Hi Guys,
Was this the general idea?

ps. I assume there was some gizmos between the supply and the coil because
the frequency for the coil may not have been intended to be the mains frequency.

 

there was a spark gap transmitter to feed the tower. there were also ultraviolet searchlights on the top of the tower. there was a very long grounding structure under the tower. not sure if the spark gap transmitter was of the quenched, rotary, or other type.

 

Was this the general idea?

Probably not. The gap between the ionosphere and earth could possibly be seen as a capacitance but not as an inductance.

By the way, your original post could do with the addition of some background information for context. Not everyone will know you're referring to the lab, experiments or theories of Nikolai Tesla.

 

the tower was supposed to excite the schuman cavity aprox 8 hz. between the earth and ionosphere.

 

By the way, your original post could do with the addition of some background information for context. Not everyone will know you're referring to the lab, experiments or theories of Nikolai Tesla.

I don't want replies in that case.

the tower was supposed to excite the schuman cavity aprox 8 hz. between the earth and ionosphere.

That sounds familiar, I'll have to look into it.

I'm not trying to reproduce anything of this scale, just to understand.
One day, I would like to produce a Tesla "display" coil though.

 

At 8 Hz was the Wardenclyffe power transmitter spark gap driven or was it alternator driven?

To me for that low of resonant frequency it would make more sense to have driven it via a massive low frequency alternator system.

 

Especially given Tesla's love of rotating/alternating mechanisms and desire to tune resonant frequencies.

 

I suppose a solid state, or even vacuum tube osc were out of the question back then
I heard he used his mechanical osc (earthquake machine) for many experiments,
as there wouldn't have been too many options.
Apparently that machine just vibrated a coil with the machines vibration.
It stands to reason that either a fixed magnet with floating coil would provide an oscillator.

 

Yeah he liked mechanical systems too. So the mechanical coil oscillator has a freq dependent on simple mechanical tuning like tweaking the spring tension and contact position etc.

 

That would be a funny clock source for a micro circuit RB
a hundred times the size of the micro circuit itself!

Coming to think of it you could probably make a pickup for 50Hz from mains that way,
if you were doing something that could run slowly like a timer.

 

Funny enough I have wanted to do a "50 Hz clocked PIC" for years now.

With full wave rectified mains that gives 100Hz for the PIC external clock, so it equals 25 PIC instructions per second execution speed.

I'm sure a workable HH:MM clock can be built at that speed, but it would take some fancy coding to make a HH:MM:SS clock with only 25 instructions per second. An interesting challenge for sure.

 

Funny enough I have wanted to do a "50 Hz clocked PIC" for years now.

With full wave rectified mains that gives 100Hz for the PIC external clock, so it equals 25 PIC instructions per second execution speed.

I'm sure a workable HH:MM clock can be built at that speed, but it would take some fancy coding to make a HH:MM:SS clock with only 25 instructions per second. An interesting challenge for sure.

the most pervert clocking i have done so far was to gate off from a dc/dc booster IC with a 12k resistor. Running a small LED chain, I never observed variation of the speed.

 

I don't know if that's possible even with seven segment drivers RB,
the change from 11:59 to 12:00.. you have to do the lookup for every digit.
The only way around it I can see is a lookup table of huge memory that
contains all digits for every possible time of day... then the EEPROM read time is another problem.

Funny enough I have wanted to do a "50 Hz clocked PIC" for years now.

With full wave rectified mains that gives 100Hz for the PIC external clock, so it equals 25 PIC instructions per second execution speed.

I'm sure a workable HH:MM clock can be built at that speed, but it would take some fancy coding to make a HH:MM:SS clock with only 25 instructions per second. An interesting challenge for sure.

 

I don't know if that's possible even with seven segment drivers RB,
the change from 11:59 to 12:00.. you have to do the lookup for every digit.
The only way around it I can see is a lookup table of huge memory that
contains all digits for every possible time of day... then the EEPROM read time is another problem.

Maybe not?

What's wrong with this (assuming 4 ports and a 7 seg display on each port);

call wait_minute;
movlw mask1 ; 11:58
movwf portA
movlw mask1
movwf portB
movlw mask5
movwf portC
movlw mask8
movwf portD

call wait_minute;
movlw mask1 ; 11:59
movwf portA
movlw mask1
movwf portB
movlw mask5
movwf portC
movlw mask9
movwf portD

call wait_minute;
movlw mask1 ; 12:00
movwf portA
movlw mask2
movwf portB
movlw mask0
movwf portC
movlw mask0
movwf portD

so apart from the 1 minute timer it takes 8 asm instructions to create AND display the time.

Then looking at that you can simplify to;

call wait_minute;
movlw mask8 ; 11:58 (only 1 digit changes this minute!)
movwf portD

call wait_minute;
movlw mask9 ; 11:59 (same!)
movwf portD

call wait_minute;
movlw mask2 ; 12:00 (3 digits change here!)
movwf portB
movlw mask0
movwf portC
movlw mask0
movwf portD

That would take very little in the way of asm instructions PER MINUTE to make a HH:MM clock.

 

Not sure if it is cheating or not, but just use a PLL to multiply the 50Hz to something more user friendly.

Have Fun

 

Oh man that's a huge amount of program memory still!

markdem, yes although much cheaper than a crystal,
I think it's cheating if the idea was some absolute minimalism

 

digital clocks have been made for years using the line frequency for time base. 50 Hz devided by 50 is 1 Hz. a little more dificult here, 60 Hz devided by 60 is 1 Hz. why would a pic chip or mcu be necessary?

 

Full wave rectify your mains and you get 100 or 120 Hz for a base frequency. Use a simple resistor and capacitor to couple that frequency base through a second full wave rectifier and you get a 200 or 240 hz base frequency to work with.

Run that through a simple comparator circuit to clean it up and make a nice clean clock pulse train and you're good to go and should have plenty of clock steps to work with during every second.