Russian Polyatron A-201 design and more information!

[Dekatron42]

Hi all,

Please remember that there might be errors in the circuit diagram I publish here, I have checked it a few times but I might have missed something anyway! Please also remember that most of the capacitors found around the Nixie and Polyatron need a voltage rating of at least 500V DC!

Please also remember that my mothers tongue is not English so it might read/sound awkward when you read what I have written below.

Now I have come as far as I feel necessary to publish my design for the A-201 Polyatron for everyone to work with and use to build clocks or whatever is your goal.

The design is in no way finished, but it works well enough for my purposes and I believe it is correct.

Unfortunately I have not been able to run the A-201 Polyatron fully according to the datasheets I have found - this is certainly true when it comes to the guide bias voltage which is stated as -45V in the datasheets (some state it as being "45V less", but it does not say less than what reference) but I never got the A-201 Polyatron to work with a negative guide bias voltage. Maybe I am doing something completely wrong here, but it is now up to you to find out what I might have missed, or to find out if I am correct. As far as I have understood the datasheets this -45V is referenced to the cathode, but it can be referenced to the anodes in which case I think my design is correct.

Over the course of the last week since I started to test my A-201's I have made very litle changes to the original design I hooked up, which was based partly on the A-201 Polyatron design and the ETL GSA10G design, and actually almost gone back to where I started as that was the best working design for my case.

I have used a burroughs B-5092 Nixie as I did not have any Russian Nixie at hand.

If you look at the circuit diagram you see that there are a bunch of 100K resistors connected together to one 100K resistor which in turn is connected to the Anode voltage of the Nixie. These resistors are crucial to the design and will have to be changed if you use other Nixies! These resistors are there to put all of the digits not lit in their extinguished state - if you remove these resistors you will see a haze around the digits not lit, they are also necessary for the A-201 Polyatron to work correctly as it will otherwise step/count incorrectly.

The original Polyatron design did not use the 100K resistor connected to the Anode voltage but rather connected all of the 100K resistors directly to the Anode voltage, but this did not work with the B-5092 - neither did it work to leave all these 100K resistors floating as other Polyatron designs show, so this depends on what Nixie you use - below you will see a Russian design which leaves them floating when connected to IN-8-2 Nixies.

One more reason why it does not work to leave them floating or connected to the Anode voltage is the same as for the GSA10G designs, the voltage on the Nixie cathodes must be in the region where it is in the cut off region, so no number will be lit when this voltage is present - it is up to the polyatron to lower the voltage enough for a digit to light up, not because of the resulting voltage from the resistors.

The 22nF capacitors from all digits to ground are necessary to minimise the variation of the load on the A-201 Polyatron - this explanation was found in the paper "A new poly-anode counting tube, the 'Polyatron'" by Y. Hatta ; H. Mase ; M. Sugawara published in Radio and Electronic Engineer, Volume 26, issue5, November 1963, p. 383 -- 388 . My first tries at getting the A-201 Polyatron up and running were based on the GSA10G design which did not have these capacitors, but when I saw the Polyatron design I just took the capacitors I had at hand which had a voltage rating above 500V and put them in and it started to work, I have a few hundred of the 22nF/630V capacitors in a bag and it was very fortunate that I used these as the same value was used in the Russian design below.

It is also necessary to connect the "screen" electrode of the A-201 Polyatron via a 100k resistor to the cathode of the A-201 Polyatron and then have the current limiting resistor connected from the cathode to ground. I have seen people mention that they use a much higher resistance and connect it to ground, this will have almost the same effect as using a lower resistance connected to the cathode but the most important difference is what you reference the screen electrode to. this is very important as the cathode voltage chnages depending on what speed you let the Polyatron run at, in my tests it changed some 40V. Since the screen electrode is positioned between the cathode and the anodes and guide electrodes it will have an effect on the electric field in the A-201 Polyatron and if you reference it to ground it will affect the counting adversely but if you reference it to the cathode it will follow the cathode voltage and not affect the counting adversely.

To get a carry signal you just connect a 22nF capacitor to the desired digit and connect the other side via for instance a 47K resistor to ground. You will then see negative pulses across this resistor of approximately 40-80V depending on the rest of your circuit. You might have to change these component values a bit depending on what Nixie and voltages you are using.

To set/reset the Polyatron it needs a positive pulse of at least 190V for 300 microseconds or more depedning on the curve form of the set/reset pulse. This positive pulse is then connected to a 470K resistor to ground and a 22nF capacitor to the digit you want to reset to - remember that if you reset to the same figure that you have connected as carry out you will get a much higher pulse on the carry output than for the usual situation when a digit is turned on by the Polyatron so you must protect your carry coupling stage for this! You can use a diode or two to protect against these positive spikes, but you will also have to take into account the effect this has on the reset pulse timing.

Also remember that the set/reset and carry circuits affect the counting speed as they load the circuit. In some circumstances you will also need a series resistor in the set7reset circuit to limit the current through the A-201 Polyatron, but for my tests I left it out of the circuit.

When the A-201 Polyatron is running in this design all of the guide electrodes as well as the rings connecting them light up during stepping, and as far as I can understand this is normal behaviour, just like when you run any other dekatron at high speed like for instance the GC10D at high speed, then all of the guide electrodes also light up.

Yesterday when I had some time left I decided to see what Google and Google Translate could do for me so I started to find out the Russian word for Polyatron and Dekatron and then started to search for more information. First I found a few old Russian books on dekatrons and stuff that I already had, but then I started to find Russian discussion forums where the A-201 Polyatron was mentioned. Suddenly I found the alias of one person "SLvik" which reminded me about a Dekatron clock he built a few years ago, and I started to read what he spoke about as he mentioned the A-201 Polyatron in one forum. Then it started to roll along really well and I found all of these references:

Russian patent on improving the counting for Polyatrons.

http://www.findpatent.ru/patent/60/609215.html

Russian patent for a combined Polyatron and Nixie (much like another patent I posted here last summer)

http://www.findpatent.ru/patent/27/278888.html

Then I stumbled upon the first schematic for the Polyatron that I have seen:

http://bar-snab.ru/poleznoe/47/1021-2-3-indikaciya-desyatichnyx-schetchikov.html

and then I found a few pictures of an actual A-201 Polyatron design:

http://radiokot.ru/forum/viewtopic.php?f=3&t=3210&start=11560 (scroll down a bit)

and then also the circuit diagram for the pictures above:

http://radiokot.ru/forum/viewtopic.php?f=3&t=3210&start=11580

All of these sources more or less confirmed my design, of course except where I had to modify it to work properly with the Burroughs B-5092 Nixie.

I have left out all timing capacitors on purpose as I want something to be up to you to figure out - I can tell you that the capacitor in the oscillator works from 330nF up to 33uF and the timing and delay capacitors I used ranged from 100nF up to 330nF. You will have to make sure that the pulses are within the times stated in the datasheet to get a correct counting sequence, the maximum counting frequency with a Nixie connected is 5KHz and without a Nixie 10KHz. The A-201 Polyatron datasheet mentions pulse widths of between 30 and 170 microseconds but it does not mention the overlap necessary for correct counting. There are unfortunately a few different datasheets available on the A-201 Polyatron but none of them mentions what differs between these different A-201 Polyatrons. Voltages as well as currents and timing parameters differ between these datasheets. as seen in the Russian design above you can drive the A-201 Polyatron with a differentiating network with just one pulse (the circuit diagram and PCB are missing the capacitor connected across the 300K resistor which is part of this differentiating network).

Well, that is all I can think of right now. Please be careful when experimenting with these high voltages and also to protect your rare A-201 Polyatrons - but most of all have fun!

I would also like to thank everyone who has helped me with material, ideas, translations and as a sounding board for my ideas!

Please be gentle if you find anything wrong, I am more than willing to discuss it and correct any mistakes, but I burn easily from being flamed! ;)

/Martin

[Dekatron42]

Hi all,

 

I saw that I hadn't put all measured voltages on the A-201 Polyatron in the schematic diagram so I add them here until I have made a new schematic diagram with them.

 

*. The voltage directly on the screen electrode is approximately +156V (pin 9)

 

*. The voltage for an auxilliary anode on the A-201 Polyatron when the Nixie digit is lit is +292V

 

*. The voltage for an auxilliary anode on the A-201 Polyatron when the Nixie digit is switched off is between +360V and +390V - this voltage varies depending on which digit is lit and how the 100K resistors work as a voltage divider, and also depends on which digit you measure the voltage on.

 

*. All resistors used are metal film resistors from Vishay with a oltage rating of 500V where necessary, otherwise the voltage rating is 350V.

 

*. Diodes marked 7' are 1N4007, transistors marked A42 are MPSA42.

 

I hope this extra information makes it easier for anyone who wants to build the design to check the circuit if something doesn't work.

 

/Martin

[Marcin]

Oh boy. It is impressing. Are you going to cascade them too?
Thanks a lot for sharing your results.
Marcin (in awe)

On 22-Mar-13 00:04, Dekatron42 wrote:
> Hi all,
>
> Please remember that there might be errors in the circuit diagram I
> publish here, I have checked it a few times but I might have missed
> something anyway! Please also remember that most of the capacitors found
> around the Nixie and Polyatron need a voltage rating of at least 500V DC!
>
> Please also remember that my mothers tongue is not English so it might
> read/sound awkward when you read what I have written below.

> ....

[Dekatron42]

Hi Marcin,

 

Yes, I'll cascade them too but right now I have no more space on my prototype board so I'll have to arrange it all in a different way, maybe hook up two nixies and two Polyatrons on a separate piece of PCB and only the logic on the prototype board.

 

I guess that people who owns Polyatrons are scrambling to show the first clock built with these and I also guess that the price for Polyatrons will multiply by 10 or even 100 now that they can be used.

 

I'll also do some more experimenting with the setup I have before I dismantle it and build something else. I want to learn more on how all of the voltages on the different electrodes of the Polyatron relate to each other and also to make a differentiating driving stage that works properly. The cascading and set/reset stages are easy compared to fully understand the timing and voltage issues properly (at least to me they are easy).

 

Cheers,

 

Martin

[Jon]

Amazing work Martin !

 

Jon.

 

 

[Dekatron42]

Thanks Jon!

I've just updated the schematic with the extra information in my post above and included the new version here.

/Martin