Getting a 1hz signal from 50hz mains using only logic gates / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Getting a 1hz signal from 50hz mains using only logic gates

previous 1 2 3 4 next

Move Thread

LAN_403

Adam Munich

Thu May 10 2012, 10:31PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

2Spoons

Thu May 10 2012, 10:31PM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Have you looked at the Dekatron counting tubes?

Marko

Fri May 11 2012, 12:17AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Wee, dekatrons seem like fun stuffs. About 8 of them should be enough to build a clock, and I think they should be able to drive nixies directly too!

If you could figure out a way to reset them without using extra vacuum tubes, maybe you could build a not only an all-tube clock but also an all-cold-cathode clock as well!

That German tube clock is ludicrous. And isn't even all tube, I spot some 1N4148 diodes underneath :)

Marko

Download

Fri May 11 2012, 12:56AM

Registered Member #561 Joined: Sat Mar 03 2007, 02:46AM
Location: Adelaide Australia
Posts: 230

Steve Conner wrote ...

Well then, what Boolean function do flip-flops perform?

The internals of a JK flip flop. Ignoring the clock;
R=0 S=0, then Q(t+1)=Q(t)
R=0 S=1, then Q(t+1)=0
R=1 S=0, then Q(t+1)=1
R=1 S=1, then Q(t+1)= not Q(t)

Grenadier wrote ...

Seems I've been beaten :P, at least I know it's possible

2Spoons wrote ...

Have you looked at the Dekatron counting tubes?

I've never even heard of one before, I'll look into it

2Spoons

Fri May 11 2012, 01:19AM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Nice page about dekatrons here

I've just bought two on ebay - for use as a hardrive light on a steampunk pc case. They go for $5 to $10 each.

Download

Fri May 11 2012, 01:48AM

Registered Member #561 Joined: Sat Mar 03 2007, 02:46AM
Location: Adelaide Australia
Posts: 230

Those Dekatrons are quite cool looking, but I think I'll look that them for another project. I'll try to stick to logic gates here

radiotech

Fri May 11 2012, 03:42AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

You want vacuum tubes? OK. Two stage phantastron. divide by ten and divide by five.

Simple: Two pentagrids and two diodes. If you can get it to work it will be the toast of the
town.

Sample circuit. This one divides 100 kHz down to 1 Hz. You just need the last two stages. Just
scale the RC in last stage from 1 to 10, to 1 to 5 .

Caution! These circuits can be voodoo-like ! Uber-Analog.

1336707743 2463 FT138169 Hp522 B Phantastron

Download

Fri May 11 2012, 04:29AM

Registered Member #561 Joined: Sat Mar 03 2007, 02:46AM
Location: Adelaide Australia
Posts: 230

By voodoo you mean a pain the the arse to get to work? :P

Steve Conner

Fri May 11 2012, 06:24AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Surely you're going with an all-tube circuit because you want the pain? :p

I see you said "ignoring the clock", I guess because you were unsure how to describe what the edge-triggered clock does. That is my point: there is no way to describe it because Boolean algebra has no notion of time. It follows that flip-flops do something fundamentally different to plain logic gates. They are "Sequential" logic as opposed to "combinational".

Dr. Slack

Fri May 11 2012, 07:07AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

As the input and output frequencies are very stable, it will be possible to make a divider from an <search term> injection-locked-oscillator </search term>. This is an oscillator set to run at approximately the correct frequency, but is synchronised to a sub-multiple of an input frequency by adding a small portion of it into the oscillator.

Consider an astable Eccles-Jordon ECC83 oscillator, designed to free-run at just below 10Hz. At some point near the end of the cycle, the voltage on one of the grids will have risen to very close to the voltage at which the device will flip. If you now inject a little of the 50Hz, one of the up cycles will push it over the limit early. The point being that one 50Hz cycle earlier, the grid voltage was lower, and the injected 50Hz was *not* enough to make it flip.

Essentially, what this technique does is select one of 5, or N, input pulses to flip the astable. This is what a divide by N digital divider does, using more stuff.

There is a down-side, stability. You have to control the amplitude of the oscillator, the injected signal, and the free-running frequency. It's the ratio between all of these things that selects the effective divider ratio N. Controlling the amplitudes is fairly easy, but a 10Hz oscillator is starting to need large resistors and capacitors which will be subject to surface leakage. The smaller the division ratio, the easier it is to guarrantee the correct division ratio. That's why I'm sort of suggesting that several stages of division, say by 5 and by 10, rather than a single by 50, would be more practical.

An interesting aspect of this arrangement is that by making the first stage go at 10Hz, you could run the clock anywhere in the world, 50 or 60Hz, with *no adjustments*. It would automatically divide by 5 or by 6, because the free-run output frequency of the first stage had been set to ~< 10Hz.

If you are going to fully embrace tubes for this clock, then I would heartily recommend using older slightly analogue techniques like this for parts of it.

previous 1 2 3 4 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob