ITS1B Clock

[Arne Rossius]

Hi,

I saw some talk about the ITS1B thyratron indicator in the archives, so
I thought I'd share my clock design I did a few months back. I got the
tubes from ebay for about 15 Eur each (it looks like they have already
doubled in price by now).

There is a russian datasheet available for these tubes at
http://www.tube-tester.com/sites/nixie/dat_arch/russian-book-0001.djvu
(starting at the bottom of PDF page 211, original page 210). After
having some fun with Google Translate's on-screen keyboard (I don't
speak russian), I figured out that these voltages are required:

1st anode: 40V ± 4V
2nd anode: 100V ± 10V
under-cathode: -250V ± 15V

I also found that the current on the first anode is negative, so
clamping it to ground with a 39V zener diode is sufficient.

To reset the fired segments, the 100V power supply has to be
disconnected for at least 500 us (I used 1 ms). The shortest pulse to
fire a segment (on the 5V CMOS compliant grid inputs) has to be 100 us
(I used 1 ms as well).

As expected, the power supply was the most difficult part of the design.
I used the UC2843 fixed-frequency (PWM) controller IC for the 12V to
100V step-up and connected a voltage multiplier circuit to to switch pin
to generate -300V (since the main output is 100V, it is impossible to
generate the required -250V directly through a multiplier) and used a
resistor + zener diode circuit to drop the voltage down to about 250V.
The current draw on the under-cathode is pretty constant, so just a
resistor with a carefully chosen value probably would have been good
enough, but I wanted to be on the safe side.

Due to the load on the switch pin, the power supply was very unstable,
especially when only drawing current on the -300V output. I finally
managed to get it stable in discontinuous mode, when in continuous mode
there would always be subharmonic oscillation (a few short pulses
followed by a long pulse or vice versa on the gate drive output, causing
audible noise). I fiddled with the operating frequency until I managed
to get the full current without the power supply entering continuous
mode at about 30 kHz. The timing capacitor value is rather critical, if
you use my design make sure to use a low tolerance cap here (definitely
not ceramic) and measure the frequency at the gate drive output. Also
use a beefy enough inductor (my first try with an axial resistor-style
1.5mH, 6.5R inductor failed miserably).

I uploaded my files here:
http://elektronik-kompendium.de/public/arnerossius/temp/its1b/
It's all there: schematic, PCB layout (both as PNG and EAGLE 4.x files),
a scan of the partially translated datasheet, the code for the AVR
microcontroller, some photos and oscilloscope screenshots (in the "oszi"
directory, ch1 = switch pin, ch2 = -300V output).

The AVR program is for use with the German DCF77 radio time transmitter,
but it should be easy to modify the program to be used as a
quartz-controlled clock (using PD5 and PD6 for the set buttons) or even
write a new one from scratch.


Best Regards,
Arne

[Joseph Bento]

Your clock is the first practical use I've seen for these tubes.  I bought a set of six years ago, if nothing else for their uniqueness. 

I would be most curious to know what their intended purpose was - why would they design a display tube that had such odd power requirements.  Perhaps on the plus side, is it's a high-voltage tube that can be controlled directly with TTL logic (if I understand other references I've read).

Joe, N6DGY

[Arne Rossius]

Hi,

Joseph Bento wrote:
> Your clock is the first practical use I've seen for these tubes.

There is one other guy who built a clock, but as far as I know there are
no details or schematics available. Here is his video:
http://www.youtube.com/watch?v=ULKdAg0mLhQ

> I would be most curious to know what their intended purpose was - why
> would they design a display tube that had such odd power requirements.

That has be baffled too. Mine even have a datecode from the year 1990,
so regular 7-segment LED displays were already available for quite a while.

> Perhaps on the plus side, is it's a high-voltage tube that can be
> controlled directly with TTL logic (if I understand other references
> I've read).

Yes, that's correct (although I'm not sure about actual TTL levels, i.e.
2V = high). It certainly works great with 0V/5V CMOS levels.
There is also the latching feature of the thyratrons, so you get the
advantage of few pins needed for a large display without the
disadvantages of actually multiplexing the tubes (flicker, reduced
brightness, increased CPU load).


Best Regards,
Arne

[Dekatron42]

Nice clock! Thanks for sharing!

I am building a similar clock myself right now and I have read a lot of what the Russian people write about these tubes and there are some small design ideas/changes that they have come up with that differs from your design - I have incorporated these in my design as the Russians say that they are essential to long life of the ITS1A/B's. I don't know if the Russians are right on this or not but they all seem to use the same design.

What they have done is adding a resistor in series with each of the second anodes (Anode-2 @ pin 14) as a current limiter. I have seen values ranging from 2.2 k up to 47 k in different designs, but with my ITS1B's a value below 10 k works best when the MPSA92 is supplied with +90V. They also put a 56k resistor between the first and the second anode (Anode-1 @ pin 2 and Anode-2 @ pin 14) as that stabilizes the zener voltage and the circuit.

I am in no way an expert on these tubes as I have only been fiddling around with them for a few weeks since I bought mine, but I think that the designs that the Russians are using are from various books and datasheets that they have access too, at least that is what I have gleaned from what they have written.

My design will also include a driving stage (MPSA42/92/zener) for each digit as that will let me control and extinguish just the digit that I am interested in changing instead of all at the same time (it will add some more components but I think it is fun to solder).

I am also making a switching power supply with a transformer. At the moment I am experimenting with winding the second coil on top of the main inductor to see if it can decrease the audible noise from the switcher.

/Martin

[Terry Kennedy]

On Saturday, June 21, 2014 11:18:58 AM UTC-4, Joseph Bento wrote:

I would be most curious to know what their intended purpose was - why would they design a display tube that had such odd power requirements. 

I've seen a low-resolution photograph of a Soviet submarine console that appeared to have displays made up of these tubes. That would make some sense as one of the last places oddball display technologies persisted in the West was for things like avionics (Minitrons), etc.

Apparently, newer sub designs went with custom displays like this: http://www.tmk.com/transient/vfddisplspecialbigsubmarine01.jpg  I also have a 512 x 512 pixel panel from another sub.

[Charles MacDonald]

On 14-06-21 11:38 AM, Arne Rossius wrote:

> There is also the latching feature of the thyratrons, so you get the
> advantage of few pins needed for a large display without the
> disadvantages of actually multiplexing the tubes (flicker, reduced
> brightness, increased CPU load).

That could also be an advantage if being used in combination with short
wave communications. you would not get multiplex harmonics interfering
with weak stations.


--
Charles MacDonald Stittsville Ontario
cm...@zeusprune.ca Just Beyond the Fringe
http://Charles.MacDonald.org/tubes
No Microsoft Products were used in sending this e-mail.