Vacuum Fluorescent Display Amplifier Tutorial

[@whitelion7976]

Your music makes me think of the stranglers - midnight summer dream. I love it considered I just discovered uses for old VFD displays but this is a surprise to me.

[@Pulverrostmannen]

@@whitelion7976 nice. If you like my music it have evolved a lot over the time that have passed from this video so if you like you can check out it on my separate channel. You should have links to it in the video description

[@magnedalaren]

I liked the intro, the laugh & music and all :P Nice video! seems complicated!

[@mackk123]

imagine a bunch of them in series for a distortion pedal for guitar.....!!

[@Pulverrostmannen]

mackk123 he he. That would be something cool. What pedal is that? That's my VFD limited special edition super pedal

[@SilverSpoon_]

technically every character acts like one at once, by this logic a VFD could be an IC but as a vacuum tube. saint celestia this is insane when you think bout it.

[@donellismusic]

Very interesting video man. Really neat, and looks very complicated.

[@28noch]

Interesting. But how supply an equipment whitout supply module. For example se-mxg7bk separated for use whith cassette deck?

[@Pulverrostmannen]

It would be possible to power this with many ways. Battery’s boost converters transformers etc. however are they very sensitive at picking up sounds so any unclean power will be heard on the output as well. The most sensitive part is to power the filament which needs as exact voltage as possible. And then the anode voltage is not really that sensitive in how much you have but you will find a range that is typically working best so having this variable is a good thing too and it affects the power output as well. The transformer output can be substituted with a decoupling capacitor if you only intend to create a low level output and connect to your cassette deck. Basically you put a resistor in series with the anode and then the capacitor output is taken out closest to the anode pin after the resistor the same way as the first stage is done here. The sound of these displays can be a bit tricky to calibrate and different displays gives different sound. But it is also the most airy spaced sound i have heard so far as well with anything I’ve built so it’s very interesting to try. It’s also good to know these displays can be incredibly microphonic so keep them away from vibrations. If you have more questions I be happy to help you out

[@alfonsogomez4725]

Hello, I would like to know how you found the anodes, cathodes and grids on those screens? The filaments are easy but the rest looks complicated, help me please

[@Pulverrostmannen]

That is the hard part indeed. It is actually a good question how to test this and I have some displays I wanted to test too so I went to my lab and tested what is a quick way to figure out the pinout on these and I think I have a good solution already. Procedure was: power the filament with enough voltage it barely glows visible. Connect a positive cable and tie this to the positive end that is powering the filament and put this to the side for now. Use a separate voltage source that must be a separate one from the first (electrically isolated) if possible very precise current limited. I have a precision lab supply capable of reducing current between 0-15ma and I don’t expect you to have this available so you can use a resistor instead. I find that 40 volts of this supply was plenty well working. If you use a 40 volt supply you could use a resistor of about 10K ohm in series with this supply. Anyway, you could even use batteries to power the filament if you don’t have two power supplies to make this test. Tie the negative end of the secondary (isolated) power to the negative end of the filament supply. Hook up your 40 volt supply in series with a multimeter dialed in to measure DC Milliamps. You are now ready to start probing your display. Make sure all legs on it is separated from each other. Also, some displays have a built in micro controller and they are useless for audio, they typically make very random weird things when you probe them. Use the positive end of your 40 volt supply and touch each leg of the display and monitor if any current shows up on your multimeter. If you touch a pin and you read a few milliamps this is very likely a Grid pin. These will most cases be in a row together but sometimes split in two sections and so forth. Just check and mark the legs of the display that gives you a reading on the multimeter. All other pins are likely to be the anodes and they will not read a current until the grid is biased. That’s why you can separate them. Final test is to grab the first positive wire you connected to the positive end of the filament. Put this wire to a leg of your choice that is supposedly the grids and hold it there. Take the 40 volt wire and touch or drag along the other pins of the display, some segments should light up now if you done this correct and there is no built in controller in the display. Each grid pin should move to a different segment of the display and if you move the anode various pixels should light up in that segment. If you get this result congrats. You figured out the pins. Now solder the grids and anode pins together to make a super triode segment of the display and you can proceed to build your amplifier with it. It may sound a bit complicated when I give this much information at once but it really worked like a charm here for me on 3 tested displays. Let me know if you have more questions or if you want me to make a video about it. Let me know how things went

[@alfonsogomez4725]

@@Pulverrostmannen Friend, thanks for your time and response, I do understand what you have explained to me but it would still be great if you made a video to make it even clearer and to be able to recycle all these ancient wonders, A hug from Chile.

[@Pulverrostmannen]

@@alfonsogomez4725 hi again! Sorry for the slow response, I have a lot going on for the time. Making a video is a pretty good idea and I will try to make one some day showing it better. I cleaned up the mess in the lab and made some more experiments with resistors and a multimeter on what I described earlier and it seems to hold up too which is good. I keep you updated if I find time to make a video for you

[@alfonsogomez4725]

@@Pulverrostmannen ok thanks

[@Plato1959]

Great!! I have some questions. What is the OPT primary impedance? Is it hard to get enough audio without pre-amp? Can you advise me?

[@Pulverrostmannen]

Plato 1959 thanks! The primary impedance of my opt is about 700 ohms. Depending how you build your amplifier it can be of higher value too but it depends on the final operating voltage. There are many ways you can make the VFD to operate but I decided to make it will less voltage than most do and have thus a positive Bias to help the electrons to move properly. Tips of what you can think of when you build is many and I list some of them here for you :) Bias: it is very important to have the correct voltage on the grid negative in most cases when you use a high operating voltage. And positive if you going to use lower voltage. Just enough voltage will increase both gain and sound quality. Gain: higher voltage will also give better gain factor and more power output. However will the VFD itself have a very big difference here too. It can only give a small amount of power even at best. A display with more pixels or letters gives more output. Pre amp: yes a two stage amplification can further give a better gain in the end. If you plan to use the VFD to power a speaker directly I would recommend a pre amplifier to ease the load of the sound source. As a VFD pre amplifier it mostly would work to use a single display and power the desired source. Voltage: operating voltage affects both the sound and the output. I have seen there is mostly a limit when the voltage gives gain but eventually start to distort the sound. Filament voltage: it is important to use correct voltage here. There is a good tolerance here still but too much voltage will up the filament resistance and lower illumination and even gain and too little will result in less power and quality of sound. I prefer to use a high frequency regulator or DC voltage to reduce hum as the filament is direct heated cathode. Protect your equipment. Capacitors is a must. AC compatible capacitors are connected in the input to protect the source from back voltage. Capacitor on the output signals if used as a pre amplifier. Sound: the signal cables are preferably screened. The displays are sensitive to vibrations or strikes when they work as a sound source and you will hear the sound output sing the filament song for a while if you knock the display he he. Output transformers needs to be used as you likely know to filter the DC current from the circuit or else it may damage your speakers or headphones. If you have more questions feel free to ask me as much as you need. I be glad to help you out the best I can :)

[@Plato1959]

Thank you for lots of and useful informations!! I thought you used 10k-50k:8ohm OPT, and as tube amps, input resistor maybe connected from Grid to GND. But your circuit was using 1M variable resistor between Grid and Anode. My last experiment was supplying only 13V for anode, using input resistor between Grid and GND. So I have to revise the voltage, circuit and OPT. I feel like trying it again. Thanks!!

[@Pulverrostmannen]

glad my information could help, I be happy if you tell me about the progress you make!

[@hannuhanhi183]

Korg nutube DIY. Ever considered to use individual digits to as one triode to get more gain ?

[@Pulverrostmannen]

Hannu Hanhi I have not tested that actually, it could be fun to test this and see if it actually works or not.

[@hannuhanhi183]

Pulverrostmannen: Yes, got inspired by this that I ordered some VFD displays from aliexpress with 9 digits. Going to try them out when I have some more spare time.

[@Pulverrostmannen]

Hannu Hanhi I wish you luck then! I can mention some stuff that are good to know that can be help during the testing. It is very important to have really clean DC voltage for this to work as it will output lots of hum or other bad sound otherwise. It goes both for the filament and high voltage. The level of power and gain you get from the display is very dependent of how the grid is biased (grid voltage) and the higher tension voltage on the plates it will be louder too. For Signal output you want a preferable output transformer from some old radio or such to filter out the DC from the signal and convert high voltage to lower with more current to drive a speaker with. Different types of trannies gonna give various results. The filament voltage needs to be very narrow on a certain level, too low voltage will lower emission and power and too much gonna make one side of the display more dimmed than the other because the resistance rises with the voltage and and more electrons leaves the thread too early on the brighter side. And as the power is very low on the displays it is possible to use them in parallel to increase the power too. Or step up the signal and voltage by using them in series as I do here. When listening to the sound from the displays trough headphones gave a very interesting airy sound that was very tube like. Poking the displays can make them oscillate too as the thin filament starts to vibrate and can keep going for a long while which can be good to know. Let me know if you have more questions and how things went in the end. Happy experimenting and be careful!

[@hannuhanhi183]

Pulverrostmannen: Thank you for this highly useful information :)

[@electronics3992]

can I make this with only one vfd display and with 30omh speaker or i must use other speaker??

[@Pulverrostmannen]

It depends a lot how easy your speaker is to power and also how big display you got, vacuum display amplifiers are very different with what result you get in the end and as in my case were I directly power a speaker with the display is very hard because the output wattage is very low, and it is not sure you get the same result in the end as I did so it takes some experimenting and tweaks. the bigger display and more digits or pixels it got the better output you will gain, and also how high voltage you can make it operate with. A decent output transformer is to recommend as you need to filter the DC current to the source/speaker some way so you don't damage your equipment. I have made a pretty large tips and tricks comment to Plato 1959 in the comment feed of this video, you may find many useful tips by reading this to get you going, I have Favorite marked the comment now so you should find it pretty easy. I be glad to hear how things went in the end. and feel free to ask any questions you may have regarding this

[@electronics3992]

Pulverrostmannen thx I will put few displays in parallel . what is max voltage I can use ?? I made yojr circuit and when I connect headphones to amp and directly to sigbal generator sound is same and.

[@Pulverrostmannen]

the gain value is depending how much voltage you can use, this is also depending with the displays how high you can go so you may need to step up the voltage until you don't gain more power or the sound goes bad, I could go up to about 70volts before the sound started to sound bad, using a higher ohm output transformer along with higher input voltage would amplify the sound, if the OPT is too low resistance the actual gain will eventually be even lower than the source in the end. you will need to make use of decent output voltage to reduce that to high current to power to the speaker. using more than one stage makes it easier to gain power as well as you can step up the voltage this way. as we work with a positive bias you also have harder to transfer the sound signal as you play using a low voltage signal to pass through. VDFs for use as a Pre amplifier may be used more easy than a actual amplifier stage for this reason. but in any case you make this I have still gotten the typical "Tube sound" trough headphones using this method even if the actual output is just above or close around the soundsource headphone jack still makes the sound interesting to listen to. I have not tested out very much using higher voltage as I don't have any better OPT for the time, but with higher input voltage the grid don't need as much bias voltage and you can get the signal through better this way, in theory

[@StephGV2]

Couldn't those segments be used discreetly for a lot more triodes out of a single display? Think of the equalizer you could make.

[@Pulverrostmannen]

Yes, technically it could if you have several segments of the grid separated from each other. The problem is that each digit so to speak each triode puts out very small amount of power so to make it useful it’s mostly better to parallel them and get your hands on more displays to get extra stages of amplification. I have this in mind actually to see how much one section could power or if I could actually make a working amplifier with displays as preamp stages instead. I probably gonna go deeper into this in the future when I have time to see what they can really do if optimized the best way

[@StephGV2]

@@Pulverrostmannen So, suppose the goal isn't power, but saturation and tone? And you get your power from MOSFET class A or some class D amps. Also, what about serial output from one triode to the input of another, over and over?

[@Pulverrostmannen]

Stephen Greenberg yes that is very possible. Since the display don’t have much output it is most suitable to use as a signal amplifier. And this type of display triodes are actually implemented in some amplifiers made by real manufacturers and goes by the name “nutube” and can be found in guitar amps such as VOX AC MV50. The sound can be very nicely distorted if the display is overdriven the right way and you can make the desired affects by controlling the bias vs signal voltages. The pros about VFD triodes are that they are compact, pulls only a small amount of power and has a long life that is typically longer than a regular preamp tube. The cons are that it is more difficult to get the right sound and especially displays are extremely “microphonic” and if you tap them or make them vibrate they can start to self resonate and go on for a long time before they get quiet again but this varies from types naturally. You also need to have a really “quiet” filament supply that is the right voltage for the vfd not to affect the sound. You can indeed use more sections in series to make it a voltage amplifier and this is what I did in this video but using a separate display as preamp and the secondary as the amplified stage. It only gives two stages of amplification but it was enough to power the speaker with only the headphone jack from my iPad as signal. You can control the level of amplification by running the anodes “plate” or the fluorinating digits in this case at a higher voltage and control the load impedance to keep the current low at all times. The grid voltage controls how much current the tube lets through. Typically on this you make the grid voltage lower as you increase the anode voltage. If you want to distort the tube you simply make the signal voltage go outside the saturating points of the display

[@StephGV2]

@@Pulverrostmannen Yeah, the idea is to get Nutube sounds, without the $50 per piece Nutube price. That can add up if you wanted to use 5 of them or more for a "tube EQ" or effects pedal or boutique small amp.

[@Pulverrostmannen]

Stephen Greenberg indeed the price leaves some to wish for in this case. Since I bought a house and been renovating and moving all my free time I had not much time to make videos but I have plans for the future. As a little sneak peek of what I have in my mind about this specific use of VFD sound I have actually gotten my hands on a special calculator namely a Remington 661•D. This calculator has 6 individual miniature VFD tubes for each digit. If I have the heart to actually salvage a “working” and rather well preserved calculator I want to make something with these tubes. Since they are so small and compact I expect the microphonic tendencies to be much smaller and more suitable for audio and also easier to work with than a display with different sized objects all over the place. This Remington has 6 equal separate triodes. And that is filled with possibilities. As calculator it’s useless anyway because you can’t really type in digits on this model without getting more or less numbers than you want. You can read about this if you so desire lol. But they were like this even new

[@Juancito_el_dulce]

Bien

[@Pulverrostmannen]

Thank you :)

[@creativestudio6844]

Hi i need help

[@Pulverrostmannen]

And what can I do for you?

[@SilverSpoon_]

Hello. You need help. But do you want help?

[@dermevaldennis3094]

Você me explicar em português pôr favor

[@TheRealDorran]

#vfddisplay

[@aficionadofootb]

PLEASE ; SUBTITLES IN SPANISH. THANKS

[@Pulverrostmannen]

jorcuar leo I don’t think I could make proper subtitles for Spanish unfortunately, I think the grammar and everything would be very strange