On 09/11/2014 12:15 PM, David Forbes wrote:
The tubes are 'on' for less time, but the current needs to be somewhat higher for a similar apparent brightness. The tube life is
proportional to something like i^-3 for higher currents. With a 1/6 duty cycle multiplexed clock running at twice the current of
an always-on clock, the tube life will be similar but the display a bit less bright appearing.
I must have come across that or other research before, because I have come to a similar conclusion, which is: When pulsing the
tube's voltage (either by PWM or multiplexing), as long as you do not increase the current above the direct drive current
specification, then tube lifetime increases. If you increase the current to compensate, then the lifetime benefit decreases before
you reach the original direct drive brightness. I wonder if the strike activity is more stressful than the sustain activity.
Also, pulsing high voltage can rattle loose lightweight metal inside of the tube. Sometimes, the frequency can be heard, as if the
tube were a speaker. I wonder if this movement can affect the buildup of resistive material (sputtering), and if there were to be
benefits, are there any negative consequences due to the vibrations, such as structural or connection failure.
It looks like this clock was built using a clock kit circuit that can power 2 of the 6 tubes simultaneously, making the duty cycle
at each tube 1/3, or 33%, not including any blanking intervals. Assuming that the picture accurately shows dimly lit segments, it
does not look like the current was increased to compensate. So the tube life should be good. Residential settings are usually
poorly lit, and nixie tubes were built for more brightly lit work environments. So a slight decrease from the rated brightness can
be perfectly acceptable.
Personally, I prefer direct drive with current regulation that decreases the brightness based on a light sensor. If the clock is in
direct sunlight, it needs to be brighter. If the room is dark, the tubes need less current. If the room is completely dark, the
tubes may as well be off. This should, at least, double the tube lifetime in most cases (assuming that you do not leave your lights
on all the time). Eventually, I would like to add room occupancy sensors to turn off the clock when there is no one there to admire
them. For clocks that have battery backup for the clock, you can do that now with a lamp motion sensor, but off-the-shelf devices
tend to have poor sensor coverage due to poor placement of the sensor, or a room layout that needs more than one sensor on a switch.
I wonder if there is a better way of limiting the current without using PWM. Some switching power supplies can adjust their voltage
based on an output from a microcontroller. However, I do not like the idea of a switching power supply failing to regulate the
voltage and it sends more voltage then what the tubes should get. I suppose a metal-oxide varistor (MOV) with the proper rating
could be added to protect the tubes. I suspect that these tubes may outlive the high voltage power supply. Maybe a resistor ladder
network would be better. However, my concern is that when digits that require less current are lit, such as the "1" digit, this
might make all other tubes receive more current and thus brighten the digits that should be static.
--
You received this message because you are subscribed to the Google Groups
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit
https://groups.google.com/d/msgid/neonixie-l/541224B9.70802%40matthewc.net.
For more options, visit https://groups.google.com/d/optout.
