2 Poisoning questions?
119 views
Skip to first unread message
redrok
Apr 2, 2018, 1:38:38 AM4/2/18
to neonixie-l
Hi All;
I experimenting with a 24Hr 6 digit Nixie clock. This will have 56 active digits and be driven with this transistor matrix operating as a current sink.
Question 1:
Of course there can be 60 digits with the 6 Nixies, but only 56 would ever be lit.
Or the any Poisoning problems associated with the 56 digits even though 4 digits will never be active?
Question 2:
This current sink array can only drive 1 digit at a time. So the circuit must be multiplexed between the 6 Nixies and drive the digits each with 6 times the current for 1/6th the time. Do the Nixies have any problems doing this? Shorter life? Increased poisoning perhaps?
BTW, the I'm running MicroMite Basic on a 28 pin PIC32MX170F256B processor. This is a wonderfull micro a so interactive when testing code.
Thanks!
redrok
Paul Andrews
Apr 2, 2018, 7:38:33 AM4/2/18
to neonixie-l
1. People like to run 'anti-poisoning' routines - cycling rapidly through all digits at full brightness. Other techniques include writing blanking routines that turn off the display when there is no one around to look at it. I remain to be convinced that anti-poising routines work.
2. Later tubes were developed specifically with multiplexing in mind - they will produce a good display with temporary higher currents. Some of this has to do with stopping the glow spreading to the support structure, some with the persistence of the glow between pulses and maybe some with the resilience of the cathodes to the higher current. For example I have seen adverts for tubes like the B5750 that explicitly mention how many digits can be multiplexed.
2. Later tubes were developed specifically with multiplexing in mind - they will produce a good display with temporary higher currents. Some of this has to do with stopping the glow spreading to the support structure, some with the persistence of the glow between pulses and maybe some with the resilience of the cathodes to the higher current. For example I have seen adverts for tubes like the B5750 that explicitly mention how many digits can be multiplexed.
I don't thing you will need 6xI for this, but something less.
Which tubes are you thinking of using?
I will try to dig up some links to specific information.
Paul Andrews
Apr 2, 2018, 7:51:04 AM4/2/18
to neonixie-l
redrok
Apr 2, 2018, 2:37:42 PM4/2/18
to neonixie-l
Hi Paul;
Unfortunately the schematic image I posted disappeared. Here's a link to the same Image.
NixieB56_Array_1.jpg a 56 transistor array controlled with 8 lines.
and the simpler
NixieB12_Array_1.jpg a 12 transistor array controlled with 4 lines.
These transistor arrays operate as constant current sinks. I've extensively tested the B12 using both IN-2 and IN-12B Nixies with a power supply up to 350V and a sink current of 1.5mA. They work beautifully.
Of course, the draw back is only 1 transistor at a time can be activated and the micro pins must be reconfigured between 2 digital outputs, one Hi one Low. and the rest as inputs. But micros are fast and do it nicely. I think I can do this with an MCP23008 I2C bus expander. Will try that later.
The question about 6/1 muxing stems from the "only 1 active transistor" thing so 6 sequential digits need about 6 times the current or about 9mA in my case.
The decodesystems reference talks about high mux ratios, 6/1 doesn't seem to hi.
Thanks for those references!
You might ask why I'm doing this? I'm a hard core experimenter and like to push ideas to the limit to understand them.
Unfortunately the schematic image I posted disappeared. Here's a link to the same Image.
NixieB56_Array_1.jpg a 56 transistor array controlled with 8 lines.
and the simpler
NixieB12_Array_1.jpg a 12 transistor array controlled with 4 lines.
These transistor arrays operate as constant current sinks. I've extensively tested the B12 using both IN-2 and IN-12B Nixies with a power supply up to 350V and a sink current of 1.5mA. They work beautifully.
Of course, the draw back is only 1 transistor at a time can be activated and the micro pins must be reconfigured between 2 digital outputs, one Hi one Low. and the rest as inputs. But micros are fast and do it nicely. I think I can do this with an MCP23008 I2C bus expander. Will try that later.
The question about 6/1 muxing stems from the "only 1 active transistor" thing so 6 sequential digits need about 6 times the current or about 9mA in my case.
The decodesystems reference talks about high mux ratios, 6/1 doesn't seem to hi.
Thanks for those references!
You might ask why I'm doing this? I'm a hard core experimenter and like to push ideas to the limit to understand them.
Reply all
Reply to author
Forward
0 new messages