CallumNunesVaz
commented
7 years ago Hi Joel,
I’m glad to see someone else doing one of these builds!
Yes its been a while since I put together the first prototype, but I do get fairly frequent emails from people asking questions or after the BOM so some stuff is still fresh in my mind.
I was using a timer to create delays between calls of the functions that drive the anodes via the MAX6920, and a separate timer (this one was tied to an interrupt) to drive the H-bridge. You could definitely use the same timer if your code that runs the anode drivers is small enough to reliably exist inside an interrupt service routine, but because mine was so long I wasn’t comfortable doing it that way and instead made use of the extra timers available.
In this “A Guide to Fundamental VFD Operation” (https://www.noritake-elec.com/technology/general-technical-information/vfd-operation), they specify timing values where the anode frequency is roughly double that of the cathode frequency, I reasoned that their usage of 50Hz and 60Hz was a simplification by driving the VFD filament via a step-down of mains AC, and that the filament should theoretically be able to go higher. Also I didn’t experience the flickering that they mentioned. Additionally, the code holds some legacy drivel, due to the fact that I was testing timing values about the time that I decided there were too many improvements that needed to be made to justify finishing the code, so that would be why the anode frequency is 4x the cathode frequency. As for your concern, the cathode filaments stretch across all four digits and the dots, so you don’t need to know which
I'm currently finishing up at uni and will be finalising my final year project over the summer (here in Australia), after which i will be finishing the second revision and hopefully be doing small batch orders of 'kits', since ordering single parts is so expensive. I mention this because I’ve identified a number of things I wanted to change in the new design…
- I have been going back and forth on whether brightness control is necessary or suitable. I may not implement it in the second revision as a result. The problem being that the VFD filament takes some time to cool up and down, and so PWM control of it is not particularly useful at the frequencies required for a nice display. Since the filament uses most of the power, it would mean that brightness control would not impact battery life (the original intention) and I personally would then keep it at full brightness all the time, rendering it a piece of redundant code.
- I've also been considering the replacing the MAX6920 with a MAX6921 so I can use a TSSOP instead of a SOIC package and drive the dots independently. I also wanted to replace the mini-usb port with type-c port. This would allow the display to sit closer to the board. I would heavily consider this if you’re going to have the parts sit underneath the display, and want it to be as thin as possible.
- I will be replacing the linear regulator for the filament with a inverting buck converter, so that I can get about ~90% efficiency and drive the filament at -2.4V. This allows me to set the LMR62421 to approximately 15V instead of 20V with no loss is brightness over the first revision, but the LMR62421 will be able to run with much more efficiency.
- … As well as all the changes listed at the bottom of the page for the watch…
Goodluck with your project! Let me know if you have any other questions.
Cheers,
Callum
From: Joel Murphy [mailto:notifications@github.com] Sent: Thursday, 16 November 2017 9:48 AM To: CallumNunesVaz/DogBoneVFD DogBoneVFD@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: [CallumNunesVaz/DogBoneVFD] Timers and H-bridge (#1)
Hey, it's been a while since you touched this. I'm building my own VFD display using the same old/new nixie thing that you did. I have a question about your use of timers.
You used an internal hardware timer to drive the H bridges which you used to even the display brightness. I want to use an independent oscillator driven circuit to power the H bridges. It's a way to accomodate less pins, and also I like oscillators...
Are you using the H bridge timer interrupts to time the display multiplexing? I'm wondering because I think I can follow the relaxation oscillator that I have with an interrupt and time my muxer that way. Hope to hear feedback from you on my scheme.
Also, in your codehttps://github.com/CallumNunesVaz/DogBoneVFD/blob/master/main.c line 71-73 you say you mux the digit display at 1Khz, but you are driving your H bridge at 250Hz. What's the logic behind that? My concern is that I will need to know when which element is 'lit' by the 2.5V to print my numbers. Does that make sense?
My current schematic and board layout can be found here https://github.com/biomurph/ILV2-7-5_VFD_Display
Great work, btw
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/CallumNunesVaz/DogBoneVFD/issues/1, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AMMYMNdO8HPYBm24WOEhmlGCZ0fVi3byks5s23E8gaJpZM4QfuVP.
biomurph
Hey, it's been a while since you touched this. I'm building my own VFD display using the same old/new nixie thing that you did. I have a question about your use of timers.
You used an internal hardware timer to drive the H bridges which you used to even the display brightness. I want to use an independent oscillator driven circuit to power the H bridges. It's a way to accomodate less pins, and also I like oscillators...
Are you using the H bridge timer interrupts to time the display multiplexing? I'm wondering because I think I can follow the relaxation oscillator that I have with an interrupt and time my muxer that way. Hope to hear feedback from you on my scheme.
Also, in your code line 71-73 you say you mux the digit display at 1Khz, but you are driving your H bridge at 250Hz. What's the logic behind that? My concern is that I will need to know when which element is 'lit' by the 2.5V to print my numbers. Does that make sense?
My current schematic and board layout can be found here https://github.com/biomurph/ILV2-7-5_VFD_Display
Great work, btw