NL-50944 Infos?

[SWISSNIXIE - Jonathan F.]

Hi folks,

I recently aquired a NL-50944 tube. Its the size of a B-5092. On first look it looks like a tube that can display two digits with a few possible combinations, but now that i've tested the tube i found out that the digits are NOT independant. They always light up together

There is 00,25,50,75

My guess is that it was used to display a percentage of a progess - or - to display a value that has only .25 precision.


Does anyone can think about a usage of this oddly tube? Or maybe even know where it was used. Its very interesting that the desing created the need to make a custom tube instead of just using two regular 0-9 tubes side by side!

The only resource on the web to this number is on Dieters site, but only the information that this tubes uses the B13B socket, no other info.

[Terry S]

My guess would be as the last tube in a scale display -- 138.25 grams or oz.

[jb-electronics]

Nice observation, Jonathan. Maybe it's the Americanized version of the GR4G ;-) http://www.tube-tester.com/sites/nixie/data/gr4g/gr4g.htm

Cheers
Jens

--
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 post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/f6e25c1f-8d6f-4f79-934a-24a3433cd5aa%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Alic]

Wow, nice find!
I’m looking forward to seeing the photos on your website!

[Terry Kennedy]

On Thursday, May 17, 2018 at 12:44:31 PM UTC-4, Jens Boos wrote:

Nice observation, Jonathan. Maybe it's the Americanized version of the GR4G ;-) http://www.tube-tester.com/sites/nixie/data/gr4g/gr4g.htm

That tube must hold the record for "most pins to display the fewest things"! I'm somewhat surprised that they populated all of the pins in the tube base instead of just using enough for positive keying - when I worked in the glass base business, we would have omitted those in the pellet pressing process in order to have a lower chance of leaks after sintering and assembly + flow in the hydrogen furnaces.

It looks like there are additional digits in the OP's 2-digit tube (possibly 5 per position, hard to tell from the picture). If that is the case, do they seem to be connected to any pins?

[Tomasz Kowalczyk]

I suppose there are two reasons why so many pins are present. One is locating the right position of the tube - if you skip most of the pins, then orientation can be located only by judging where the front of the tube is. If there is only one way to put something in something, then the risk of something wrong happening is minimized.

Also, if the socket has all 27 slots present, but the tube doesn't, someone might thing his tube is broken, as it has less pins. 

The other reason is that as far as I know the stems weren't produced by the same companies, or at least not in same factories, so ordering unique bases was more problematic than just purchasing same type as for the other tubes. 

Similar thing goes in signal tubes - many tubes do not utilize all pins there are present, but to keep with the standard sockets and to provide good grip all pins are produced anyway. For example, there are many single triodes with noval base - and those require only 5 pins out of 9.

Voltage stabilizers use extra pins just for rigidness of internal structure - they are essentially neon tubes, like nixies, but have larger working areas to support currents varying between 5mA and 30-40mA. So their model has only two electrodes, but as there are more present in the envelope, then why not use them as extra mechanical support.

[GastonP]

Actually, in most if not all of the voltage regulator/voltage reference tubes, at least a pair of the pins are used as security feature. This is just a short circuit between two pins that are used to disconnect the output of the power supply when the regulator is not in the socket to avoid unregulated high voltage going into the load when the regulator/reference is not in.
It is interesting to see that very few of the new circuits that use this voltage regulators do not make use of this security feature.

[JohnK]

Great value in that reminder Gaston; well done.

 

Another regulator arrangement that needs great care is this.

I have seen instances where the shunt regulator is designed to deal with the expected change in current in the load.

For instance the load might always draw at least 70mA and never more than 100mA.

Should the load go open-circuit [or be unplugged] then the whole 100mA has to flow in the regulator which in this case

was designed to only cope with 30mA plus a minimum current that allows regulation to occur.

[Back in the 1970s I saw a zener supply let out the magic smoke.]

 

OT:  but might apply to valve clocks. I have also seen screen bypass capacitors rated for the screen voltage and not for the usually higher rail voltage.

When the valve is removed from the socket [or goes opencircuit] there is supposedly no DC in the capcitor so the rail voltage appears across it.

In practice many capacitors survive because electrolytics are notorious for conducting. [Obviously there are safety margins in ratings etc too.]

 

John Kaesehagen

 

--

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 post to this group, send email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/00e6bda1-b54b-43d4-b347-889a28fb735e%40googlegroups.com.

[Terry Kennedy]

On Friday, May 18, 2018 at 2:02:32 AM UTC-4, Tomasz Kowalczyk wrote:

I suppose there are two reasons why so many pins are present. One is locating the right position of the tube - if you skip most of the pins, then orientation can be located only by judging where the front of the tube is. If there is only one way to put something in something, then the risk of something wrong happening is minimized.

In the referenced tube, the entire inner ring of pins could be omitted without losing positive keying, and none of the inner ring pins are used on that tube.

 

Also, if the socket has all 27 slots present, but the tube doesn't, someone might thing his tube is broken, as it has less pins. 

I doubt this was much of an issue in the original application, as assembly (and most likely repair) would be done by people familiar with the product.

 

The other reason is that as far as I know the stems weren't produced by the same companies, or at least not in same factories, so ordering unique bases was more problematic than just purchasing same type as for the other tubes. 

This is more likely. Economy of scale may mean that it would be more expensive to omit the pins. And if the bases were made by a different company, the rejected part ratio was not the tube company's problem, as long as rejects were found by the supplier before the tube company assembled the final product, and the supplier's prices remained low enough. That can sometimes be problematic - I was called in to fix a disk drive design problem in the 80's where the drive manufacturer was having a 90%+ incoming reject rate on outsourced hybrid modules which were made by one of the most well-known companies in that field.

Back on tubes, the omitted-pin business seems to be more common on tubes where a) the base was not part of the evacuated system and b) there was already some positive keying method (such as octal base) in place. It was very common for TV picture tubes to not have all pins populated, for example.

At the glass company I worked at, most production was for relatively high-cost / low-volume products. One big exception was a part (something like BC1237 - I forget the exact number) which was the base + lead assembly we made for Motorola, used for a 3.58MHz crystal in TV sets. Those went out in orders of 10's of thousands at a time. We did all sorts of esoteric stuff, like crystal housings where everything but the lead wires and the base / cap seal ring were clear glass. That was for a military application.

[John Rehwinkel]

> At the glass company I worked at, most production was for relatively high-cost / low-volume products. One big exception was a part (something like BC1237 - I forget the exact number) which was the base + lead assembly we made for Motorola, used for a 3.58MHz crystal in TV sets.

That's an interesting part: it's a specialty item with an assortment of specific requirements, but made in huge quantities.

> We did all sorts of esoteric stuff, like crystal housings where everything but the lead wires and the base / cap seal ring were clear glass. That was for a military application.

I'll bet those were pretty! I have a few old glass-cased crystals, but they look like miniature tubes.

- John

[5-ht]

FYI, crystals in glass cases are still being manufactured.
Total Frequency Control in the UK list some in the H/C-26U type case.

See: http://www.tfc.co.uk/pdfs/HC-26-U_TFC.pdf

( I have no connection to this company.)

Mark

[Tomasz Kowalczyk]

W dniu piątek, 18 maja 2018 15:50:41 UTC+2 użytkownik GastonP napisał:

Actually, in most if not all of the voltage regulator/voltage reference tubes, at least a pair of the pins are used as security feature. This is just a short circuit between two pins that are used to disconnect the output of the power supply when the regulator is not in the socket to avoid unregulated high voltage going into the load when the regulator/reference is not in.
It is interesting to see that very few of the new circuits that use this voltage regulators do not make use of this security feature.

That's really simple and clever! I haven't thought about that. It seems soviet engineers did not, too, I have a piece (and have seen more pieces) from soviet military equipment and they connect only one electrode or wire up all electrodes of same type together. 

John, valve amplifiers produced today fortunately implement capacitors able to survive all voltages present in circuit. At least last 10 or so designs I was looking into (with plan to build each one of them) had properly thought through voltage requirements.

Which, by the way, brings one thing to my mind: many of nixie power supplies use SMD components. But how many of them take voltage rating of SMD resistors into account? I think I've seen few designs using 0805 resistors in feedback circuit, which (almost all of them) have only 150VDC rating, while the upper resistor in feedback divider will be seeing 170VDC or more all the time. Using two resistors in series would fix the problem, but first, the problem has to be noticed.

On TME I could find only two resistors smaller than 1206 which are rated for voltages present in nixie PSU.

 

 

[Paul Andrews]

Speaking for the two power supplies I have actual experience of: My own and the Taylor HVPS. Yes they both take account of the voltage rating of resistors.