I that 555 supply is to power 6 IN-18s direct drive, then it ain't
gonna cut it. IN-18s need at least 4mA, minimum. X6 that's 24mA.
You're really pushing the 555 designs. Yes, the MAX1771 can be
unstable. If you're going to have a PCB etched, then just copy how
nick did his, and life should be okay. There is always the middle
option, of using the MC34063. Just make sure you use the design that
has a pull-down transistor (MK1.5), or push-pull transistor pair (MKII
two). Just having a 330 ohm resistor trying, to dump the FETs charge
(MKI one), will give poorer performance than the 555. Efficiency,
determines which is best. A 555 will get you 65% at best. That's 35%
going up in heat, or 2.33W of heat. With a MK1.5 or MKII design,
efficiency can be as good as 80% (75% typical), so that will be 1.44W
of waste heat (assuming 75%). The MAX1771 should give you at least 85%
efficiency, or 0.76W of waste heat. This is all base on 4.32W going to
your nixies (and anode resistors).

How bright those tubes need to be depend on where this thing will be
sitting. If its in the TV room, or any other room in your house, other
than the kitchen, or workshop, then the ambient light will be pretty
low. I have 1x4 mux'd clocks, and all of those are still pretty
bright, in a normal home environment. Maybe too bright. Best thing to
do is to light up a single tube and stick in the room that's intended
as the home for this clock. If you want to be thorough, pulse it on
and off at varying duty cycles, to mimic multiplexing.

To avoid flicker, make the full frame rate faster than 60Hz. Maybe a
lot fast if you have young eyes. For example: 1x6 muxing. the full
frame (all 6 tubes) must be accessed in 16.7mS or less. 16.7/6=2.78mS
max per tube. Again you can try this out with a simple pulse circuit
on one tube, and see when your eyeballs no longer can see the flicker.
For my old eyes, that's a month and a half. For most normal people,
that's somewhere over 25Hz (40mS).

Don't run the tubes under their specified minimum current. that's just
as bad as overdriving them. You'll get cathode poisoning over time,
and some numerals won't illuminate fully, maybe in a month or two. You
can adjust brightness, by varying the duty cycle. Say its a 1x6 mux'd
clock, and each tube gets 2.5mS of ON time. During this period, vary
the actual ON time from say 200uS to 2.5mS, in 100uS increments. Let
all tubes stay off for the remainder of that 2.5mS. BTW: nixies are
kinda slow, as display devices go. They take about 50uS to turn on,
after the power is applied. Its not an LED, that can respond in 10s of
nanoseconds.

To see if it will sing, again run an experiment. Pulse each of your
six tubes individually, and see at which frequency, if any, that those
tubes will sing at. Avoid those frequencies, when you pick your
multiplexing frequency.

As you can see a lot of this will be determined by experiment, which
is how it should be. If there's one thing I keep ranting about, is
don't take anything as dogma. Try it out, and see what works for you.

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To post to this group, send an email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/neonixie-l?hl=en-GB.

Reply via email to