Diving VFD such as IV-22 etc DC Vs AC Filament drive

[Richard Scales]

Greetings all,

I am starting to look at VFD displays and am reading all I can about them.

I am trying to get a better understanding of the pros and cons of using a AC filament drive rather than an DC one.

I am aware that on larger displays (like those monster ones from Russia) it is necessary in order to create a more even glow across segments. I wonder however how important this is for smaller segments like those in IV-22 for example.

Making the filament drive AC adds a small degree of complexity but if the difference is visible and beneficial then I'd like to go that way.

My project would be direct drive, no multiplexing required.

Any advice would be greatly appreciated.

- Richard

[Adam Jacobs]

You'll be fine with DC. If you can't see a visible difference in brightness across the tube then it probably doesn't matter. People say there are lifespan issues with DC but I haven't seen that yet on my VFD clocks. I just use DC and bias the grid ~ -3vdc.

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/37248fe9-3f82-4d36-be12-bdba154e7364n%40googlegroups.com.

[Christian Riise Wagner]

The IV-22 has very low filament voltage. Something like 1.2V IIRC. That makes the voltage gradient across the display face low and I didn't notice a brightness gradient while running one with DC. Used a simple buck converter for the filament and a boost converter for the anodes and grid.

[Paul Andrews]

Assuming you drive the filaments in parallel rather than in series, a DC voltage would be fine. If you drive them in series then (I assume) the voltage gradient might be visible. A decision to drive in parallel has a knock-on effect - I couldn't come up with an AC drive that had a constant voltage with varying load. With filaments in parallel, the load will change if one or more filaments burn out. You don't want a filament burn out to increase the voltage across the remaining tubes, otherwise they might burn out too - so a regulated DC voltage is easiest. With a non-regulated AC supply, you will also get a higher voltage at startup when the filament is cold. Another reason to go with a regulated DC supply. Which brings me to the potential issue of in-rush current. Either a constant current setup or a current-limiting resistor in series with the filament would be a good idea. The latter also helps with achieving a voltage offset to the grid - i.e. making the grid negative with respect to the filament.

[gregebert]

Adding series resistors also reduces the current spike during power-on. Hard to say what fails first with VFD's; ones that I've seen such as the displays on kitchen appliances which run 24/7 show phosphor degradation after a few years. If you frequently power-cycle the filaments, then you definitely want to take steps to reduce the current spike.

Be sure to include a PIR sensor so you can shut off the segments when it's not in-use; your decision to keep filaments on or off.

For my NIMO clock, I shut off the anode supply after 2 minutes of no PIR activity to reduce phosphor burn, and the filaments go off after 12 hours of no activity.

[Moses]

Richard,

Same here! I started to experiment with the IV-21 tube and how to drive it recently. I have some drivers on the way. I don't have much experience with VFDs at the moment.. but I did light one up on the bench recently.

I used the datasheet recommended filament voltage/current on the IV-21 of about 2.7v at 35ma, and I lit up all the segments/girds manually without a driver IC. With everything on, I was able to perceive a difference in the brightness but I had to be looking for it, it was not obvious. Now in operation as a clock, as opposed to a static display with all segments on, I would expect it to be less noticeable.

The IV-21 is one of those 8 digit small calculator type tubes, the filament is about 40cm long or so. I believe the IV-22 single digit tube would have a shorter filament, and hence less of an effect? Maybe.

Keep us updated on your progress.

Regards,

-Moses

[Richard Scales]

Many thanks for all your replies and advice.

DC seems t be the way then, with some series resistance on the filament supply.

I'll start in that direction and get something working on the bench.

 - Richard

[David Pye]

For the longer tubes where the brightness gradient is problematic, the usual DC-DC converter to get the voltage followed by an H-bridge to generate an alternating polarity square wave is another option.

David

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/20b78aec-a35e-4d5f-a677-921eaa14ca0fn%40googlegroups.com.

[Benoit Tourret]

in eastern europe, they are often using a handmade transformer driven by a sn6501

the ferromagnetic core receives windings of two to forty turns with wire of 0.1, 0.2 or 0.4mm

see on PCBWAY project of Konstantin Repnikov from BELARUS.

I will try one day, I will try.... I have the core and the wire...

In my youth, I was a redhead. can you imagine the hair? oh well it's the same...

[Moses]

Richard,

An update for you on my experience. I got VFD drivers (max6921) and built up my first prototype using an IV-21 tube.

I cannot see any different from running it on DC while running the *RECOMMENDED* filament current. I'm running the filament at a recommended 2.65 volts on DC giving me 35mA (through 5v using a series resistor). Grid and anode voltages are at 35v from the recommended 27-50v of the datasheet.

*NOTE: I initially had the filament driven at around 2.3-2.4 volts and 30mA, this did cause a noticeable brightness gradient from one side of the tube to the other, not too bad though. I swapped polarity and the gradient moved. But by dropping the series resistor and giving the filament another 0.25v / 5ma or so I can stare at it all day and I can't see a difference. I'm surprised at the difference such small voltage/current made.

Picture attached. Hope this helps.

Regards,

-Moses