from the top of the rack, and it disloged
a Cenco "Lab Jack" which fell corner first on the Sun keyboard which
switches between the Mac Mini and the Sun Fire 280R (Rack mount server
version of the Sun Blade 1000/2000).

Ouch!

I know the feeling. I think I need to build a robot that is smart
enough to move my cup of tea a second or two _before_ my forearm
knocks it over onto my desk. (Did you notice the 10-minute gap in
my typing? grin!)

:-)

... It totally destroyed the keycap
for "Page Down" -- so I went to the peanut can full of keycaps from
another keyboard of the same construction, and a bit of digging through
found the proper replacement key cap. Just a bit lighter in color. :-)
Luckily, the keyswitch below it survived, and it was just the key cap.
I've had to disassemble some of these keyboards and take the flexible
printed circuits and other parts down to the shop and use compressed air
to remove the accumulation of cat hair which had worked between the
layers.

So you were able to finally justify the twenty-odd keyboards in
various states of disrepair that had been cluttering up your
basement for the past ten years? (Sorry -- thinking about my own
basement, not yours. grin!)

Well ... I probably have that many keyboards (mostly Sun)
scattered around -- dating back to the one for the Sun 2/120 Two changes
of interface connector back. :-)

But when one keyboard got damaged by an Exabyte EZ-17 tape
jukebox falling on my head and then to one of the mail server's
keyboards (sticking out a bit from the bottom of the server rack), it
broke the frame among other things, so I popped off and saved all the
surviving keycaps, since I sometimes find keyboards at hamfests with a
keycap or two missing. They all (except for the spacebar) fit in a
single Peanuts can. I've had to use these a few times since. (The
keyboard was a fairly recent style -- available both in Sun's DIN
connector interface and in USB, so being able to keep these going is
nice.

But seriously... if you're interested in an overpriced and
generally impractical solution to your Damaged Keytop Problem, a
solution that would suck up _hours_ of your time, check out the
latest issue of "Make" magazine (v21, January 2010); its theme is
"Your Desktop Factory: 3D Manufacturing at Home".

Here's what you would do:

1) Build the SplineScan "DIY 3D Scanner" from p.54
2) Find an indentically-shaped keytop. If it's a specially-shaped
Enter key, glue and sand the broken one; if you can't find all
the pieces, patch it with wax or plaster.
3) Scan the keytop.
4) Use the scanned model to create a 3D model.
5) Create a replacement keytop using a 3D plotter like the
mis-named "CupCake CNC" device from MakerBot described on p.46,
or one of a number of similar gadgets described on assorted 'web
sites.
6) Sand, drill, letter, and paint the result.

I agree that it is what we *should* do -- but there is one thing
left out of that approach which I consider important. The Sun keyboards
have two-shot molded keycaps -- in which a different color of plastic
lines the underside of the keycap, and protrudes through to make the
visible marking. Painted (or silkscreened) lettering wears off rather
quickly, as I have seen on some cheap PC keyboards. (One reason why I
am using a Sun keyboard on my Mac Mini as well.)

But here's the "leverage" part: if you post your 3D model on a
publicly accessible site, the next person with a broken Sun keyboard
can skip steps 1-4, since he only has to download your model.

If you think about it, this is a marvelous way for manufacturers to
save on parts inventory, the ultimate JIT method for keeping a huge
number of plastic parts readily available for decades (or even
centuries -- "You, Too Can Build Your Own 2059 Edsel!"). The only
ongoing expenses are the 3D printer, "toner", a stack of read-only
32Gb SD chips, and someone to do the lookup and kick off the machine
as required.

Hmm ... make a jukebox for the SD chips so it can select the
proper one automatically? And make them micro-SD chips to make the
storage density even higher. Since each chip could have an identifier
in a standard file -- you could even just ship a handful of chips, dump
them in, and let the computer sort them and store them in the proper
place. :-)

Oh, and someone to collect the money. grin!

:-)

Seriously, I realise that you can't cost-justify such a setup to
replace one broken keytop, any more than you can justify building
even a 78RPM phonograph disc cutter and a playback unit simply to
make a permanent copy of your child's first word, even if it's
something as world-shattering as "photosynthesis". But... record a
thousand songs and make it possible for others to listen to music
whenever they want, and you've created an entire industry.

Actually -- I am planing to make a Edison Cylinder playback
machine using a low-speed servo motor.

You should be able to pick up a copy of Make v21 at Borders or
Barnes&Noble, or check their 'web site at www.makezine.com.

I'll keep an eye open for that.

* * *

Jumping back to long-snipped digression in this thread, a discussion
which might even have been with you -- pencil-drawn resistors -- the
other day I stumbled on a DIY project called "DrawDIO!" that
actually _uses_ this effect as a part of yet another re-invention of
the Theremin:

http://www.ladyada.net/make/drawdio/

The circuit is, like most really good ideas, simple enough once you
know the secret: a 555 multivibrator controlled by pencil-drawn
lines. What I found most impressive was the number of different ways
people had found to _use_ the device. I think someone earlier in
this thread felt that creativity required manual dexterity; see if
you agree after viewing the video clips showing the DrawDIO off.

Interesting. I remember building an array of oscillators using
CMOS gates as driving elements, and a high valued resistor and a trimmer
capacitor as tuning elements. The major problem was that it was not
stable with temperature. I wonder whether using low TC mica capacitors
and hand-drawn resistors would produce something more stable. (I didn't
use trimmer resistors because I had a lot of the trimmer capacitors from
a surplus deal, but not enough trimmer resistors -- and that many would
have been *expensive*. :-)

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

On Jan 27, 5:00*pm, "DoN. Nichols" wrote:
...I wonder whether using low TC mica capacitors
and hand-drawn resistors would produce something more stable. *...
* * * * * * * * DoN.

I was curious and checked the temperature change of several types of
caps with a Boonton meter. Baked-bean micas changed perhaps 5% from
ice water to a heat gun blast. Multilayer ceramics dropped about in
half when heated, tantalums much more. The 1% metal film resistors
that company used for everything didn't change much at all.

jsw

Jim Wilkins wrote:
On Jan 27, 5:00 pm, "DoN. Nichols" wrote:
...I wonder whether using low TC mica capacitors
and hand-drawn resistors would produce something more stable. ...
DoN.

I was curious and checked the temperature change of several types of
caps with a Boonton meter. Baked-bean micas changed perhaps 5% from
ice water to a heat gun blast. Multilayer ceramics dropped about in
half when heated, tantalums much more. The 1% metal film resistors
that company used for everything didn't change much at all.

1% metal film chip resistors are a miracle of the
modern world. What else in the world costs 3 for
a penny and has 1% accuracy?

On Jan 27, 6:13*pm, Jim Stewart wrote:
...

1% metal film chip resistors are a miracle of the
modern world. *What else in the world costs 3 for
a penny and has 1% accuracy?

Coins themselves are very accurate in weight:
http://www.usmint.gov/about_the_mint...specifications

IIRC I found a variation among unworn nickels of less than +- 5
milliGrams. There is never a weighing scale in an electronics lab, but
I could weigh small parts within a few percent with nickels and a
ruler balanced on a round pencil. I did it to figure out specific
gravity to identify materials such as small hardware, which is
difficult to guess if it's too small to feel the weight or thermal
conductivity.

jsw

On 1/27/2010 5:00 PM, DoN. Nichols wrote:

Interesting. I remember building an array of oscillators using
CMOS gates as driving elements, and a high valued resistor and a trimmer
capacitor as tuning elements. The major problem was that it was not
stable with temperature. I wonder whether using low TC mica capacitors
and hand-drawn resistors would produce something more stable. (I didn't
use trimmer resistors because I had a lot of the trimmer capacitors from
a surplus deal, but not enough trimmer resistors -- and that many would
have been *expensive*. :-)

Back when I used to do this sort of thing, the standard method was to
use a varistor with a TC inverse to that of the capacitor. If you wanted
an LC oscillator, there was a ferrite core material that had a
complementary TC to polystyrene caps.

Kevin Gallimore

On Jan 27, 9:13*pm, axolotl wrote:
On 1/27/2010 5:00 PM, DoN. Nichols wrote:

* *Interesting. *I remember building an array of oscillators using
CMOS gates as driving elements,...

Back when I used to do this sort of thing, the standard method was to
use a varistor with a TC inverse to that of the capacitor. If you wanted
an LC oscillator, there was a ferrite core material that had a
complementary TC to polystyrene caps.

Kevin Gallimore

You were making low frequency sine oscillators?

The ones I worked with were all digital, even for 60Hz to synchronize
A/D sampling to the power line. The standard method was a VCO
referenced to a crystal or ceramic resonator, or the power line zero
crossing. I designed one circuit that counted down a coffee-cup-sized
rubidium beam reference, AKA an atomic clock.

A clever engineer taught me several analog tricks with spare CMOS
gates. One 4050 buffer with a resistor from out to in is a flip-flop.
Connect the input to the center of a SPDT switch, NC and NO to GND and
VDD, and it's a switch debouncer that holds the last momentary input
spike.

Schmitt trigger gates are very versatile:
http://www.fairchildsemi.com/an/AN/AN-140.pdf

jsw

On 1/28/2010 7:53 AM, Jim Wilkins wrote:


You were making low frequency sine oscillators?

Mostly V to F converters at the analog computer factory. I was fortunate
to work with _real_ good analog designers. My son studied my Handbook of
Analog Computation (printed for in house use at the factory) when he was
in school, and because of that ended up in the awkward situation of
having to correct his interviewer's circuit explanation during a job
interview.

(At a ummm.. research facility in New Hampshire)

Kevin Gallimore

On 2010-01-27, Jim Wilkins wrote:
On Jan 27, 5:00*pm, "DoN. Nichols" wrote:
...I wonder whether using low TC mica capacitors
and hand-drawn resistors would produce something more stable. *...
* * * * * * * * DoN.

I was curious and checked the temperature change of several types of
caps with a Boonton meter. Baked-bean micas changed perhaps 5% from
ice water to a heat gun blast. Multilayer ceramics dropped about in
half when heated, tantalums much more. The 1% metal film resistors
that company used for everything didn't change much at all.

O.K. So the micas would probably have been an improvement. The
trimmer caps were ceramic ones in the low pF range. However, the
resistors which I had available were the old carbon composition
resistors (RC20GF???J IIRC). And I don't know how the CMOS thresholds
might have changed with temperature.

I *am* amazed that the multilayer ceramics had that much change.

Thanks,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

On 2010-01-28, axolotl wrote:
On 1/27/2010 5:00 PM, DoN. Nichols wrote:

Interesting. I remember building an array of oscillators using
CMOS gates as driving elements, and a high valued resistor and a trimmer
capacitor as tuning elements. The major problem was that it was not
stable with temperature. I wonder whether using low TC mica capacitors
and hand-drawn resistors would produce something more stable. (I didn't
use trimmer resistors because I had a lot of the trimmer capacitors from
a surplus deal, but not enough trimmer resistors -- and that many would
have been *expensive*. :-)

Back when I used to do this sort of thing, the standard method was to
use a varistor with a TC inverse to that of the capacitor. If you wanted
an LC oscillator, there was a ferrite core material that had a
complementary TC to polystyrene caps.

Interesting approaches. I think that I might have done well to
simply set up a divider chain from a crystal oscillator. All the target
frequencies were in the audio range, and not particularly sensitive to
harmonics, so the square-wave output would be easy enough to use.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

On Thu, 28 Jan 2010 23:47:03 +0000, DoN. Nichols wrote:
On 2010-01-27, Jim Wilkins kb1dal... wrote:
On Jan 27, 5:00Â*pm, "DoN. Nichols" ... wrote:
...I wonder whether using low TC mica capacitors and hand-drawn
resistors would produce something more stable. Â*...

I was curious and checked the temperature change of several types of
caps with a Boonton meter. Baked-bean micas changed perhaps 5% from ice
water to a heat gun blast. Multilayer ceramics dropped about in half
when heated, tantalums much more. The 1% metal film resistors that
company used for everything didn't change much at all.

O.K. So the micas would probably have been an improvement. The
trimmer caps were ceramic ones in the low pF range. However, the
resistors which I had available were the old carbon composition
resistors (RC20GF???J IIRC). And I don't know how the CMOS thresholds
might have changed with temperature.

I *am* amazed that the multilayer ceramics had that much change.

For several kinds of high-dielectric MLCC's[*], capacitance
decreases as higher voltage is applied, which for many apps is a
worse problem than temperature dependence. For example, see
chart near middle of http://my.execpc.com/~endlr/ceramic.html
where it shows that Z5U-dielectric MLCC's have 120% of rated
capacitance at 10% of rated voltage; 50% at 50%; and 30% at 100%.

As the article says later: "Why use them if their electrical
properties are inferior to C0G? Size and cost. You will rarely
see C0G larger than 0.047 uF (at least as a standard product),
but Z5U can be found as large as 22 uF, even in SMD. Poor as it
is by C0G standards, a 22 uF Z5U can compete with the equivalent
electrolytic in many high-frequency applications with a much lower
ESR."

Also see www.murata.com/cap/measure.pdf and
www.niccomp.com/catalog/nmc2.pdf.

* MLCC = one or more of "multi-layer ceramic capacitor",
"multi-layer ceramic chip", or "multi-layer chip capacitor"
SMD = surface mount device.
--
jiw

James Waldby wrote:

On Thu, 28 Jan 2010 23:47:03 +0000, DoN. Nichols wrote:
On 2010-01-27, Jim Wilkins kb1dal... wrote:
On Jan 27, 5:00Â pm, "DoN. Nichols" ... wrote:
...I wonder whether using low TC mica capacitors and hand-drawn
resistors would produce something more stable. Â ...

I was curious and checked the temperature change of several types of
caps with a Boonton meter. Baked-bean micas changed perhaps 5% from ice
water to a heat gun blast. Multilayer ceramics dropped about in half
when heated, tantalums much more. The 1% metal film resistors that
company used for everything didn't change much at all.

O.K. So the micas would probably have been an improvement. The
trimmer caps were ceramic ones in the low pF range. However, the
resistors which I had available were the old carbon composition
resistors (RC20GF???J IIRC). And I don't know how the CMOS thresholds
might have changed with temperature.

I *am* amazed that the multilayer ceramics had that much change.

For several kinds of high-dielectric MLCC's[*], capacitance
decreases as higher voltage is applied, which for many apps is a
worse problem than temperature dependence. For example, see
chart near middle of http://my.execpc.com/~endlr/ceramic.html
where it shows that Z5U-dielectric MLCC's have 120% of rated
capacitance at 10% of rated voltage; 50% at 50%; and 30% at 100%.

As the article says later: "Why use them if their electrical
properties are inferior to C0G? Size and cost. You will rarely
see C0G larger than 0.047 uF (at least as a standard product),
but Z5U can be found as large as 22 uF, even in SMD. Poor as it
is by C0G standards, a 22 uF Z5U can compete with the equivalent
electrolytic in many high-frequency applications with a much lower
ESR."

Also see www.murata.com/cap/measure.pdf and
www.niccomp.com/catalog/nmc2.pdf.

* MLCC = one or more of "multi-layer ceramic capacitor",
"multi-layer ceramic chip", or "multi-layer chip capacitor"
SMD = surface mount device.
--
jiw


Z5U are crap grade capacitors intended for bypass applications. They
are not suited for tuned circuits.



--
Greed is the root of all eBay.

On 2010-01-29, James Waldby wrote:
On Thu, 28 Jan 2010 23:47:03 +0000, DoN. Nichols wrote:

[ ... ]

O.K. So the micas would probably have been an improvement. The
trimmer caps were ceramic ones in the low pF range. However, the
resistors which I had available were the old carbon composition
resistors (RC20GF???J IIRC). And I don't know how the CMOS thresholds
might have changed with temperature.

I *am* amazed that the multilayer ceramics had that much change.

For several kinds of high-dielectric MLCC's[*], capacitance
decreases as higher voltage is applied, which for many apps is a
worse problem than temperature dependence. For example, see
chart near middle of http://my.execpc.com/~endlr/ceramic.html
where it shows that Z5U-dielectric MLCC's have 120% of rated
capacitance at 10% of rated voltage; 50% at 50%; and 30% at 100%.

Hmm ... sounds like a basis for making a FM transmitter, if the
temperature coefficient were not too bad -- or if kept in a stable
temperature environment.

As the article says later: "Why use them if their electrical
properties are inferior to C0G? Size and cost. You will rarely
see C0G larger than 0.047 uF (at least as a standard product),
but Z5U can be found as large as 22 uF, even in SMD. Poor as it
is by C0G standards, a 22 uF Z5U can compete with the equivalent
electrolytic in many high-frequency applications with a much lower
ESR."

That sounds good.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---