--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To post to this group, send an email to neoni...@googlegroups.com.
To unsubscribe from this group, send email to neonixie-l+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/neonixie-l?hl=en-GB.
The rings are quite fickle
-joe
Hi Robert,
I am currently in dialogue with an Australian person, who is on this
list and who has successfully made a ring counter clock using IN-3s.
His information and help is proving very useful to me. The clock is
based largely on Pieter-Tjerk de Boer's clock, but uses transistors
to drive the nixies rather than the opto-coupling system used by Pieter:
http://wwwhome.cs.utwente.nl/~ptdeboer/ham/neonclock/
Michael has kindly documented his project and has a web page at:
http://members.iinet.net.au/~mminns/NeonBulbClock/NeonBulb.html
The IN-3s are a good choice in that they have a wide striking to
maintaining voltage. This voltage difference is reasonably constant,
although it varies up and down the voltage scale from neon to neon
and thus the neons do need adequate burning in and careful matching
into sets. Michael has an Excel spreasheet documenting his
measurements for the neons, available from the website above.
The neons are slightly photo-sensitive in themselves and change
depending on the ambient light levels. Bear this in mind and try to
batch match the neons under constant lighting conditions. They also
fall foul of the problem that they will not light in the dark, again
imposing a level of complexity. Pieter-Tjerk de Boer describes this
issue on his own website.
How old-school you want to be depends on you. Pieter-Tjerk de Boer
has been very purist in his design. I am going the other way for
mine. PCBs, maybe even some surface mount stuff. Unsure if I am going
to use a real time clock and microcontroller yet, but I might, to set
the clock, sync up the rings and provide battery backup.
Be sure to buy at least four times the number of neons than you will
actually use in one clock, so you have plenty of choice.
John S
{ http://www.dos4ever.com/ring/ring.html ]
[ http://wwwhome.cs.utwente.nl/~ptdeboer/ham/neonclock/ ]
Look also for a (pdf copy of the book of J.B. Dance (M.Sc., B,Sc.):
'Electronic Counting Circuits'.
eric
If you wanted to do some real serious play with an xacto knife and some wire,
you could rewire the switches on the board to allow you to set it w/o the
processor and then just let it run with no processor.
Maybe on the next run of boards I'll consider making the jumpable. I kind of
remember that was much easier to say than do :)
-joe
I have the same feeling in developing my clock. I first wanted to use only
realy's in the conters, utill I discovered that the number of swichings
should be far beyound the life expectance of the relay's. Now I think only
the hours and minutes counters with relays and the rest by dekatrons (I have
some 12 counters!) Hopefully they have a long enough life.
Your idea to use a µP as a back-up realtime clock sounds fine, however My
main clock is a verry stable TCXO of 4 MHz that I have to devide to quater
seconds. The first stage has to be a TTL or CMOS divider, followed by a
Elesta EZ10B and some Sylvania 6879's.
Then a GS10B and a GZ12D for the seconds and the same for the minutes.
It should be nice to get soem more information about your µP setup as time
correcting element.
eric
-----Original Message-----
From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On
http://www.sgitheach.org.uk/nixie3.html
As a clock it worked well as long as it was kept in the light. Left in
total darkness overnight it would be 50:50 what you would find in the
morning. I ran it under a small UV lamp and it survived every night
(leds are not allowed). Now its just a demonstration clock. I had
started with two electrode neon bulb rings but found them more unreliable.
I recently bought a bunch of Z700U trigger tubes which have a keep alive
electrode for total darkness operation. I've laid out (in Eagle) a
mains to 1 Hz divider card but not had time to brew and test yet. I
bought the trigger tubes from
But I don't know what his stock levels are like.
BTW if anyone in the group has a few Z700W (dual trigger) trigger tubes
I'd be very interested in a few
Grahame
> Thank you! I really had to swallow hard to put the processor on the
> clock, but
> it was needed. It also added a little eye candy with the quarter
> hour marker
> lamps as well as being able to add in a rtc so that it could
> survive power
> interruptions.
Joe,
Can you explain why you needed to have a three transistor stage
trigger between the later stages - it seems like overkill. Was it
really necessary to condition the signal between rings in this way?
If so, why?
John S
I would have top go back and play with it once again (this is where better
note keeping on my part would've paid off) to be sure, but what I remember is
that I would get fluctuations and noise on the signal feeding the stages and
with the simple 1 transistor circuit, the rings were getting triggered. These
signals would stay high for up to 10 mins depending on the ring. I put in the
extra stage to make the actual trigger for the ring a short pulse. This made
it significantly less prone to fire on signal variations. It may be a tad bit
overkill but I have found it to be reliable.
-joe
--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/a42aeaa9-12cc-4823-b479-655c19aa1140%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.