On Monday, August 24, 2015 at 9:35:17 AM UTC-4, Hugh Pryor wrote: > > Thanks for the replies - that's all useful - particularly about the > problem with multiplexing and the kit available. > I enjoyed reading Terry's Electronika clock restoration pages. > > I would be interested to know what components were used in the kit - would > they use the tubes with the joined together spots like in the Electronika > clock? >
Unlikely. The videos posted for the kit show things like "slot machine" effects. So all of the dots need to be individually controllable > Is the problem with multiplexing to do with requiring a pulsed current > through the tubes? Would that shorten the life span? > No; the data sheets provide values for both steady-state current and pulsed current. The multiplexing issue is like this (perhaps someone else can do a better job explaining it): In a tube with a control grid, you can prevent the tube from displaying anything simply by controlling the grid. So you connect the dots on each tube together (bottom dot on each tube is one connection, and so on). You need 7 pins for that (one for each dot on the tube). On a 4-digit clock made up of 4 tubes for each digit, you need 16 pins for the control grids (one on each tube). You then enable the control grid on one tube, light the dots you want, switch to the next tube's control grid, light those dots, and repeat. Other combinations are possible (for example, tie the 2 hours tubes together and the 2 minutes tubes together and use 14 drivers). Since these tubes don't have a control grid, you end up needing 112 pins (7 dots per tube x 4 tubes/digit x 4 digits). This gets rapidly out of hand for larger displays (336 pins for a 4-digit 12-tube Elektronica). That's why the Elektronika "cheats" and ties various dots together inside the tube envelope. There are VFD driver chips from Supertex that have very large numbers of pins available to drive tubes, but once you get to a certain point you're dealing with fine-pitch surface-mounted components which is probably not what you want to start with for your first design. I was talking to someone who suggested that I used relays to provide the > current and shift registers to control the illumination of the dots. The > relays would make a noise which would make it more interesting, and > possibly quite annoying depending on how much noise. > For something that changes once per minute with no transition effect, people near the clock will probably be able to remain sane 8-} If you're going to have a seconds display, slot machine or other transition effects, etc. it will likely get annoying very fast. Plus, even with microminiature relays, this will easily be the biggest current consumption in the whole project. By the way, I once toured the relay room (pretty much the whole floor, actually) of the "zipper" on the Allied Chemical Building at 1 Times Square. This was long before it was modernized. VERY loud. Would it be possible to vary the illumination of the dots for greyscale > values? If so would it be with pulsed width modulation (PWM), or > controlling the voltage? > From what I understand it's technically quite tricky even with LEDs > because our perception of brightness is logarithmic. However a crude 2-bit > with 4 shades (including off) would improve the appearance of animations. > Yes. It will require some fiddling to get the brightness steps to where you'd like. The Elektronika doesn't use multiplexing, so it just drops the voltage when in dim mode. -- 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 [email protected]. To post to this group, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/b129efdd-40d7-48db-9cfb-d2dc5b17ef7a%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
