B7971 questions

[newxito]

In 2016 I bought some B7971 to build a clock, not the best tubes for a beginner with no skills in electronics. Until now I have not even tested these tubes. But now, after building some clocks with numeric nixies I think I'm ready to design a board for a 6-digit clock.

I know that these tubes are hard to find a very expensive these days, so I don't want to mess it up.

The design will have a fuse for the 12V, another fuse for the 170V, reverse, over and under voltage protection and a temperature sensor for the HV PSU. Just to be on the safe side :-)

I found the datasheet and a lot of information on the internet. I know now, that I should use different cathode resistors for the different segment lengths, but I still have questions about the max. current limit (21mA):

- I don't want to go to the limit, 18mA max. Is it ok to choose the resistors to be able to light up all 15 segments at the same time? (average 1.2 mA/seg). Or is it better to drive the segments with a bit more current (2mA) and limit in the software the number of concurrent segments to 9? (0 has 8 segments + underscore according to the font in the datasheet)

- Do I need some kind of anode current limiter?

[Kevin A.]

I would start by taking a look at the "Mod Six" 7971 clock by the team over at badnixie.com. This has pretty much been the clock for 7971s, as it is loaded with features and is by all means a professional quality product.

http://www.badnixie.com/MOD-SIX_Info_Page.html

Check out the assembly guide for schematics of all the circuits. You will get a good sense of the general topology required to drive the tubes. 

You can also watch a video I made of assembling my own mod six here: https://youtu.be/jd8zp5yjdWA

This might give you a better tangible representation of what the circuits look like and how they are combined. 

--
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/8940d162-56fd-4403-a954-b4b6dd03766a%40googlegroups.com.

[Paul Andrews]

I used the resistor values listed here. Estimate a maintenance voltage of, say 140V. Then the total current used, if all segments are lit using those values is < 21mA. The brightness of the segments when using these values seems to match those of other displays I have seen.

In other words, pick some total current < max and then distribute that current between all cathodes such that the ratios between them stay the same as the maximum values in the datasheet.

[newxito]

The Smart Socket and the MOD_SIX use the same resistor values and as far I can see they don't have an anode current limiter.

They can not be wrong, so I will start testing using those resistor values.

Thanks Kevin, thanks Paul

[gregebert]

I designed and built my own 8-tube 7971 clock; it's been running over 2 years now and the only tube failure was caused by a segment moving in the tube and contacting another one. Nothing has failed due to the actual design of the clock.

So, yes, the datasheet specifies a maximum current for all segments, and you should not exceed that.

As you probably noticed, there is a spec for the various segment currents because there are several different lengths, hence different operating currents. If you add-up the total current for all segments at their rated maximum, you will draw about 40mA, which is almost twice the spec limit.

I did some spreadsheet analysis of how much current each displayable character consumes, and not surprisingly most of them fail the 21mA limit.

What I did was to scale-back the individual segment currents so that most characters are below 21mA.

For the few that would otherwise exceed a total of 21mA, there is a second current regulator on the anode set to 21mA. For those characters, the current gets shared so each segment gets slightly dimmer.

For each tube, I plotted I-V curves and found that the observable brightness does not vary significantly between 50% to 100% of rated current. I also used this data to confirm tubes have similar characteristics.

There is potentially some risk that cathode poisoning will occur at reduced current, but I have seen no hint of it. On the other hand, I definitely see it with IN-18's but that's a different story....

I gave away my extra PC boards; I dont know if the folks on neonixie who have these boards have built theirs yet.

I like the fact that these tubes are enormous and you can display text, but I have to say they are quite ugly. Nothing compares to the beauty of a fully-figured traditional nixie.

[Terry S]

I like the fact that these tubes are enormous and you can display text, but I have to say they are quite ugly. Nothing compares to the beauty of a fully-figured traditional nixie.

Ahhh thank you for saying that. I concur.  Still impressively large tubes.

[Nicholas Stock]

I’m going to (respectfully) disagree here. I used to be of the same opinion (when they were 10 dollars a tube) and hence didn’t think they were a real nixie tube. I now regret that stance. They’re fascinating devices with an apparent ‘infinite’ life span (still haven’t had one die of ‘natural’ causes). 

Sent from my iPhone

On Aug 9, 2019, at 18:24, Terry S <tschw...@gmail.com> wrote:

I like the fact that these tubes are enormous and you can display text, but I have to say they are quite ugly. Nothing compares to the beauty of a fully-figured traditional nixie.

Ahhh thank you for saying that. I concur.  Still impressively large tubes.

--

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/8855b0e1-cb8a-41d6-b52a-5dbb5564e33c%40googlegroups.com.

[Richard Scales]

I am fully prepared to offer a loving and caring home to all ugly B-7971s!

[newxito]

After 3 years I finally managed to test the tubes. The good news: 7 seem to be ok so I can start designing the board. The bad news: 1 has a broken connection on segment 2. I’ve applied some heat to the pin, but I think there is no way to fix it. I got 3 tubes with Burroughs label, 4 with Ultronic label and 1 with no label, quite a mix…

[gregebert]

If you have a capacitance meter, you can compare capacitance readings on this pin vs the same on good tubes.

If the values differ significantly, it suggests the break is closer towards the pin.

I suspect it's somewhere with the connections to/from the internal 'circuit board'. Tapping the tube is risky because it could cause segments to short against eachother (I have a tube like this).

I also have a tube with a dead segment, and nothing is obviously wrong from a visual inspection. Gentle tapping did not help, and cranking the voltage way up did not produce any glow.

Since other segments are good, this failure is not a gas-leak. It's unlikely to be cathode poisoning, either. It's still a very useful tube so take care of it.