You can always look at the getter material - if it's nice and shiny, the vacuum is good. If it's white and corroded, the vacuum is gone. This applies for all vacuum-devices with a fired getter.
28:06 The pulsing is not because of the power supply, it's because you are using Pin1, which is reserved for serial comms along with Pin0. And that's why it's pulsing. Never use pin0 and Pin1 for anything else, unless you know you want to do that.
VFD's have always been a favorite of mine, especially the VU meters in top end cassette decks. Back in the mid eighties I was just starting out with electronics, and I spent a week trying to get one of these to work, finally giving up due to lack of knowledge and support equipment (I was in my teens). Awesome video, keep up the great work !!
The reason why the screen was blinking (27m39) is because you are using the TX and RX lines of the arduino as output lines. It will work but evertytime you upload to the arduino or the arduino is trying to send data over the serial, you will see the screen blinking.
It wasn't quite obvious at first but it seems like the grid selects the column of characters, while each dot of the anode is driven separately. Early in the video it sounded like the grid was the only thing controlling the dots. The interface also reminds me of the HD44780. One of the lines probably selects between data and commands, and there might be a read/write pin as well.
Now, wouldn't that be interesting if they emulated the interface of an HD44780? :) One of the remaining pins for which Dave has not discovered a use could be to switch to command sending mode.
Andrew Litt They do - I used to have one on my server until a friend bumped it and sent it to a shattering death behind the stand the server sits on. Dave if you look up HD44780 chipset, you should be able to do lots of stuff like progress bars (using custom characters), position the cursor, turn off the blinky line cursor (or turn it into a block) etc etc etc. :)
I've been watching your channel for a few years now (while attending school for comp E), and it's awesome to say that I finally understand the content in the videos.
Thanks for this Dave - I love VFDs reminds me of my youth as an electronics apprentice at GEC Traction. By the way, Babcock are a multi-faceted company... even building nuclear reactors for submarines.
I can't believe only one other person thought the Dr. Emmett Brown quote was worth a comment. Thank you for the smile along with the information. You are a genius.
Excellent video, I always enjoy seeing hacky DIY stuff on the EEVblog, rather than the usual mailbags and teardowns (not that there's anything wrong with those).
This is a typical VDF found in things like Gas Pumps and Cash Registers (The external customer display on the pedestal). We have boxs of these types of displays that came from everything from calculators & cash registers to alarm panels and old retired ATMs. varying colors from traditional blue/green to amber/orange
Working these interfaces out is (often) pretty straightforward. Some of the older ones require supply voltages that are difficult, such as 22V@250mA for one of my IEE's, but by the late 1980's, they were pretty much 5V all the way. The ones I have worked with fall into two general camps: Those that are designed for systems going VFD from the get-go, and those to connect instead of a printer. The first type generally have command/data selection via either a separate latch or a select (low is data, high is command on the one I am looking at). The second type is `Centronics compatible', with commands embedded int he data stream. They may honor backspace, LF, CR, and a few other control chars, as well s other special. Worked with a display years ago that did Epson dot matrix compatible graphics that inerfaced this way.
I have a pair of large 2x20 character VFDs made by IEE. I've been wanting to build a "command module" controller for Kerbal Space Program. There are existing projects that provide both hardware input controls (buttons, switches, joystick axes, etc) and readouts (character displays, analog meters, graphic readouts... even an Apollo style "8-ball" Flight Director Attitude Indicator). All I can say, is that every command module ends up different. The creativity of each builder really shines, and display choice can make a CM controller really stand out! Anyway, I LOVE the enormous 11mm character height of my VFD modules, and the two I have would provide 4x20 in total characters! I also have several smaller VFDs that could be used to display other flight data. I have two retail display posts that have either 2x16 or a 2x20 VFDs I can salvage... I'd have to dig them out. I'd like to use linear panel meters for my fuel gauges, and I was gonna use 7 segment LEDs for most of my numeric readouts (I'm considering doing the altitude in nixies... can't afford to do ALL of it in Nixies though... Bummer :P ) The large 2x20 modules I have feature a 26 pin header that is attached right next to a 40 pin NEC chip as well. Promising. The power connector is 6 pin, 2 pins are clearly power, but there is also a third pin connected to a very thin trace. Unsure as to it's function, or why it's on the power connector. *UPDATE:* I emailed IEE, and they sent me the original datasheets for my VFD module! YES!!! No serial interface on mine, but VERY similar to the one here, possibly the same processor. It uses an NEC 40 pin DIP and a 12 MHz crystal, with an interface that features the same number of pins (but mine is already brought out to a header, rather than a card edge connector). Now that I have a solid application and datasheets, I kinda wish I had bought more of these! I only paid 20 bucks each for the pair I have! I'll post the link on the blog.
One of the better videos in a while - I for one much prefer the videos were you are actually DOING something to things like the mail bag. Also, more Fundamental Fridays, please!!!!!!!!!!
Awesome! My dad's vending machines have tons of these, I could not find out for the life of me how to operate one. Now I can update the machine circuits without buying all new displays :)
I think the pulsing is due to the serial UART on pins 0 and 1 (one's Rx and one's Tx). The Arduino software includes bootloader on the ATMega (which is usually pre-flashed on complete boards) which allows it to be programmed over the UART (the USB is simply an FTDI serial port on those pins, with a reset on serial connection). When the board resets the bootloader waits a while for UART data before proceeding to the program so uploading and verifying ends up toggling the serial (it also flashes what's an LED on pin 13 most boards to show the bootloader's starting (happens irrespective of whether you're uploading anything).
These were my favorite display's back in the day. All my older amateur radio gear used VFD's but years back everything went LCD because it was far cheaper and used way less power.
Pioneer's dual-channel 24 segment dB displays were among the most visually rewarding. It's no small part of the attraction of their classic "Silver, quarter-inch, milled aluminum faceplate" dreadnought weight division Tape Decks like the RT-909 and the CT-F1250. As the years went by the resolution dropped, the number of segments plummeted and what was once a great thing, got cheapened to the point of the slower responding, overshoot prone physical "needle meters" once again prevailing, also owing to improvements in execution/design. (At least in neo-vintage gear.) But nothing reproduces the look of those great Flouroscan (their trade name) displays.
Thank you for your videos, as an undergrad in EE I was wondering how some things were done but was not quite sure how to go about it. Your videos are making things I somewhat understood make much more sense.
WTF Dave. I literally only just bought a VFD a couple days ago and been fumbling trying to get a pinout having never used a VFD display before... then you bring this video out. It's like you're physic
Thanks Dave , I had just posted on the forums trying to trouble shoot a VFD from Noritake and had one response so far to get me started on seeing if I can repair , and now you video comes up :) This should help with a couple steps in finding a fault in mine . I'm reversing without the engineer ;) John
Really enjoyed this video, Dave. Nice to hear the thought processes you used and listen to your explanations. Please keep doing videos like this. Very educational and entertaining!
The construction of VF devices allows them to be used as a single vacuum triode capable of amplification at up to 100kHz or so. Good for a novel audio preamplifier stage with a mellow 1940s sounding output.
IvanIvan1974 Unfortunately not at the same time. There is effectively only one anode and one cathode for all the grids. Hence only one triode. Two triodes would need two VFDs.It's more of an interesting curio than a really cutting edge thing. Of course the transconductance (gm in mA/V) would need to be measured first. The designing part is then a lot like designing with an n-FET.
I'm going to quibble with the control description around minute 6! The grid typically covers many anodes in a VFD (at least the old Soviet VFDs like the IV-22). You supply anodes individually to light them up (typically through a high-side driver IC), and can control the brightness of the entire display (a single 7-segment cell in the case of IV-22) by varying the grid voltage.
Nicely done! I always find these videos interesting. I have nowhere near the electronics knowledge to do this kind of stuff, but I have enough to grasp the concepts in your explanations and I appreciate your enthusiastic approach.
Awesome! I could watch videos like this (reverse engineering hardware) for days. Thank u Dave, and please do more like this. I am your padawan and learned a lot from this one vid and your narrative!
I have one of these lay around, out of some old, broken HP measurement device I got a few months ago. I was thinking of throwing it away due to lack of storage space, but now you inspired me to hack it, and see if it still works. Let's see what I can get :) Greetings from Germany!
So awesome to see you get this this to work. Had such a module for ages in a drawer from an old commercial scale with red light coming off it. Shame that I got rid of it because after seeing this I would have love to try and get it to work myself as well. Really enjoyed the video! Thanks!
The joys of using the minus sign - for electrons, as it was used by Benjamin Franklin who invented the notation and also knew that electricity flows from - (negative) to + (positive) side of things. :)
My digital clock broke recently so I’ve been researching clocks and it occurred to me I could make my own with an Arduino and a display, so now I’m researching VFDs since I love the look. I might try to salvage one from some broken AV equipment if I can any on eBay or any dumpsters near me, good ol’ reusing and recycling right!
27:55 One must keep in mind pin 0 and 1 are the serial programming input on the arduino, so if you need to use them, be ready for your project to have lots of pulsing on them during programming, and your serial monitor may poo the bed if you are driving them.
These things have their own character table, best check if there is a datasheet to show the table or else mod a new chip in place of the controller with a new character lookup table to suit your needs. The nice thing about this vfd is that it is early design which allows every bit of it to be highly modular. Wish all technology today is like that.
5:35 You say if you want to turn that dot off, you can simply make the grid voltage the same as the cathode. Understood, but the problem is that the grid is not as fine as the dots; there is often only one grid on for many many dots - how can these dots, under a shared grid, be individually controlled? You might say that you can do this by manipulating the voltages on the dots, but if this is the way, why should be need the grids at all? Thanks.
I would say that display was designed to be interfaced with an 8051/6805 parallel system bus. So, the line your toggling is probably the R/W line. But there is either a command/data line, Or... you can try rotating through all the non-ASCII bytes, some are likely to be control codes. Such as brightness, CLS, char position, etc.. Some displays even use an Escape code. So you would send the Escape char, and then the control code.
Fascinating!!!!! I watched, enthralled, mesmerised, honestly. I saw VFD's coming in (slowly) in the early 80's then they became the norm for a while, included on HiFi equipment, etc. VFD's were beautiful, largely due to the color but also the unknown, mysterious, they were like magic; now I am old, they have NOT lost their beauty and charm. When I watched your video, it was like watching a magic trick being explained, but, due to the way my brain works, you have added to the wonder :-) Well done and thank you All the best Jay (South Wales)
In the test sequence printing the character set it looked like it had some non ASCII characters: do you think that test sequence showed the character set in order (thus showing how to get the extra characters)? My other question would be if the controller accepted ASCII cursor control bytes, to move it around, etc.
EEVblog I just stumbled upon another VFD module datasheet, and apparently one of the unknown lines on the board is most likely a toggle between Character input and Command input, so you can change things like brightness duty cycle if you put the module into Command mode and write the relevant bytes. Will see if I can find it again, not that there's any guarantee they use the same commands...
That's exactly what I thought when watching it. Probably using characters under 32 for control codes like the old dumb terminals attached to mainframes/mini computers used to use. It'd also make it simple to connect to an old modem :)
Really good work there Dave! Watching your videos make wanna blow-off the dust from my E.E. degree. Too bad I am in another business and got very little time to do that =(. Keep up the good work mate, I really enjoy your videos. Greetings from Greece!
Am I the only one who read UFO's instead of VFD's up there in the corner of the whiteboard? =) Anyway, awesome and interesting video as usual Dave, very much appreciate it !
That was just a lot of fun. Really enjoy your stream of consciousness reverse engineering videos. It's good for the newcomers too since it shows the dead ends, the oops, etc. The other canned videos where everything goes perfectly can be frustrating for beginners because they think they are doing something wrong. Nice vid Dave
I fix pinball machines for a living and this appeals to me a lot. Gottlieb Pinballs used green VFDs similar to this one, which reminds me of the first Sys 80B. displays. Also, Babcock used to manufacture orange 128 x 32 DMDs for newer generation pinballs in the 90s.
Late 70's early 80's, a LOT of hand-held games used the VFD. This was before cheap black and white LCD displays were used. Most games didn't use dots like pixels, but actual shaped elements for the game; say a space shooting game would have spaceships, and shots, and stars all spaced out over the display (very jumpy motion, yes.) Also, many were quite colorful...the "standard" output color is light blue/aqua (still common, even today) and orange. By using shaped filters either inside or colored strips outside you can get blue,green,red,and pale yellow. These were fairly expensive to make, so by the mid 80's things were going toward LCD displays....I have a lot of games like this in 3 boxes in storage!
The reason for arduino to have a non-standard header spacing is because they accidentally put it like that on REV 1.0 and they will keep it like this to not break compatibility with boards/modules made to plug directly into the arduino
Well, at first there was a vacuum in there, then there was not. It must've gotten OUT somehow! This is actually an interesting case, since the more intuitive way to see a vacuum is as a something than the absense of everything. We create a vacuum for specific purposes and once we have it we would less likely think about it as "an absense of air", but rather as "a presence of lack of air" When a vacuum goes missing it's kinda cryptic like "Nothing just stopped existing. It took so much energy to make this nothing, too!" And because of that, I find a description like vacuum leaking out more sensible than air leaking in which doesn't get the point of the vacuum disappearing across quite as well. Proper descriptions I believe would need more words, which I also find dumb when literally everyone gets what the meaning of what is said is.
I've never had the opportunity to play with one of theses, so i really enjoyed this video. As for it not being ASCII, well it could be some variant of extended ASCII. The easiest way would be to map out the characters in the form of a loop. Start with 32 (SPACE) and end with 255. 32 through 127 are the printable ASCII characters, and 127 - 255 are the 'extended'. I'd also like to see if one of the first 31 characters generates a clear screen.
I'll look forward to it! I really enjoy these "hacking"/hobbyist videos. I think it really opens peoples minds to what people can do with whatever piece of hardware they have sitting in the bottom of their closet.
Awesome video, Dave. I managed to salvage one of those types of displays from an old shop till. So I'll definitely be putting your theory to the test on mine :)
That was a really nice logical approach to gain some understanding of a bit of unknown gear and in making it useful again. I enjoy seeing this kind of recycling of electronics. I myself like the VFD's, IMHO, LCD's seems cheap and nasty by comparison to such an elegant bit of engineering as a VFD.
EEVblog VFDs are sexy, sexy displays. IIRC Ford is using them again in their cars and other car mfgrs are following suit, because they last longer than LED/LCD diaplays for that sort of stuff, in practice. yes, they can be a bit tempermental when cold, but they've stood the test of time since the 70s (even when they start to fade a bit)...! I have a couple calcs here from the late 70s from Casio and Omron with VFD bulbs in them and I know you've done at least one video on one in the past. plus, the color they throw off is just amazing and all kinds of futuristic. I've been considering making or finding a 5.25" mountable unit I could stick in a CD-Rom bay in a tower for a long time, so I can push a button and display stats such as drive/system temperatures, as I have a lot of drives I want to keep in check, and summer will be here in the States soon enough... your video here has only fueled that want for such a system exponentially, mostly because VFDs look pretty good from any angle, whereas an LED-backlit monochrome LCD display will have wash-out issues from horizontal and vertical angles that aren't head-on... This is very annoying sitting next to the computer in question. So yeah, if you want to try your hand at trying to reverse-engineer the display from a recent Ford or other brand 2010 or newer, that isn't a Falcon, I'd go over and see if you can pull some out of some totaled cars at the local scrapyard, if there's any to be had there. (I dunno, maybe a Falcon will have some.) but if the place is reasonable, that just may be the place to pick more up, if you're willing to drip some elbow grease in the process.
Would have been nice to see you using the Analog Discovery for that.. because as you might remember you have a pattern generator in this and even in the static-I/O menu you can have a slider from 0 to 255 for a whole 8bit :)
I was wondering if one could replace the mesh with a conductive polymer membrane or a gauze woven from ITO (ISnO) coated glass fibers. I enjoy the look of VFDs but would be very interested in modern technology being applied to them as well. It also seems possible to make a classic VFD at home from scratch.
Dave, thank you so much. I've been learning a lot from you. This one is very cool. I was planning to make a vu-meter using vfd and this inspired me a lot :)
Nice one Dave, I love VFD's, They were great in early Casio calculators. This video is going to very useful. They look so much cooler than LCD's and have advantages in dark rooms that LCD's fail at. It's a pity they draw so much current huh?
Why are they driven from the front? Surely the wires and the mesh could be put behind the phosphor elements? (or do the electrons have to pass through the phosphor material onto the conductive plane - that would explain it)
One thing you forgot to mention Dave, with multiplexed VFDs, either the Cathode has to be biased slightly above ground with the grid being pulled to ground, aka Cutoff. Or the other way around, leaving the Cathode at 0 potential, while biasing the grid negative. This is required to prevent ghosting.
Presumably the safety concern with the board's high voltage section would be more a discomfort issue than actually dangerous? Also, while I understand the basic cathode->grid->anode behaviour of an early valve I don't see how you light/dark the individual pixels of the display element. It looked like the grid was only discrete for each whole element and not the individual pixels. Was there some way to energise the individual sections of the grid that stood over the corresponding pixel?
I'm a bit of a n00b so forgive the stupid question... 14:30 in the video - you state it's likely to be 5v, then you test it using your multi-tester. Your multi-tester says 000.2 ohms - how did you work out 5v from that? You start off with 0.1 amp on your oscilloscope and it fails to draw in 5v, so you crank up to 1a to get the 5v power feed you need. But I can't see how you concluded it was 5v from your multi-meter?
He already knows that that controller uses 5v. The power input is shorted to that pin (close to zero ohms). Therefore 5v is what must go in there. If it wasn't a dead short, then the power would be going through some other components first and more research of the board would be necessary.
I've used these displays in instrumentation systems. Nothing to hack to use these displays; just a parallel 8 bit word with a strobe line to load a character. IEEE makes a one that instead of just characters it has a dot matrix with built-in graphics. For example, send it a couple simple commands and a circle is displayed. You can set the circle size and location on the screen. Also has text size function built-in with various video effects, such as a curtain-like effect that "opens up" on the display to change screens.
I found a VFD in my disassembled Nakamichi Amp (it *was* broken beyond recovery, otherwise I would not have done that). Unfortunately, it's symbols are extremely specific and don't lend themselves well to repurposing.
You can always look at the getter material - if it's nice and shiny, the vacuum is good. If it's white and corroded, the vacuum is gone. This applies for all vacuum-devices with a fired getter.
***** Yes, good point.
28:06 The pulsing is not because of the power supply, it's because you are using Pin1, which is reserved for serial comms along with Pin0. And that's why it's pulsing. Never use pin0 and Pin1 for anything else, unless you know you want to do that.
True story.
VFD's have always been a favorite of mine, especially the VU meters in top end cassette decks. Back in the mid eighties I was just starting out with electronics, and I spent a week trying to get one of these to work, finally giving up due to lack of knowledge and support equipment (I was in my teens). Awesome video, keep up the great work !!
The reason why the screen was blinking (27m39) is because you are using the TX and RX lines of the arduino as output lines. It will work but evertytime you upload to the arduino or the arduino is trying to send data over the serial, you will see the screen blinking.
It wasn't quite obvious at first but it seems like the grid selects the column of characters, while each dot of the anode is driven separately. Early in the video it sounded like the grid was the only thing controlling the dots.
The interface also reminds me of the HD44780. One of the lines probably selects between data and commands, and there might be a read/write pin as well.
Now, wouldn't that be interesting if they emulated the interface of an HD44780? :) One of the remaining pins for which Dave has not discovered a use could be to switch to command sending mode.
Kevin Cozens I think some Noritake VFDs have an HD44780 compatible driver
Andrew Litt They do - I used to have one on my server until a friend bumped it and sent it to a shattering death behind the stand the server sits on. Dave if you look up HD44780 chipset, you should be able to do lots of stuff like progress bars (using custom characters), position the cursor, turn off the blinky line cursor (or turn it into a block) etc etc etc. :)
I've been watching your channel for a few years now (while attending school for comp E), and it's awesome to say that I finally understand the content in the videos.
Thanks for this Dave - I love VFDs reminds me of my youth as an electronics apprentice at GEC Traction. By the way, Babcock are a multi-faceted company... even building nuclear reactors for submarines.
we make vfd. ssot company. thanks😊
I can't believe only one other person thought the Dr. Emmett Brown quote was worth a comment. Thank you for the smile along with the information. You are a genius.
VFD's are BEAUTIFUL displays. That crisp, plasma-like colour is truly a sight to behold!
Excellent video, I always enjoy seeing hacky DIY stuff on the EEVblog, rather than the usual mailbags and teardowns (not that there's anything wrong with those).
This is a typical VDF found in things like Gas Pumps and Cash Registers (The external customer display on the pedestal). We have boxs of these types of displays that came from everything from calculators & cash registers to alarm panels and old retired ATMs.
varying colors from traditional blue/green to amber/orange
Working these interfaces out is (often) pretty straightforward. Some of the older ones require supply voltages that are difficult, such as 22V@250mA for one of my IEE's, but by the late 1980's, they were pretty much 5V all the way. The ones I have worked with fall into two general camps: Those that are designed for systems going VFD from the get-go, and those to connect instead of a printer. The first type generally have command/data selection via either a separate latch or a select (low is data, high is command on the one I am looking at). The second type is `Centronics compatible', with commands embedded int he data stream. They may honor backspace, LF, CR, and a few other control chars, as well s other special. Worked with a display years ago that did Epson dot matrix compatible graphics that inerfaced this way.
I have a pair of large 2x20 character VFDs made by IEE. I've been wanting to build a "command module" controller for Kerbal Space Program. There are existing projects that provide both hardware input controls (buttons, switches, joystick axes, etc) and readouts (character displays, analog meters, graphic readouts... even an Apollo style "8-ball" Flight Director Attitude Indicator). All I can say, is that every command module ends up different. The creativity of each builder really shines, and display choice can make a CM controller really stand out!
Anyway, I LOVE the enormous 11mm character height of my VFD modules, and the two I have would provide 4x20 in total characters! I also have several smaller VFDs that could be used to display other flight data. I have two retail display posts that have either 2x16 or a 2x20 VFDs I can salvage... I'd have to dig them out. I'd like to use linear panel meters for my fuel gauges, and I was gonna use 7 segment LEDs for most of my numeric readouts (I'm considering doing the altitude in nixies... can't afford to do ALL of it in Nixies though... Bummer :P )
The large 2x20 modules I have feature a 26 pin header that is attached right next to a 40 pin NEC chip as well. Promising. The power connector is 6 pin, 2 pins are clearly power, but there is also a third pin connected to a very thin trace. Unsure as to it's function, or why it's on the power connector.
*UPDATE:* I emailed IEE, and they sent me the original datasheets for my VFD module! YES!!!
No serial interface on mine, but VERY similar to the one here, possibly the same processor. It uses an NEC 40 pin DIP and a 12 MHz crystal, with an interface that features the same number of pins (but mine is already brought out to a header, rather than a card edge connector).
Now that I have a solid application and datasheets, I kinda wish I had bought more of these! I only paid 20 bucks each for the pair I have!
I'll post the link on the blog.
One of the better videos in a while - I for one much prefer the videos were you are actually DOING something to things like the mail bag. Also, more Fundamental Fridays, please!!!!!!!!!!
This is an absolutely amazing video! I love getting an insight into your thought process. Excellent job!
How does Dave know Bob is my Uncle! It's like he makes me aware of it every video I watch!
MyWillyboi I thought u were AvE
Dave's video on any subject is always master grade. Thanks.
Awesome! My dad's vending machines have tons of these, I could not find out for the life of me how to operate one. Now I can update the machine circuits without buying all new displays :)
I think the pulsing is due to the serial UART on pins 0 and 1 (one's Rx and one's Tx). The Arduino software includes bootloader on the ATMega (which is usually pre-flashed on complete boards) which allows it to be programmed over the UART (the USB is simply an FTDI serial port on those pins, with a reset on serial connection).
When the board resets the bootloader waits a while for UART data before proceeding to the program so uploading and verifying ends up toggling the serial (it also flashes what's an LED on pin 13 most boards to show the bootloader's starting (happens irrespective of whether you're uploading anything).
These were my favorite display's back in the day. All my older amateur radio gear used VFD's but years back everything went LCD because it was far cheaper and used way less power.
Pioneer's dual-channel 24 segment dB displays were among the most visually rewarding. It's no small part of the attraction of their classic "Silver, quarter-inch, milled aluminum faceplate" dreadnought weight division Tape Decks like the RT-909 and the CT-F1250. As the years went by the resolution dropped, the number of segments plummeted and what was once a great thing, got cheapened to the point of the slower responding, overshoot prone physical "needle meters" once again prevailing, also owing to improvements in execution/design. (At least in neo-vintage gear.) But nothing reproduces the look of those great Flouroscan (their trade name) displays.
This is the type of video i look forward to Dave. Second to teardown tuesday
Thank you for your videos, as an undergrad in EE I was wondering how some things were done but was not quite sure how to go about it. Your videos are making things I somewhat understood make much more sense.
WTF Dave. I literally only just bought a VFD a couple days ago and been fumbling trying to get a pinout having never used a VFD display before... then you bring this video out. It's like you're physic
Thanks Dave , I had just posted on the forums trying to trouble shoot a VFD from Noritake and had one response so far to get me started on seeing if I can repair , and now you video comes up :)
This should help with a couple steps in finding a fault in mine .
I'm reversing without the engineer ;)
John
Really enjoyed this video, Dave. Nice to hear the thought processes you used and listen to your explanations. Please keep doing videos like this. Very educational and entertaining!
Just found this while trying to repair an old vfd, this is priceless info, thanks Dave
The construction of VF devices allows them to be used as a single vacuum triode capable of amplification at up to 100kHz or so. Good for a novel audio preamplifier stage with a mellow 1940s sounding output.
...and use the fluorescent part as a spectrum display?
IvanIvan1974 Unfortunately not at the same time. There is effectively only one anode and one cathode for all the grids. Hence only one triode. Two triodes would need two VFDs.It's more of an interesting curio than a really cutting edge thing. Of course the transconductance (gm in mA/V) would need to be measured first. The designing part is then a lot like designing with an n-FET.
Electro It was a joke, but yes, there would be a few parameters to work out for a proper design, if even possible.
I'm going to quibble with the control description around minute 6! The grid typically covers many anodes in a VFD (at least the old Soviet VFDs like the IV-22). You supply anodes individually to light them up (typically through a high-side driver IC), and can control the brightness of the entire display (a single 7-segment cell in the case of IV-22) by varying the grid voltage.
Nicely done! I always find these videos interesting. I have nowhere near the electronics knowledge to do this kind of stuff, but I have enough to grasp the concepts in your explanations and I appreciate your enthusiastic approach.
I am very happy to have found this UA-cam channel. You are a genius. My respect and admiration. Very good video, very good job. Best regard.
Awesome! I could watch videos like this (reverse engineering hardware) for days. Thank u Dave, and please do more like this. I am your padawan and learned a lot from this one vid and your narrative!
I have one of these lay around, out of some old, broken HP measurement device I got a few months ago. I was thinking of throwing it away due to lack of storage space, but now you inspired me to hack it, and see if it still works. Let's see what I can get :)
Greetings from Germany!
I really enjoyed this one. It was interesting in that it described how VFDs work AND it had a bit of a voyage of discovery. Great stuff Dave, thanks.
Fantastic video Dave! Reverse engineering old bits of kit like this is really interesting, I'd like to see more of this on your channel, keep it up!
So awesome to see you get this this to work. Had such a module for ages in a drawer from an old commercial scale with red light coming off it. Shame that I got rid of it because after seeing this I would have love to try and get it to work myself as well. Really enjoyed the video! Thanks!
Loving the vacuum leaking out comment really had me in stitches :D
maybe he imagined to be inside the display
New from EEVBlog Studios: the VFD Shield.
Very cool to see this work in action.
"if the vacuum leaked out" good one Dave. We know what you meant :D
The joys of using the minus sign - for electrons, as it was used by Benjamin Franklin who invented the notation and also knew that electricity flows from - (negative) to + (positive) side of things. :)
My digital clock broke recently so I’ve been researching clocks and it occurred to me I could make my own with an Arduino and a display, so now I’m researching VFDs since I love the look. I might try to salvage one from some broken AV equipment if I can any on eBay or any dumpsters near me, good ol’ reusing and recycling right!
27:55
One must keep in mind pin 0 and 1 are the serial programming input on the arduino, so if you need to use them, be ready for your project to have lots of pulsing on them during programming, and your serial monitor may poo the bed if you are driving them.
Or trigger the firing SCR in my case when I changed some code on my capacitor bank project. OOPS :)
dash8brj at least you were firing the scr at no more than 50% duty cycle and 9600 baud ;)
frollard Or more to the point, at least the bank wasnt charged much at the time (about 30V)
These things have their own character table, best check if there is a datasheet to show the table or else mod a new chip in place of the controller with a new character lookup table to suit your needs. The nice thing about this vfd is that it is early design which allows every bit of it to be highly modular. Wish all technology today is like that.
5:35 You say if you want to turn that dot off, you can simply make the grid voltage the same as the cathode. Understood, but the problem is that the grid is not as fine as the dots; there is often only one grid on for many many dots - how can these dots, under a shared grid, be individually controlled? You might say that you can do this by manipulating the voltages on the dots, but if this is the way, why should be need the grids at all? Thanks.
+EEVblog 28:00 The pulsing is due to pin 13 pulsing on boot of the Arduino
Fair play Dave, you're a treasure trove of brilliant info.
I would say that display was designed to be interfaced with an 8051/6805 parallel system bus. So, the line your toggling is probably the R/W line.
But there is either a command/data line, Or... you can try rotating through all the non-ASCII bytes, some are likely to be control codes. Such as brightness, CLS, char position, etc..
Some displays even use an Escape code. So you would send the Escape char, and then the control code.
For every idea I have or part I find there always a great eevblog video accompanying it it's great!
Fascinating!!!!! I watched, enthralled, mesmerised, honestly.
I saw VFD's coming in (slowly) in the early 80's then they became the norm for a while, included on HiFi equipment, etc. VFD's were beautiful, largely due to the color but also the unknown, mysterious, they were like magic; now I am old, they have NOT lost their beauty and charm. When I watched your video, it was like watching a magic trick being explained, but, due to the way my brain works, you have added to the wonder :-)
Well done and thank you
All the best
Jay (South Wales)
@14:05 you could just look at the voltage on the capacitor and stay below thevoltage limit. But nice work
In the test sequence printing the character set it looked like it had some non ASCII characters: do you think that test sequence showed the character set in order (thus showing how to get the extra characters)? My other question would be if the controller accepted ASCII cursor control bytes, to move it around, etc.
Good thought on the cursor control bytes, that'd be worth testing for sure.
rpavlik1 No doubt it would be in sequence, just a for loop in the code.
EEVblog I just stumbled upon another VFD module datasheet, and apparently one of the unknown lines on the board is most likely a toggle between Character input and Command input, so you can change things like brightness duty cycle if you put the module into Command mode and write the relevant bytes. Will see if I can find it again, not that there's any guarantee they use the same commands...
It's likely Extended ASCII. en.wikipedia.org/wiki/Extended_ASCII
Regular ASCII is only 7 bits whereas Extended makes use of all 8.
That's exactly what I thought when watching it. Probably using characters under 32 for control codes like the old dumb terminals attached to mainframes/mini computers used to use. It'd also make it simple to connect to an old modem :)
I've paused and counted: it's 40x2
One of your best videos, Dave! Couldn't find any bigger thumb than the default though. ;-) Thanks a lot!
Great video!
Really good work there Dave! Watching your videos make wanna blow-off the dust from my E.E. degree. Too bad I am in another business and got very little time to do that =(. Keep up the good work mate, I really enjoy your videos. Greetings from Greece!
Well done Dave, thanks for spending your time on this, keep up the good work.
Great video - I salvaged a VFD from a late 70's battery arcade game - gonna hook it up!
Am I the only one who read UFO's instead of VFD's up there in the corner of the whiteboard? =)
Anyway, awesome and interesting video as usual Dave, very much appreciate it !
The vfd explanation was very easy to understand. Thanks!
man that was great fun! more videos like this :) getting into old electronics like this is so coool
17:44 I particularly liked Dave doing the "Duanggggg ...".
That was just a lot of fun. Really enjoy your stream of consciousness reverse engineering videos. It's good for the newcomers too since it shows the dead ends, the oops, etc. The other canned videos where everything goes perfectly can be frustrating for beginners because they think they are doing something wrong. Nice vid Dave
I fix pinball machines for a living and this appeals to me a lot. Gottlieb Pinballs used green VFDs similar to this one, which reminds me of the first Sys 80B. displays. Also, Babcock used to manufacture orange 128 x 32 DMDs for newer generation pinballs in the 90s.
Late 70's early 80's, a LOT of hand-held games used the VFD. This was before cheap black and white LCD displays were used. Most games didn't use dots like pixels, but actual shaped elements for the game; say a space shooting game would have spaceships, and shots, and stars all spaced out over the display (very jumpy motion, yes.) Also, many were quite colorful...the "standard" output color is light blue/aqua (still common, even today) and orange. By using shaped filters either inside or colored strips outside you can get blue,green,red,and pale yellow. These were fairly expensive to make, so by the mid 80's things were going toward LCD displays....I have a lot of games like this in 3 boxes in storage!
The reason for arduino to have a non-standard header spacing is because they accidentally put it like that on REV 1.0 and they will keep it like this to not break compatibility with boards/modules made to plug directly into the arduino
16:26 the vacuum leaked out? I know you what you meant, everyone mess up now and then. Other than that, thanks for telling me how these work.
I hear that so much xD I always just say I hate when my vacuums leak out and they realize their mistake, usually :P
madgod 117 LOL! Yeah. It's a magic vacuum.
Well, at first there was a vacuum in there, then there was not.
It must've gotten OUT somehow!
This is actually an interesting case, since the more intuitive way to see a vacuum is as a something than the absense of everything. We create a vacuum for specific purposes and once we have it we would less likely think about it as "an absense of air", but rather as "a presence of lack of air"
When a vacuum goes missing it's kinda cryptic like "Nothing just stopped existing. It took so much energy to make this nothing, too!"
And because of that, I find a description like vacuum leaking out more sensible than air leaking in which doesn't get the point of the vacuum disappearing across quite as well. Proper descriptions I believe would need more words, which I also find dumb when literally everyone gets what the meaning of what is said is.
This video is my favorite EEVBlog. Love it!!
I've never had the opportunity to play with one of theses, so i really enjoyed this video. As for it not being ASCII, well it could be some variant of extended ASCII. The easiest way would be to map out the characters in the form of a loop. Start with 32 (SPACE) and end with 255. 32 through 127 are the printable ASCII characters, and 127 - 255 are the 'extended'. I'd also like to see if one of the first 31 characters generates a clear screen.
"vacuum leaked out" :D:D
Finally some actual electronics :-) well done ...
Fantastic! Thanks for taking the time to do this!
The polarity switching is on the phosphor elements not the mesh. It is likely that they are isolated for multiplexing.
More videos like this please. Truly enjoyed it.
Why not program it to some sort of Tweetdeck display for the office?
Well it's either that or teen spirit.
Although I'm not sure of the wireless capabilities of teen spirit...
RedRamzor I have bigger plans for a twitter display, literally...
I'll look forward to it! I really enjoy these "hacking"/hobbyist videos. I think it really opens peoples minds to what people can do with whatever piece of hardware they have sitting in the bottom of their closet.
EEVblog I'm not sure if I should be scared or not.
Great explanation. I would love to have one of those for some of my projects. Super-cool! Thanks.
Oh wow awesome timing, I just got several of these but much larger, and one big one that says plasma on the back but looks like a VFD.
Awesome video, Dave. I managed to salvage one of those types of displays from an old shop till. So I'll definitely be putting your theory to the test on mine :)
That was a really nice logical approach to gain some understanding of a bit of unknown gear and in making it useful again. I enjoy seeing this kind of recycling of electronics. I myself like the VFD's, IMHO, LCD's seems cheap and nasty by comparison to such an elegant bit of engineering as a VFD.
Awesome video, as usual. I wonder why those digital scopes are not floating ground. It would be so handy at times, just like when using a multimeter.
TD62783 is a great choice for a high-side driver IC for a VFD 7-segment display. Still available-ish.
EEVblog VFDs are sexy, sexy displays. IIRC Ford is using them again in their cars and other car mfgrs are following suit, because they last longer than LED/LCD diaplays for that sort of stuff, in practice. yes, they can be a bit tempermental when cold, but they've stood the test of time since the 70s (even when they start to fade a bit)...! I have a couple calcs here from the late 70s from Casio and Omron with VFD bulbs in them and I know you've done at least one video on one in the past.
plus, the color they throw off is just amazing and all kinds of futuristic.
I've been considering making or finding a 5.25" mountable unit I could stick in a CD-Rom bay in a tower for a long time, so I can push a button and display stats such as drive/system temperatures, as I have a lot of drives I want to keep in check, and summer will be here in the States soon enough... your video here has only fueled that want for such a system exponentially, mostly because VFDs look pretty good from any angle, whereas an LED-backlit monochrome LCD display will have wash-out issues from horizontal and vertical angles that aren't head-on... This is very annoying sitting next to the computer in question.
So yeah, if you want to try your hand at trying to reverse-engineer the display from a recent Ford or other brand 2010 or newer, that isn't a Falcon, I'd go over and see if you can pull some out of some totaled cars at the local scrapyard, if there's any to be had there. (I dunno, maybe a Falcon will have some.) but if the place is reasonable, that just may be the place to pick more up, if you're willing to drip some elbow grease in the process.
Would have been nice to see you using the Analog Discovery for that.. because as you might remember you have a pattern generator in this and even in the static-I/O menu you can have a slider from 0 to 255 for a whole 8bit :)
I was wondering if one could replace the mesh with a conductive polymer membrane or a gauze woven from ITO (ISnO) coated glass fibers. I enjoy the look of VFDs but would be very interested in modern technology being applied to them as well. It also seems possible to make a classic VFD at home from scratch.
Dave, thank you so much. I've been learning a lot from you. This one is very cool. I was planning to make a vu-meter using vfd and this inspired me a lot :)
Nice one Dave,
I love VFD's, They were great in early Casio calculators.
This video is going to very useful.
They look so much cooler than LCD's and have advantages in dark rooms that LCD's fail at.
It's a pity they draw so much current huh?
Nice Back to the Future III reference thrown in there, Dave. :D
ms_enj Every 2nd or third video has a reference I think!
+EEVblog i have a vfd and i need help on to make it display what i want but i know how to wire it
Asked for it and got it. Thank you very much Dave. :)
Why are they driven from the front? Surely the wires and the mesh could be put behind the phosphor elements? (or do the electrons have to pass through the phosphor material onto the conductive plane - that would explain it)
One thing you forgot to mention Dave, with multiplexed VFDs, either the Cathode has to be biased slightly above ground with the grid being pulled to ground, aka Cutoff. Or the other way around, leaving the Cathode at 0 potential, while biasing the grid negative.
This is required to prevent ghosting.
I love VFD; great video as always Dave :)
Awesome video, as usual!!! Congrats!!! I really hope you continue doing stuffs like these!!! Thanks!
omg i am a display obsessed one, especially vfd's, this is very satisfying thx
Presumably the safety concern with the board's high voltage section would be more a discomfort issue than actually dangerous?
Also, while I understand the basic cathode->grid->anode behaviour of an early valve I don't see how you light/dark the individual pixels of the display element. It looked like the grid was only discrete for each whole element and not the individual pixels. Was there some way to energise the individual sections of the grid that stood over the corresponding pixel?
+morelenmir The grid is for multiplexing each block not each pixel.
I'm a bit of a n00b so forgive the stupid question... 14:30 in the video - you state it's likely to be 5v, then you test it using your multi-tester. Your multi-tester says 000.2 ohms - how did you work out 5v from that? You start off with 0.1 amp on your oscilloscope and it fails to draw in 5v, so you crank up to 1a to get the 5v power feed you need. But I can't see how you concluded it was 5v from your multi-meter?
He already knows that that controller uses 5v. The power input is shorted to that pin (close to zero ohms). Therefore 5v is what must go in there. If it wasn't a dead short, then the power would be going through some other components first and more research of the board would be necessary.
Great video ! I got really excited when the test characters popped up ! LoL. Cheers!
I bought one from digikey. Best project ever.
And this is how we do it. I would have done it just the same way with my adurino attached....:)
Thanx for this one Dave.
The rest of the pins may provide direct cursor addressing, again using the latch to "execute" it. Are there enough pins for that?
i love vfd's
and i use them whenever i can
I've used these displays in instrumentation systems. Nothing to hack to use these displays; just a parallel 8 bit word with a strobe line to load a character. IEEE makes a one that instead of just characters it has a dot matrix with built-in graphics. For example, send it a couple simple commands and a circle is displayed. You can set the circle size and location on the screen. Also has text size function built-in with various video effects, such as a curtain-like effect that "opens up" on the display to change screens.
I found a VFD in my disassembled Nakamichi Amp (it *was* broken beyond recovery, otherwise I would not have done that). Unfortunately, it's symbols are extremely specific and don't lend themselves well to repurposing.