Clock Divider - Making an 8 Bit pipelined CPU - Part 79

preview_player
Показать описание
In this video I build the 3rd clock source for the main clock/reset circuit. Because I don't yet know exactly how fast this will run I build a clock divider circuit using a pair of 74LS193's to reduce the output of the a 5Mhz crystal oscillator.

This would have been an extra short video, but I go on to tidy some of the wiring up. Replace a mostly redundant latch chip with a simpler 74LS74. I also add a couple of capacitors to bias the start-up state of both the reset and mode selection latches.
Рекомендации по теме
Комментарии
Автор

this is so nice :) I myself had been working on a similar topic making my own crystal clock with not gates and dividing it down twice (divide by 4) used a 7474 to do that. it works on breadboard but sometimes it has trouble starting. i really love watching these videos.

dualityy
Автор

The ringing is from the massive ground loop you have there. The Gnd from the scope goes all the way across the top rail, jumps to the other rail, and then goes all the way to the oscillator.

It's the loop area of a circuit that determines the inductance. This is why breadboards suck for high speed. They waste a lot of area making signal wires longer than they need to be, and encourage routing power and gnd to opposite power rails creating a nice big inductive loop.

drlazy
Автор

That's going to be a pretty fancy clock!

FrankGevaerts
Автор

Thanks James., very interesting. It will be interesting to see if you can smooth out the clock. Take care.

jerril
Автор

As always very nice circuit, I will probably use this in my build.

mathiasvermeire
Автор

Part of that ringing will be caused by the scope ground lead it's quite long. If you have a clip ring with a pin use that in place of the ground lead and it might go away.

TheEmbeddedHobbyist
Автор

You mentioned using a Schmidt trigger to clean up the waveform. What is the advantage of that over a zener diode? (Be gentle, it's been 15 years since I last dabbled in electronics.)

IslandHermit
Автор

You shouldn't need a Schmitt trigger to clean up that signal as long as it rises above the Vih threshold within the rise time constraint listed in the datasheet. Not sure if that matters as much with TTL as it does with CMOS...

Likewise Vil and fall time.

TomStorey
Автор

Very nice. Have you given any thought to an interrupt handler for your UART?

michaelheuss
Автор

Very nice indeed. I guess you are going to have an instruction to pause the processor? Interesting!

lawrencemanning
Автор

Are you familiar with the 556? It's two 555's in a single package.

AndyGoth