Author Topic: spark gap latch  (Read 947 times)

0 Members and 1 Guest are viewing this topic.

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
spark gap latch
« on: April 15, 2021, 12:55:00 am »
Is it possible to make a latch/flip flop/register out of two switching spark gaps?
 

Offline Manul

  • Frequent Contributor
  • **
  • Posts: 546
  • Country: lt
Re: spark gap latch
« Reply #1 on: April 15, 2021, 01:13:50 am »
I would say no, because latch/flip flop/register should, by definition, have at least two stable states. I hardly can image two (stable) states of a spark gap.
 

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
Re: spark gap latch
« Reply #2 on: April 15, 2021, 01:28:27 am »
I mean switching spark gap as in it has a base and collector and emitter, just like a transistor,   the base starts the arc, and then the other line will travel over its plasma.

like this picture->

 1 pos    ------------*   spark gap   * -------------  neg 1
                         |                           |                           
                         |                           |
                         |                           |
                        pos                       neg
                         2                           2

so the 1 line hasnt enough voltage to cross the gap.
but the 2 line (the base) does,   and when the base fires,  the 1 line will cross over once the 2 line has started the arc.


I was wondering if u could set up 2 in a latch style,  similar to transistors.
 

Offline Manul

  • Frequent Contributor
  • **
  • Posts: 546
  • Country: lt
Re: spark gap latch
« Reply #3 on: April 15, 2021, 01:55:21 am »
What you describe is close to a thing called triggered spark gap. I think the problem is, that the spark, if it can be made sustainable (using huge power) would not be any more controllable. So it acts like thyratron or SCR. That is not very useful for building logic.

This is quite far from my field of speciality, but somehow my gut says that it is not possible (at least from practical / engineering perspective). Interesting idea, though.
« Last Edit: April 15, 2021, 01:57:38 am by Manul »
 
The following users thanked this post: Capernicus

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
Re: spark gap latch
« Reply #4 on: April 15, 2021, 02:05:45 am »
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16954
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: spark gap latch
« Reply #5 on: April 15, 2021, 04:59:16 am »
Sure, use the two-SCR circuit, the one where they're coupled with a capacitor so one turning on pulls the other off.  The spark gaps must be somehow rated for continuous discharge (arcing) and exact design (minimum holding current, voltage drop) depends on gas pressure and composition.  And probably electrode material and condition (hot electrodes ignite easier, thermionic emission).

Given how unreliable spark gaps are, good luck on getting consistent operation out of it, I suppose.  But it should function, when it does.

A single spark gap also functions as a bit storage, if given a fixed voltage (with current limiting) below the breakdown voltage but above the arc voltage.  It will be bistable: once extinguished, it remains off; once ignited (say by mechanical contact, or a small trigger arc to one of the electrodes), it remains on.

A direct application of this, in two ways, is a welder: One, obviously, when the electrode is at great distance from the work, nothing happens; when ignited, it keeps going, etc.  Two, when HF start is employed, it functions the same as applying a large AC bias to any other hysteretic system -- like the AC bias used for recording audio tapes, which overcomes magnetic hysteresis.  In this case, a high voltage overcomes the arc's hysteresis, making it much easier to use -- no striking needed.

Tim
« Last Edit: April 15, 2021, 05:03:43 am by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: Capernicus

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14433
  • Country: us
Re: spark gap latch
« Reply #6 on: April 15, 2021, 05:22:23 am »
It's possible to make flip-flops and counters with NE-2 lamps so it ought to be possible to do the same thing with a spark gap. Due to the way an arc behaves, as soon as it strikes the gas will ionize and the resistance drops dramatically which will pull down the voltage and prevent another one on the same power supply from striking. I don't see it as being useful but I think it could work.
 
The following users thanked this post: Capernicus

Offline Berni

  • Super Contributor
  • ***
  • Posts: 3486
  • Country: si
Re: spark gap latch
« Reply #7 on: April 15, 2021, 06:00:06 am »
http://www.massmind.org/techref/mem/neonlamp.htm

This is technically spark gap memory, just that the arc happens inside the neon lamp.

Basically the idea is that if you apply a voltage that is lower than the striking voltage but higher than the running voltage it becomes a memory element. Applying the striking voltage writes a 1 into it, while pulling the voltage below running voltage writes a 0 into it. To read the bit out of it you just measure the current trough the neon bulb, if there is current it reads as 1 if no current it reads as 0.
 
The following users thanked this post: boB, Capernicus

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
Re: spark gap latch
« Reply #8 on: April 15, 2021, 06:12:18 am »
goolies coolies.


Ive definitely got to try it,  Ive got all my equipment I need to go see if it works then!

So extinguishing the bit might be reversing the current to deactivate it.

This seems really cool because it acts like a relay! and thats the best form of switch,  all my baby logic designs all worked using relays and this would make them all practical but solid state still.

But I spose I have to try it to find out.
« Last Edit: April 15, 2021, 06:14:18 am by Capernicus »
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14433
  • Country: us
Re: spark gap latch
« Reply #9 on: April 15, 2021, 08:38:08 pm »
http://www.massmind.org/techref/mem/neonlamp.htm

This is technically spark gap memory, just that the arc happens inside the neon lamp.

Basically the idea is that if you apply a voltage that is lower than the striking voltage but higher than the running voltage it becomes a memory element. Applying the striking voltage writes a 1 into it, while pulling the voltage below running voltage writes a 0 into it. To read the bit out of it you just measure the current trough the neon bulb, if there is current it reads as 1 if no current it reads as 0.

It would be interesting to scale that up to something borderline usable, for example a 128*8 bit static RAM, 1,024 NE-2 lamps from China wouldn't cost a whole lot although it would certainly be tedious to wire it all up.

They used to make special trigger tubes that were neon lamps with a third electrode in them that could be used as a trigger, those haven't been made in a long time though. Somewhere I had a digital copy of an old book on neon logic. It was used to make ring counters and oscillators and all sorts of other stuff that made more practical sense back before modern semiconductors.
« Last Edit: April 15, 2021, 08:40:37 pm by james_s »
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16954
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: spark gap latch
« Reply #10 on: April 15, 2021, 10:26:40 pm »
Now I wonder if you can hack a plasma TV into something more than a WOM... :-DD :-DD

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
Re: spark gap latch
« Reply #11 on: April 16, 2021, 12:53:50 am »
http://www.massmind.org/techref/mem/neonlamp.htm

This is technically spark gap memory, just that the arc happens inside the neon lamp.

Basically the idea is that if you apply a voltage that is lower than the striking voltage but higher than the running voltage it becomes a memory element. Applying the striking voltage writes a 1 into it, while pulling the voltage below running voltage writes a 0 into it. To read the bit out of it you just measure the current trough the neon bulb, if there is current it reads as 1 if no current it reads as 0.

It would be interesting to scale that up to something borderline usable, for example a 128*8 bit static RAM, 1,024 NE-2 lamps from China wouldn't cost a whole lot although it would certainly be tedious to wire it all up.
They used to make special trigger tubes that were neon lamps with a third electrode in them that could be used as a trigger, those haven't been made in a long time though. Somewhere I had a digital copy of an old book on neon logic. It was used to make ring counters and oscillators and all sorts of other stuff that made more practical sense back before modern semiconductors.


1k ram can be a bit difficult and a little small to use, but 1024 home made gate asic competes with 1k gate fpgas.   and its more useful in that form.
« Last Edit: April 16, 2021, 01:06:08 am by Capernicus »
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 3486
  • Country: si
Re: spark gap latch
« Reply #12 on: April 16, 2021, 05:10:08 am »
There are "nixie matrix" displays out there too. The one in the photo is a 100x100 russian display.

Tho not sure if you can make the voltage levels work out for the whole matrix and reading a bit might be difficult because that requires sensing the current trough that particular pixel. And i would guess that the striking voltage tolerance for the pixels is too large. But if you could then this would be 10kbits or 1250 bytes of memory.

There is also another form of triggered spark gap commonly used. The xenon photo flash tube. Supply voltage is constantly applied to the ends of the tube while the flash is actually triggered by applying high voltage to a additional trigger electrode (Tho its not really an electrode since its just wire wrapped around the outside of the tube). If you limit the current then the flash tube can probably be kept lit like a neon bulb, might affect the tube lifespan tho since they are not designed to be used in such a continuous fashion.
 
The following users thanked this post: Capernicus

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14433
  • Country: us
Re: spark gap latch
« Reply #13 on: April 16, 2021, 11:14:24 pm »
1k ram can be a bit difficult and a little small to use, but 1024 home made gate asic competes with 1k gate fpgas.   and its more useful in that form.

Depends on what you're using it for. The sound board used in many Gottlieb pinball machines and video games like Qbert is its own 6502 based computer system with 128 bytes (1kbit) of RAM, along with 2-4 kbytes of ROM. That's all it needs. Many much newer microcontrollers have only 128 bytes of RAM, the code runs from flash and the RAM is for holding the stack and variables. A neon lamp based static RAM is not going to be truly useful in the modern world, only a novelty so it doesn't really matter what it's actually doing, just some kind of proof of concept that "does something" is enough.
 
The following users thanked this post: Capernicus

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
Re: spark gap latch
« Reply #14 on: April 17, 2021, 01:50:00 am »
A microprocessor doesn't even need an ALU,   if u can just AND mask the memory together and INVERT, those 2 instructions operating on the ram is all you need.    Making a beefy ASIC means ur giving it heaps of ALU space and sacrificing ram/rom space for it.
So 1kbit of ram equates to 1k of gates, at some ratio between for the finished circuit.

So a microprocessor turns into an ASIC, and the other way round depending on this ratio.

A neon lamp based static RAM is not going to be truly useful in the modern world, only a novelty so it doesn't really matter what it's actually doing, just some kind of proof of concept that "does something" is enough.

Making ur computer properly from scratch means doing it discretely how I want to do it,  I dont mind if its a volatile and has alot of system crashes, and occupying very little estate, thats just adding extra fun element,  only then can I say I made my own computer by myself from the beginning.  I also want to fab all the components myself, buying nothing but raw material.  I even want to make my own sensors, inputs and monitors,  all from scratch.
« Last Edit: April 17, 2021, 02:07:06 am by Capernicus »
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 7301
Re: spark gap latch
« Reply #15 on: April 17, 2021, 04:23:42 am »
There are "nixie matrix" displays out there too. The one in the photo is a 100x100 russian display.
Isn't that basically a monochrome plasma display?
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 3486
  • Country: si
Re: spark gap latch
« Reply #16 on: April 17, 2021, 09:13:28 am »
There are "nixie matrix" displays out there too. The one in the photo is a 100x100 russian display.
Isn't that basically a monochrome plasma display?

Not really sure where the border between these sort of neon displays and plasma is.

What is typically seen as a plasma display is conductive patterned glass with gas chambers and a RGB phosphor pattern to generate the appropriate colors.

These old russian displays are built more like nixie tubes where they have a structure of metal electrodes sitting in a glass box fixed with neon. So then a nixie tube is also a plasma display but with weirdly shaped pixels and a neon bulb is a single pixel plasma display. So i suppose it depends on who you ask. But in any case its neat cool tech from probably somewhere is the 60s to 70s
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16954
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: spark gap latch
« Reply #17 on: April 17, 2021, 10:09:04 am »
There's one display tech that's something like a graphic dekatron, though I forget how multiple glows are maintained in parallel.

It's a pretty slick idea, a 2 or 3 phase clock drives the columns in sequence (the columns are wired together 0, 1, 2, 0, 1, 2, ...).  Rows are lit from the edge, and pixels are clocked in just like you'd do LEDs on shift registers.

There's also something about a hidden discharge which serves as the data shift register, then a full panel power signal or something that brings the glow to the front where it's visible.  Much like strobing a '595 shift register... :)

I can't quite remember or reason out how exactly they work, but they're just that easy to drive; if you don't mind the higher voltages, of course.

AFAIK, they were popular for sign boards, before LEDs took over.  Not sure if they were made with small enough pixels to be useful graphic displays on equipment, probably?

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: Capernicus

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14433
  • Country: us
Re: spark gap latch
« Reply #18 on: April 19, 2021, 11:23:31 pm »
There are "nixie matrix" displays out there too. The one in the photo is a 100x100 russian display.
Isn't that basically a monochrome plasma display?

Not really sure where the border between these sort of neon displays and plasma is.

What is typically seen as a plasma display is conductive patterned glass with gas chambers and a RGB phosphor pattern to generate the appropriate colors.

These old russian displays are built more like nixie tubes where they have a structure of metal electrodes sitting in a glass box fixed with neon. So then a nixie tube is also a plasma display but with weirdly shaped pixels and a neon bulb is a single pixel plasma display. So i suppose it depends on who you ask. But in any case its neat cool tech from probably somewhere is the 60s to 70s

They're both plasma, and both operate on a similar principal. Neon plasma displays achieved some degree of widespread use for a while, I have a Compaq suitcase portable that has a neon plasma display, they were also used in quite a few 90s pinball machines. To make a color plasma display they use a gas that generates UV to excite a phosphor, but for monochrome they can use a gas that generates visible light directly. Same underlying principal though, any glow discharge is plasma.
 

Offline BrokenYugo

  • Regular Contributor
  • *
  • Posts: 161
  • Country: us
Re: spark gap latch
« Reply #19 on: April 19, 2021, 11:51:44 pm »
1k ram can be a bit difficult and a little small to use, but 1024 home made gate asic competes with 1k gate fpgas.   and its more useful in that form.

Depends on what you're using it for. The sound board used in many Gottlieb pinball machines and video games like Qbert is its own 6502 based computer system with 128 bytes (1kbit) of RAM, along with 2-4 kbytes of ROM. That's all it needs. Many much newer microcontrollers have only 128 bytes of RAM, the code runs from flash and the RAM is for holding the stack and variables. A neon lamp based static RAM is not going to be truly useful in the modern world, only a novelty so it doesn't really matter what it's actually doing, just some kind of proof of concept that "does something" is enough.

The Atari 2600 is probably the most famous exame of getting by with 128 bytes, probably the same MOS RIOT chip? But yeah, for a just for fun machine it's plenty.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14433
  • Country: us
Re: spark gap latch
« Reply #20 on: April 22, 2021, 11:33:16 pm »
The Atari 2600 is probably the most famous exame of getting by with 128 bytes, probably the same MOS RIOT chip? But yeah, for a just for fun machine it's plenty.

Yes, it is actually. The Gottlieb sound board is very much like an Atari 2600 but with a latch driving a DAC instead of having a TIA. Some of them also have a Votrax SC-01 speech synthesizer.
 

Offline Renate

  • Super Contributor
  • ***
  • Posts: 1044
  • Country: us
Re: spark gap latch
« Reply #21 on: April 23, 2021, 11:50:46 am »
If you wanted to make a real hack... Make an electronic "Etch-a-Sketch".
1024 NE2 running on barely sub ionization voltage.
Use a servo controlled laser to give them the kick to fire.
Just duck the voltage to erase.
When you've used the display too much, just reverse the polarity to get another half-life.

Hmm, I may just build a proof of concept.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16954
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: spark gap latch
« Reply #22 on: April 23, 2021, 08:14:09 pm »
Or if you don't have a laser handy, a small Tesla coil will do. :)  Also works as a "spray can" if running higher voltage...

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Capernicus

  • Regular Contributor
  • *
  • Posts: 94
  • Country: au
Re: spark gap latch
« Reply #23 on: April 24, 2021, 04:11:30 am »
Dont knock the spray can.   That thing makes beautiful dithered sketches.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf