[time-nuts] GPS for Nixie Clock
kb8tq at n1k.org
Sat Jul 16 13:50:25 EDT 2016
The gotcha is that it’s the sum of the radiation arriving in the vicinity of the gas.
Supplying a bit can flood small variations, but they still are present. You are trying
to get what is essentially a neon bulb to trigger accurately to a very tight budget.
There is a lot of prior art on the pitfalls. Since you do not have a structure that was
custom designed as a high frequency transmission line, there are a lot of lumps and
bumps along the way…
The only real point is that going from millisecond to microseconds is a leap for these
gizmos. Going from microseconds to nanoseconds (or picoseconds) is an even bigger leap.
There *is* a way to do it, if your budget it big enough. It’s going to be a major effort all by
it’s self. Being able to forward predict the process (which is the goal) at these levels has
a lot of variables in it. Feedback does not help if you have second to second jitter that
is all over the place….(gas tubes are used as noise generators for a reason ….).
> On Jul 16, 2016, at 11:36 AM, David <davidwhess at gmail.com> wrote:
> The point of measuring the actual ignition point is to predictably
> remove delay by driving the element earlier. CRT grid structures
> support transition times in the 5 to 20 nanosecond range; the smaller
> distances involved with a nixie tube should support faster operation.
> Something which just occurred to me is that ultraviolet can be used to
> provide ionization to the gas without radioactivity. Flame detector
> tubes work like this so bathe the tubes in a small amount of UV. I do
> not know how transparent the nixie tube envelope to UV is though.
> On Sat, 16 Jul 2016 11:04:22 -0400, you wrote:
>> Since we have moved into synchronizing this stuff at the nanosecond level
>> (maybe we are even lower than that by now ..), simply getting a wide band
>> enough signal off of a Nixe socket is going to be interesting. An array of picosecond
>> photo diodes on each tube may be the only way to go. How many channels
>> this all will take depends a bit on how many digits past the second the display
>> will show. Is it 9 digits past the second?
>> Since you will only know the ignition point *after* it has happened, the system
>> only works to a certain degree. Trigger point *is* dependent on the light level.
>> You will need to collect real time data to keep things consistent.
>>> On Jul 16, 2016, at 10:49 AM, David <davidwhess at gmail.com> wrote:
>>> Use AC coupling to each digit to measure the ignition waveform and
>>> detect the breakdown point like with a tunnel diode trigger. Use a
>>> higher compliance voltage and greater negative resistance (constant
>>> current drive?) to lower breakdown jitter.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts