Loved this video, especially as I worked for Packard Instrument Company (I started in 1978). Sadly, the company no longer exists, it was sold several times and I actually work for the company that currently owns them. I don't recall that board though it is a classic Packard design, right down to the color dots on the ICs. Codes indicated internal part number of the chip. They actually paid factory workers to hand paint them on every IC in stock. I worked in several departments, depot repair, field service, design engineering, now IT. The diodes are necessary when you multiplex Numitrons to prevent back-feeding current through the segments causing segments on multiple displays to glow. LEDs don't have this problem as they only conduct current in one direction. I will see if one of my former Packard Alumns recognize it. I am guessing it would have been part of a display for a radioactivity counter.
The numitron clock I built has been running continuously for just over 4 years on my work office desk with no failed segments either. It's got 6 tubes so plenty of chance for failure too. I do have it automatically turn the brightness way down with an LDR when they turn off the lights though so I'm sure that helps.
Fran is absolutely right about the diode blocks. Since the numitron would light up no matter which way the current passed through. The diodes made the tubes unidirectional. Now the display can be matrix scanned just like a modern LED display. This allows 7 wires to control 49 segments. Scan it fast enough and you have a solid 7 digit display.
one of the things I love about your channel is that you show when things go wrong and then talk about why they went wrong and how you fixed it. great to see the process and demystify the realm of making by actually including the troubleshooting portion
The Oklahoma Region of the Sports Car Club of America used a home-brew lap timer system based on Numitrons back in the 70s. On race days, they used a central clock wired to 10 or so displays at observer stations, each one had a handheld switch the observer would click to lock in the time when their car crossed the line. Found the display units at a swap meet and researched the markings. I built a couple of arduino-based clocks out of them and they've been running nonstop for years. I enjoy the "warmth" of the tungsten segments and the way they kinda fade in and out as they change. Numitron calculator would be kinda cool! Maybe I'll try building one with the ones I have left.
I made a Pi powered numitron clock-radio. Numitrons came out of a arcade game of some sort. Pinball perhaps. I drive them at 3V to prolong their life. Also not on during the day, since it's a bedside clock.
Makes me think of the time (1996) a numitron BART ticket vendor machine ate a $5 bill. Enough of the numitron segments were dead so I couldn't see that it didn't register and it just spat the ticket back out with no value increase. This was in 1996 , $5 was worth $10 back then so I went to see the station agent. She gave me a look like something wasn't right with me, but I filled out a form. Several weeks later I got a real 'chain printer printed' check from BART for $5. No hard feelings, it created this very visual memory.
So I burned up 3 IV-15 VFDs today. Fun times. Note to self, although a 1117 may allow 20V max input, on failure you may end up with 20V on those precious filaments. I'm cranky af lol.
Killed an IV-12 tube recently while testing because one of my wonkily attached wires got lose and shorted with the filament. Bit of magic smoke inside the tube and that was that. Very unfortunate.
You can see "Packard Inst Co, Inc." on the board. This is Packard Instruments who made a lot of scientific measuring stuff and I don't think were related at all to Hewlett-Packard or, for that matter, Packard-Bell either. I had seen some items over the years, but don't know the company history, and couldn't find much. But at the Smithsonian "National Museum of National History" found a long listing of products such as, "..."Flame Oxidation Sample Oxidizer Model 307"; "Auto-Gamma 5500 Benchtop Gamma Counters" ; "Packard Pesticide Analyzer"; Tri-Carb Liquid Scintillation Analyzer; Microplate Scintillation and Luminescence Counters." I suspect this same board may have been used in multiple of their products. Just a guess.
This display *might* be from a scintillation counter. In one of the labs I worked int many years ago, we had a 7 digit Packard Inst. counter. IIRC, part of Packard was bought out by Perkin Elmer around 2000.
I have a whole bunch of T3 500w halogen tubes and thought it would be neat to assemble them into 7-segment displays like the numitrons, then build a clock out of them. They have a fantastic glow when ran between 15 - 40 watts, but I probably couldn't afford to keep the clock powered up for any length of time, which is kinda the whole point of a clock. I guess it sounds more practical if you frame it as a space heater that also displays the time.
It’s multiplexed. You can’t multiplex incandescent filaments without diodes to prevent sneak paths (LEDs have the diodes inherently, and they’ll generally block at least 5V). If you want 4V (RMS) across each filament you need to have 1/7 duty cycle of sqrt(7) * 4V peak. If the multiplexing stops the filament life will be impacted negatively. Probably something like a 12V or a bit higher supply. Thanks for an interesting video!
Beautiful displays! I remember when I was young seeing a fuel bowser in my town with Numitron display readout, certainly wasn’t Nixie tube architecture but had wired segments for the digit elements. Wonder where it ended up, scrapped and dumped no doubt.
I was always curious when they 'tin' the traces on the component-side, do they 'mask out' the holes? I can't imagine those poor ladies having to solder-wick all those holes! Talk about cost-of-manufacture!
Nice unit! Packard made many instruments, I've used their Scintillation counters. They were out of La Grange and Downers Grove Illinois and unrelated to HP I believe. Perkin Elmer bought them in 2001 way after they mis-ground the Hubble mirrors...;)
I have several similar old TI calculators, but I don't think I have the exact same one. I used to find them almost every time I went to the thrift store near me.
Wow Fran, I JUST received a 6-digit DR2010 Numitron board that I'm guessing came from a gas pump. And now you've posted a video on Numitrons! This is perfect! Excited to watch!
Would not the obvious solution for infinite Numitron filament life be to adopt the approach in Fran's "Keep Them Bulbs On!!!" video? ie. add a resistor to "trickle" only enough current to each filament, when it is "off", to keep it warm but not quite incandescing?
you can greatly extend the lifespan of any incandescent "thing" by never having the current completely cease through the filament. "Off" current should be enough to keep it warm without making any light. In this way you can cycle the filaments on and off for a much longer time. I'm sure you know already, but just in case...
I'm surprised to see a multiplexed Numitron display. I've been told it's not a good idea because of the thermal lag of the segments - it's going to be dim, and then maybe allright for a clock because they will not be cooked too much running a 1:6 multiplex ratio.
Could you implement a software based always on by essentialy creating a virtual PWM by driving the inputs at a high frequency and varying their duty cycle in each refresh? Or would the BCD decoders not work at high speed?
I don't think that a tungsten element would withstand pulsed current like that over time. It's not like sinusoidal AC.... but always on would imply constant low current to provide longevity.
… I think that the way you’re explaining it, description of what & how it does it, etc., that’s probably how the original designer of the thing thought about it, as he sat down to begin this project.
@@KeritechElectronicsI realy hold my thumbs for a collaboration with Fran in the future, that would be realy awesome! You and Fran would be a perfect match for a collaboration!
@@sheep1ewe I'd absolutely love to do that! It's been my dream for almost as long as I ran my channel. Maybe someday... online or IRL. If I ever go to the States, FranLab will be the #1 on my list of places to visit.
Fran, could the second power rail on the board be designed for a different voltage? Multiplexing the displays would result in much dimmer digits. I think you'd need about 12-15 volts to reach the same filament temperature as a constant 5v would provide.
I have several dual in line displays, in one way I'd like to build something with them, unfortunately as they are now unobtainium, it would be totally unrepairable. So they just sit in a tub.
I have some similar minitron displays salvaged from old aircraft radios. However most are worn out - Fran is right on this one. Can't use them for an always on display like a clock. They also get hot!
No, nixies and numitrons are two fundamentally different animals. Nixies use the glow of ionised neon gas to produce light, while numitrons are just incandescent lightbulbs, with filaments arranged in a 7-segment display fashion.
Loved this video, especially as I worked for Packard Instrument Company (I started in 1978).
Sadly, the company no longer exists, it was sold several times and I actually work for the company that currently owns them.
I don't recall that board though it is a classic Packard design, right down to the color dots on the ICs. Codes indicated internal part number of the chip.
They actually paid factory workers to hand paint them on every IC in stock.
I worked in several departments, depot repair, field service, design engineering, now IT.
The diodes are necessary when you multiplex Numitrons to prevent back-feeding current through the segments causing segments on multiple displays to glow.
LEDs don't have this problem as they only conduct current in one direction.
I will see if one of my former Packard Alumns recognize it. I am guessing it would have been part of a display for a radioactivity counter.
My clock with numitrons has been running for 6 years, no failed segments yet. For tungsten filaments, little bit less voltage means WAY more life.
Yes, but that plus always on adds perhaps an order of magnitude more longevity - ua-cam.com/video/31rUB8njeck/v-deo.html
The numitron clock I built has been running continuously for just over 4 years on my work office desk with no failed segments either. It's got 6 tubes so plenty of chance for failure too. I do have it automatically turn the brightness way down with an LDR when they turn off the lights though so I'm sure that helps.
Fran is absolutely right about the diode blocks. Since the numitron would light up no matter which way the current passed through. The diodes made the tubes unidirectional. Now the display can be matrix scanned just like a modern LED display. This allows 7 wires to control 49 segments. Scan it fast enough and you have a solid 7 digit display.
Also filaments have thermal inertia allowing use of fairly low frequency._
one of the things I love about your channel is that you show when things go wrong and then talk about why they went wrong and how you fixed it. great to see the process and demystify the realm of making by actually including the troubleshooting portion
The Oklahoma Region of the Sports Car Club of America used a home-brew lap timer system based on Numitrons back in the 70s. On race days, they used a central clock wired to 10 or so displays at observer stations, each one had a handheld switch the observer would click to lock in the time when their car crossed the line. Found the display units at a swap meet and researched the markings. I built a couple of arduino-based clocks out of them and they've been running nonstop for years. I enjoy the "warmth" of the tungsten segments and the way they kinda fade in and out as they change. Numitron calculator would be kinda cool! Maybe I'll try building one with the ones I have left.
love seeing the oklahomies in the wild
I made a Pi powered numitron clock-radio. Numitrons came out of a arcade game of some sort. Pinball perhaps. I drive them at 3V to prolong their life. Also not on during the day, since it's a bedside clock.
Makes me think of the time (1996) a numitron BART ticket vendor machine ate a $5 bill. Enough of the numitron segments were dead so I couldn't see that it didn't register and it just spat the ticket back out with no value increase. This was in 1996 , $5 was worth $10 back then so I went to see the station agent. She gave me a look like something wasn't right with me, but I filled out a form. Several weeks later I got a real 'chain printer printed' check from BART for $5. No hard feelings, it created this very visual memory.
So I burned up 3 IV-15 VFDs today. Fun times. Note to self, although a 1117 may allow 20V max input, on failure you may end up with 20V on those precious filaments. I'm cranky af lol.
Killed an IV-12 tube recently while testing because one of my wonkily attached wires got lose and shorted with the filament. Bit of magic smoke inside the tube and that was that. Very unfortunate.
You can see "Packard Inst Co, Inc." on the board. This is Packard Instruments who made a lot of scientific measuring stuff and I don't think were related at all to Hewlett-Packard or, for that matter, Packard-Bell either. I had seen some items over the years, but don't know the company history, and couldn't find much. But at the Smithsonian "National Museum of National History" found a long listing of products such as, "..."Flame Oxidation Sample Oxidizer Model 307"; "Auto-Gamma 5500 Benchtop Gamma Counters" ; "Packard Pesticide Analyzer"; Tri-Carb Liquid Scintillation Analyzer; Microplate Scintillation and Luminescence Counters." I suspect this same board may have been used in multiple of their products. Just a guess.
This display *might* be from a scintillation counter. In one of the labs I worked int many years ago, we had a 7 digit Packard Inst. counter. IIRC, part of Packard was bought out by Perkin Elmer around 2000.
Loved it! I like these displays. Made a digital clock with them back in the 70's. I miss the 70's.
Look at this Mechanism fran.
You can connect your + / - Numitron to the extra transistor circuit...
I have a whole bunch of T3 500w halogen tubes and thought it would be neat to assemble them into 7-segment displays like the numitrons, then build a clock out of them. They have a fantastic glow when ran between 15 - 40 watts, but I probably couldn't afford to keep the clock powered up for any length of time, which is kinda the whole point of a clock. I guess it sounds more practical if you frame it as a space heater that also displays the time.
Thanks for talking me out of making a numitron clock! I like the idea of a calculator, gonna go for it
Do the clock! I've had one running for years with no burnouts
It’s multiplexed. You can’t multiplex incandescent filaments without diodes to prevent sneak paths (LEDs have the diodes inherently, and they’ll generally block at least 5V). If you want 4V (RMS) across each filament you need to have 1/7 duty cycle of sqrt(7) * 4V peak. If the multiplexing stops the filament life will be impacted negatively. Probably something like a 12V or a bit higher supply. Thanks for an interesting video!
Beautiful displays! I remember when I was young seeing a fuel bowser in my town with Numitron display readout, certainly wasn’t Nixie tube architecture but had wired segments for the digit elements. Wonder where it ended up, scrapped and dumped no doubt.
I was always curious when they 'tin' the traces on the component-side, do they 'mask out' the holes? I can't imagine those poor ladies having to solder-wick all those holes! Talk about cost-of-manufacture!
Nice unit! Packard made many instruments, I've used their Scintillation counters. They were out of La Grange and Downers Grove Illinois and unrelated to HP I believe. Perkin Elmer bought them in 2001 way after they mis-ground the Hubble mirrors...;)
I have several similar old TI calculators, but I don't think I have the exact same one. I used to find them almost every time I went to the thrift store near me.
Activa los subtítulos para otros países,gracias.
Gorgeous!!! Make it say FrAnLAb!!!!
Wow Fran, I JUST received a 6-digit DR2010 Numitron board that I'm guessing came from a gas pump. And now you've posted a video on Numitrons! This is perfect! Excited to watch!
Would not the obvious solution for infinite Numitron filament life be to adopt the approach in Fran's "Keep Them Bulbs On!!!" video? ie. add a resistor to "trickle" only enough current to each filament, when it is "off", to keep it warm but not quite incandescing?
you can greatly extend the lifespan of any incandescent "thing" by never having the current completely cease through the filament. "Off" current should be enough to keep it warm without making any light. In this way you can cycle the filaments on and off for a much longer time. I'm sure you know already, but just in case...
As I said... ua-cam.com/video/31rUB8njeck/v-deo.html
I have some nixie tubes that I want to build some project with too and a clock has been done so much!
5:20 "Nibble" was a magazine for Apple computers in the early 1980s.
I'm surprised to see a multiplexed Numitron display. I've been told it's not a good idea because of the thermal lag of the segments - it's going to be dim, and then maybe allright for a clock because they will not be cooked too much running a 1:6 multiplex ratio.
I'm happy to see you got it working, really pleasing displays, love them
Could you implement a software based always on by essentialy creating a virtual PWM by driving the inputs at a high frequency and varying their duty cycle in each refresh? Or would the BCD decoders not work at high speed?
I don't think that a tungsten element would withstand pulsed current like that over time. It's not like sinusoidal AC.... but always on would imply constant low current to provide longevity.
@@FranLab A multiplexed display would also pulse the filaments, I don't see why it would be a problem.
Really cool Fran well done for getting it working 😊
… I think that the way you’re explaining it, description of what & how it does it, etc., that’s probably how the original designer of the thing thought about it, as he sat down to begin this project.
"Are Fran's Electric?" by Gary Numitron.....
Lovely project! Hope you'll develop it into something Fran-tastic
Love You Keri!
@@sheep1ewe thanks - I'll Keri on as long as it takes :)
@@KeritechElectronicsI realy hold my thumbs for a collaboration with Fran in the future, that would be realy awesome! You and Fran would be a perfect match for a collaboration!
@@sheep1ewe I'd absolutely love to do that! It's been my dream for almost as long as I ran my channel. Maybe someday... online or IRL. If I ever go to the States, FranLab will be the #1 on my list of places to visit.
@@KeritechElectronicsYou wisit her lab (or she made a wisit to Your lab) would be awesome!
Fran, could the second power rail on the board be designed for a different voltage? Multiplexing the displays would result in much dimmer digits. I think you'd need about 12-15 volts to reach the same filament temperature as a constant 5v would provide.
There will be a part two...
Fully TTL desktop calculator, 60s/early 70s style.
I have several dual in line displays, in one way I'd like to build something with them, unfortunately as they are now unobtainium, it would be totally unrepairable. So they just sit in a tub.
Ooh RCA tubes!! I love RCA!!🥰🥰
I know the CD4511 will blank on BCD values above 9, do those 7 segment decoders blank or show odd segments with BCD above 9?
The SN7446A used here will show odd segments. They are well defined in the '46 datasheet.
Love the board, the Numitron display tubes and listening to your interesting analysis :-) Way beyond me but always enjoy watching you and learning.
So pretty 🧡
Make a wall or bench mounted aux voltmeter or multimeter display 😊
Thanks for a cool project.
Can the life of the numitrons be extended by running them at a lower current ?
Of course yes, just like any incandescent lights.
What would this be used for if it can only drive one digit at a time?
When the display scanning frequency is driven up over 60hz you just see all the digits on at the same time - persistence of vision.
Technically almost every LED clock does exactly this, just much, much faster.
Clock with capacitive touch to activate circuit.
Resurrecting..I see what you did there, happy Peeps Day ; )
Love this episode
And so did I!
I have some similar minitron displays salvaged from old aircraft radios. However most are worn out - Fran is right on this one. Can't use them for an always on display like a clock. They also get hot!
You rock! 👍
Great video, Fran...👍
you need to make a divergence meter! like in Steins;Gate! El Psy Kongroo!!
(@2:33) FranCAD, I see. 😊
very cool Fran 😀
only thing with the Diode Matrix is, you have to decide
whether to connect the Red wire or the Blue wire HA HA
Very cool!!
Subbed, Australia is watching 😇
TI-30. Swoon!
Ate these called " Nixie tubes" in the u.k. ?
No, nixies and numitrons are two fundamentally different animals. Nixies use the glow of ionised neon gas to produce light, while numitrons are just incandescent lightbulbs, with filaments arranged in a 7-segment display fashion.
@@mrnmrn1 ok. Thanks 👍
hi fran !
Excelent vídeo
❤️🔥
Come parli ?
👏👏👏👏👍👍👍👍