[sdiy] Building stock (try #3)
René Schmitz
uzs159 at uni-bonn.de
Wed Jun 7 01:06:56 CEST 2006
Hi Spencer and all,
Spencer Johnson wrote:
> i'm having this same problem. I'm trying to get together a "lab" of
> sorts, for basic expirementation. really all i need is a list of the
> most common values of passives (i.e. - caps, resistors, pots, maybe
> even crystals, etc.) that i would need to use.
I find stocking caps is difficult since there are not only varieties in
value, but also in size, dielectrics and voltages (anything above 63V
will generally work). So one projects' PCB layout might call for a 10n
in 5mm the other in 7.5mm... While all you might have at hand is in 15mm
with 630V rating (if, like me, you'd build tube stuff)... Of course if
you do perfboard this is no problem.
I have occasionally bought bags of randomly assorted caps and roughly
sorted them into "decade bins". Not what I would count on this for
projects, but usually it gives you lots of values (or at least enough
material to parallel) at hand that you can use in a pinch. Deciphering
the codes can sometimes be difficult though.
The most common cap would be the 100n ceramic, which is frequently used
as bypass cap. (Bought a bag of 500 of them for 3 DM some years ago, and
the drawer still has some left.) Wouldn't hurt if you found a bargain of
a value in that neighborhood...
The electrolytics values I keep in stock are 0.22u, 0.47u, 1u, 2.2u,
4.7u, 10u, 22u, 47u, 100u, 220u, 470u, 1000u. The three largest are
practically only used in power supply circuits.
I think this same scheme of 1, 2.2, 4.7 would also make a nice base for
caps between 10p and 470n. You can fill up the values in between later.
The ones below 100p could be ceramics. The other ones mylar.
Caps for VC-filters or time constants in VCOs should be higher quality
dielectrics, like styrene or polycarbonate. The most interesting range
is probably 220p to 2.2n, with only a few values like 220p, 330p, 470p,
1n, 2.2n.
For resistors, I'd at least buy the E6 or E12 values from 100 ohms to
1Meg. Use 1% metal film resistors. Some other values like 4.99k, 49.9k
and 2k, 20k, 200k would be handy too. The most common will be the 100ks.
I wouldn't recommend "random" assortments here, if you buy in 100s,
resistors are fairly cheap. Sometimes you can find complete sets with
all values in seperate bags even with a nice drawer box.
The most common pot values would be 100k, 47k, 22k, 10k in both log and
lin. Some ADSRs call for larger valued pots like 1 or 2.2Meg log.
Mostly these values are somewhat uncritical, so that you can sometimes
substitute one value for another. Infact some manufacturers make only
50k and 20k instead of 47k and 22k.
Trimpots have more variety in values, and again there is the difficulty
of varying footprints. Standing, laying, 5mm 7.5mm, 19mm, 2.5mm inline
and so on. The values from 100ohms to 1Meg in a 1, 2.2, 4.7 scheme might
be useful. Though its probably best to buy these when you need them.
Generally speaking, it is often not very critical what exact value you
use for a particular passive. A VCO might call for a 2.2n while 2.7n or
1.8n would also work. Sometimes it is that the ratios are to be kept.
(Dividers, gain of opamps...) It pays to have a critical eye on a
circuit and see if some substitutions can be made, if you're out of stock.
> i can always locate ics
> as need comes up, but its annoying to have to order five caps just to
> put together a simple oscillator or something. i'm also wondering how
> badly an oscilliscope is nessecary... oh and what functions i need in
> a VOM. any input would be great!
Oscilloscope is a must IMO. In a pinch use a "softscope" using your
soundcard. Adjustment of an oscillators waveform for example, would be
possible with this alone, but finding RF dirt or anything which calls
for DC coupling doesn't work with this.
One thing I wouldn't want to miss on my scope is its component tester.
It gives you the V vs. I characteristic of a device, so with one look
you know if something is a diode, zener, transistor, FET and so on...
And if it works or not. Other than that, XY-Mode, dual channel, 20MHz
(or more) would be minimum requirements IMO.
My mostly used VOM is a el-cheapo one. I only use it for measuring
voltages, currents and resistors anyway. AC is going to the scope (and I
prefer to use a scope anyway). Frequency measuring features on VOMs are
often very "interestingly spec'd". Like 0.5% accuracy.... Nothing to
invest money into IMO. (I use a Philips PM6667 7-digit counter for
that.) hFE and diode tests are "go no-go" indicators mostly, mere
add-ons that you'll find on practically all VOMs today, not accurate
enough for serious device characterisation, though handy at times.
Cheers,
René
--
uzs159 at uni-bonn.de
http://www.uni-bonn.de/~uzs159
More information about the Synth-diy
mailing list