I cant find where to buy spark gap "circuit interrupters"

[ratstar]

Maybe my laptop is too slow, or I'm hopeless at the internet, but I cant find anywhere to buy spark gaps for doing clocked logic with, I'd like something that just has a negative and a positive and a gap of air in the middle, with very little capacitance.

I do have a cool way of making them myself, just two pieces of copper tape with a rizla (rolley cigarette paper, or even a piece of bible paper would work) "frame" inbetween gets a nice tight 25 micrometres. I could even make it semi-permanent with a bit of thick plastic top and bottom, but I think buying them is probably better.

Also, my electronics stores near I live don't have them listed in their stocks. Can anyone please post a link to something if they would be so kind?

Theres nothing in Australia, I'm going to have to order over-seas I'm pretty sure.

 

[bertus]

Hello,

There are sparkgaps at Mouser.
The voltages are from 60 Volts and higher:
https://au.mouser.com/Circuit-Prote...-Arrestors/_/N-5g3f?Keyword=spark+gap&FS=True

Bertus

 

[ratstar]

Thanks, I could never have found that...

Woah 60 volts, pretty good, I want to go as low as I can.
With a 60 volter I probably only need a 3 capacitor marx generator, cause my novice power supply goes up to 30 volts. (I have to count for losses, 2 caps for my system is more like 1.7x, not 2 x)

The lower the volts, the less Q loss, im pretty sure.

 

[Audioguru]

Spark gap oscillators and sequencing circuits were used 97 years ago. For the last 30 years we use transistors and ICs.

 

[ratstar]

Spark gap oscillators and sequencing circuits were used 97 years ago. For the last 30 years we use transistors and ICs.

Is there any reason why a spark gap machine can't do as good if not better than a transistor machine?

If I were to make all my components from scratch, it would be easier if I didn't use transistors because its one less thing on the list and I'm also not sure how to make them,

Recently though, I saw on you tube how to make copper diodes by sticking copper tape in an oven/kiln at 400 degrees and it comes out with photo voltaic properties, not so good for energy harvesting but it would be a cool way to make a camera sensor.

One thing I want to make a super ultra high speed camera, and put hardware computer vision logic into it. (I was an AI programmer before I got into hardware.)

 

[Harald Kapp]

Is there any reason why a spark gap machine can't do as good if not better than a transistor machine?

A spark gap is a comparatively crude device with not very closely controlled parameters.
A spark gap has no gain.

Recently though, I saw on you tube how to make copper diodes by sticking copper tape in an oven/kiln at 400 degrees and it comes out with photo voltaic properties, not so good for energy harvesting but it would be a cool way to make a camera sensor.

What would be the size of the sensor with millions of these homebrew diodes?
Besides: There's much more to a camera sensor than just a bunch of photodiodes. You need heaps of amplifiers and logic circuits to convert the signals of the millions of individual sensors to a useable digital format.

One thing I want to make a super ultra high speed camera, and put hardware computer vision logic into it.

This is not going to happen anytime in the near future with your current knowledge and your attitude towards established technologies. I'll bet on it.

 

[ratstar]

A spark gap ha no gain.

I know u dont like it... ... but being able to convert from parallel to series and back again like a marx generator forwards and backwards with capacitors gives you amplification abilities.

 

[ratstar]

heres a schematic of it.
the extra spark gaps can be diodes if you want, they are to stop the capacitors prematurely discharging, but it sorta spoils it adding pn junctions.

As you can see its the reverse of a marx generator, charge series, discharge parallel.
[edit] i think the circuit is a bit wrong, it needs to be fixed. but the general idea is there. [/edit]

About the sensor, you probably could get 512x512 in 50x50 cm, but its pretty tight. Its like an insects compound eye.

 

[Harald Kapp]

What you show is not amplification as we usually designate it. We're back to your not using the terms of the trade as (almost) everyone else does. Or using the terms in a non-conventional (to use a benign expression) way. Thus you create misunderstanding at the least.
Amplification in the sense usually used is the effect of inputting a small signal and outputting a signal at the same waveform but with a higher amplitude (e.g. higher voltage, higher current, or both). Think of an audio amplifier.

I am not going to analyze your circuit in detail, but if you were to input a sine wave at the "base", you're not going to get a sine wave with higher amplitude (voltage) at your so called "emitter". The Fidelity of the input signal will by no means be preserved.
This circuit will not work as a transistor.

 

[ratstar]

What you show is not amplification as we usually designate it. We're back to your not using the terms of the trade as (almost) everyone else does. Or using the terms in a non-conventional (to use a benign expression) way. Thus you create misunderstanding at the least.
Amplification in the sense usually used is the effect of inputting a small signal and outputting a signal at the same waveform but with a higher amplitude (e.g. higher voltage, higher current, or both). Think of an audio amplifier.

I am not going to analyze your circuit in detail, but if you were to input a sine wave at the "base", you're not going to get a sine wave with higher amplitude (voltage) at your so called "emitter". The Fidelity of the input signal will by no means be preserved.
This circuit will not work as a transistor.

Yes you are correct, it works best with binary signals, but I bet if you teamed up 8 of them with an adc going into them, you could bring back an analogue signal composed of so many binary amplifications. But it definitely responds off with off, and on with on, but maybe halfway between wont work so well, and the system runs off a spark gap, its not continuous, it needs to be oscillated at the nyquist.

 

[Nanren888]

Neons?

 

[ratstar]

Neons?

I've never used a neon how much volts do they need?

I'm keeping the system as low voltage as possible until I get to the electrostatic actuators, then I might get to 3000 volts max, only because I have to, I'd rather not put myself in danger, but the other thing other than transistors I'm avoiding using is magnets, (for motors) I want the whole thing iron free.

A funny thing was happening because my spark gaps were just floppy things was I was actually getting contractions on them, and they were closing on me, I think it was 1000 or so volts, so I wasn't just seeing things, I think. So ESA's are on the way already.

 

[bertus]

Hello,

Neon will need about 90 volts to ignite.
From the wiki:
Voltage tester

+DC (left), -DC (center), AC (right) supplied to NE-2 type neon lamps
Most small neon (indicator-sized) lamps, such as the common NE-2, have a break-down voltage of around 90 volts. When driven from a DC source, only the negatively charged electrode (cathode) will glow. When driven from an AC source, both electrodes will glow (each during alternate half cycles). These attributes make neon bulbs (with series resistors) a convenient low-cost voltage tester. By examining which electrode is glowing they can reveal whether a given voltage source is AC or DC, and if DC, the polarity of the points being tested.

https://en.wikipedia.org/wiki/Neon_lamp

Bertus

 

[Martaine2005]

I was an AI programmer before I got into hardware

I now believe in reincarnation

Martin

 

[Nanren888]

runs off a spark gap,

No idea what you mean.

Just an aside. Just a comment. Don't let it stop you, but, ...
I see that multiple people are happy to help with answers to how components and circuits work or will work. Sites like this attract such useful people.
.
Some other discussions from time to time are about people wanting perpetual motion or free energy, or similar and can be identified early, so answers can be given, but more importantly, the impossible nature of the overall aim can be explained, or an approach that is never going to work can be shown and alternatives suggested.
.
Discovery is great. Exploration is great.
Spark gaps? Neons are nonlinear and basically spark gaps. They are also cheap.
No magnets.
No iron.
No transistors.
A "Computer".
ADCs from spark gaps.
A preference for low voltage.
An emphasis on energy efficiency, yet spark gaps, which must be generally one of the least energy efficient things I can think of in normal use, being usually a protective device to dissipate excess energy and then only for higher voltage. Lower voltage nonlinear devices do exist.
.
Can't tell where you are going. Can't tell whether you are trying to go somewhere that physics says one cannot go, in part or in total aim. Can't offer out of the box suggestions on how to get there, as don't know where you are trying to get to. You have a right to keep things close to your chest. Your ideas, aims, IP.
.
Can continue to answer questions on individual components and individual circuit configurations. Enjoy.

 

[Nanren888]

PS:

Transformers are also amplifiers. (sort of)

 

[ratstar]

PS:

Transformers are also amplifiers. (sort of)

I mean by amplification as an increase in power whilst maintaining the on off nature of a signal that's being amplified.

Not just changing amps for volts... im not totally crazy.

 

[Nanren888]

capacitors gives you amplification abilities.

"Power"? Maybe, but energy-wise you can only get out what you put into passive components.

What you show is not amplification as we usually designate it.

Anyway, happy to keep guessing. Maybe we will eventually chance upon something useful to you. Enjoy.

 

[ratstar]

Say I have 3 capacitors in series, but the 1st capacitor is either charged or uncharged before hand, the other 2 capacitors are empty. (Thats the thing everyone always misses.)

If the 1st capacitor is charged, no current can go though, and the 2 other capacitors remain empty, but if it isn't, all 3 caps will charge, the 2 capacitors will charge from the single capacitor being what decides them to charge.

Then you have 2 capacitors charged from 1 capacitor, an increase in power.
But note its not the capacitor thats charging the 2 other capacitors... its the battery.

 

[Nanren888]

"Power" is energy over time. You are not making energy, so you can use that energy at any rate you can manage, but net energy is just the same, unless you are getting more from somewhere.
Capacitors store charge, at some voltage. They store energy. You can put it in & take it out, but it does not increase.
If you put it in slowly then the power is low, energy = power by time.
If you take it out quickly, then yes, higher instantaneous power than you had putting it in, but the same energy. The resulting power * time (integrals for changing quantities) will be the same.
.
Maybe you understand this. Might just be the difference language usage others have mentioned.