My first GSA10G dekatron driving a GR10K Nixie!

[Dekatron42]

I just finished my first GSA10G tester which is connected to a GR10K Nixie. These Dekatrons are really finicky when it comes to driving them. There are errors in the datasheet schematics and component values, but the book "Cold Cathode Tube Circuit Design" by D.M. Neale has described the necessary calulations for the component values on page 229-237. I also found references for the German Z572S to be helpful, those Dekatrons are even more finicky in their driving requirements. I accidentally forgot to include the Anode resistor when I hooked up the first Z572S and it promptly lit up like a Christmas tree! It got very warm from just a few seconds with full power, every wire inside the Z572S was glowing like it was a neon lamp, fortunately it wasn't destroyed!

 

From the Youtube page: "GSA10G Dekatron driving GR10K Nixie counting! Dekatron driving Nixie directly via auxilliary anodes. Sine wave drive from oscillator" : http://www.youtube.com/watch?v=DAnyKhPUpgk

[threeneurons]

Nice ! I was starting to think those were "vapor ware". Good to see one come to life !

[Dekatron42]

Thanks! Yes, I was a little bit afraid they wouldn't come to life.

 

Today I tested a bunch of Z572S's with equally good results as with the GSA10G's when using the same resistor values that I found in the D.M. Neale book. I also tested all of my GSA10G's, both the types with blue print and no "fins" between the auxilliary anodes and the older engineering samples with the paper lable with serial numbers and they all work very well. I also found that the Z572S's comes in two different styles, one with straight auxilliary anodes and one with bent auxilliary anodes - I hadn't spotted that difference earlier. Now I'll have to modify the tester so I can test my GCA10G's too, they have a very different pin out compared to the GSA10G which is a shame as they can't substitute each other straight away.

 

Finally my GSA10G clock has gotten one more step closer to being realised!

[Dekatron42]

I shot a new movie today with an improved driving stage with transistors driven by a square wave generator. I could run the GSA10G up to 2KHz with this stage, higher speeds can be attained with higher voltage on the guide pulses if necessary. Here's the clip: http://www.youtube.com/watch?v=bwfLZyhhRzE

 

/Martin

[Dekatron42]

A last movei before bedtime: http://www.youtube.com/watch?v=owPnu6B-xHg I used the GSA10G to drive a GR10J large Nixie at just 2mA anode current for the Nixie as that is the limit in the datasheet for the GSA10G. All figures lit up beautifully at 2mA anode current but by adjusting each cathode with a small resistor the current could be controlled better so that each figure get exactly 2mA, now the average was 2mA but some figures consumed a lot less, close to 1.5mA which made them glow less than others. The GSA10G was made with the GR10K & GR10H in mind which erquire up to 2mA anode current whereas the GR10J is made for up to 4mA anode current. I had to raise the voltage to 500V on the GSA10G to make sure that it could provide enough current and the correct voltage level for the GR10J to light up, at 475V thew figures were not fully covered with glow even though I decreased the anode resistor for the GR10J, the bias network for the cathodes (auxilliary anodes) could possibly have been adjusted too get the same effect.

 

I am very pleased the GSA10G works with the GR10J as I can use them in my clock even though adding an extra current limiting resistor for each cathode will make it trickier to wire it up and quite possibly also tricker to adjust the glow on each figure. 

 

I also tested the speed of the GSA10G with the GR10K and by raising the guide pulse voltage and guide bias voltage I could get it to count up to 8KHz, a bit faster than the specified 5KHz but I might have gotten a bit outside of the specifications of the voltage limits for the pulse and bias voltages on the guides. I had to raise the pulse voltage to 225V(normally 110V-150V) and the bias voltage to 100V (normally 60+/-5V).

 

Finally something is going my way! Now I'll just have to learn to use the cold cathode trigger tubes for the coupling stages and reset stages for the GSA10G so I can make the stages for seconds, minutes and hours. Making these stages with transistors is no problem but it will be my first try with cold cathode trigger tubes for those stages.

 

/Martin

 

[Grahame Marsh]

Martin

Looks good - I'm envious of your stash of dekatrons but it is so good to see them being used.

Which trigger tube/circuit are you going to use for the interstage coupler? In experiments I have reproduced the standard GTE175M and Z700U stage couplers, but you're using what must have been the last development step in dekatrons before the transistor took over.

Look forward to seeing more results :)

Grahame

--
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.
To view this discussion on the web, visit https://groups.google.com/d/msg/neonixie-l/-/t4BAzep5PZcJ.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Dekatron42]

Hi Grahame,

 

I will use the Z865W and its counterpart GT21. They are low trigger voltage cold cathode trigger tubes that were designed partly with this kind of Dekatrons in mind as these Dekatrons only produce pulses of 6-8V over the cathode resistor. In a circuit where you want to be able to reset the Dekatron to any figure you want you will loose the forward voltage drop of one diode, so you'll end up maybe 0.6-0.7V lower. You need this diode in the circuit as the GT21 can only withstand a maximum of -80V on the trigger electrode whereas the negative pulse to reset a Dekatron needs to be aminimum of -110V. To my knowledge there are no other cold cathode trigger tubes that can trigger consistently at these low voltages, usually somewhere above 20-30V is necessary to trigger other types.

 

Tubedata for the GT21 can be found here: http://tubedata.tubes.se/sheets/174/g/GT21.pdf and for the Z865W here: http://tubedata.tubes.se/sheets/182/z/Z865W.pdf

 

As with many cold cathode trigger tubes you bias these so they are in the cut-off region and then let the positive cathode pulse put them into the turn-on region. With the GT21/Z865W this region is between -10V and 0V. Typical circuits show that a negative bias of -8V works fine if you do not have the diode in the circuit for reset purposes, if you have the diode you'll have to raise that voltage to maybe -7V or -6V to have consistent triggering, but not much more as you then enter the high slope portion where the valve will possibly trigger by itself. I believe I will have a lot of testing in front of me with these cold cathode trigger tubes to get them to work properly in this circuit. I haven't seen any real designs with them, usually normal valves or transistors were used.

 

I've made a lot of simulations in LTSpice with these circuits (without any model for the GT21/Z865W) to check what voltage levels I need. It turns out that to get a proper circuit working I will probably have to use a lot of trim potentiometers to adjust the different bias voltages so that each cold cathode trigger tube triggers properly. The span in trigger voltage might make this a necessity even though the span is quite small but I do no tknow how age has affected these trigger tubes, there is a primer electrode so I hope that the design is so good that age will not make too much difference. I'll also need other negative bias voltages for the logic circuits for the alarm I want to include in the design. I just hope I don't run out of these trigger tubes before my ideas for the design are finished. I might skip using these trigger tubes in the mains operated 1Hz generator and use standard GTE175M there to save some GT21's.

 

/Martin