MPU Clock circuit

[Danno]

Hi,

Attached is a pic of the clock circuit on a Motorola MC6802, most circuits I have seen have the crystal across XTAL & EXTAL. On this example it's only going to EXTAL.

Could somebody explain how this works please.

 

[Harald Kapp]

A crystal between XTAL and EXTAl uses the internal oscillator of the chip.

In this circuit teh NAND gates ar euses as Inverters (both inputs connected) and the inverters are biased to linear operation by the feedback resistors R49 and R50 (this is an abuse of CMOS inverters seen not seldom). These amplifiers and teh crystal in the overall feedback loop form the oscillator as described here. The output from IC6, pin 11 is used to clock the MCU.

 

[Old Steve]

I wonder why they elected to use an external oscillator? Perhaps to run the chip at a frequency that was outside the normal (frequency) range of the internal oscillator?

Usually, I only see an external oscillator used when synchronisation with another device is necessary.

 

[Danno]

A crystal between XTAL and EXTAl uses the internal oscillator of the chip.

In this circuit teh NAND gates ar euses as Inverters (both inputs connected) and the inverters are biased to linear operation by the feedback resistors R49 and R50 (this is an abuse of CMOS inverters seen not seldom). These amplifiers and teh crystal in the overall feedback loop form the oscillator as described here. The output from IC6, pin 11 is used to clock the MCU.

Ah I see, thanks for explanation

 

[Danno]

I wonder why they elected to use an external oscillator? Perhaps to run the chip at a frequency that was outside the normal (frequency) range of the internal oscillator?

Usually, I only see an external oscillator used when synchronisation with another device is necessary.

There was a early prototype to this system that used a 6800, maybe a carry over

 

[Old Steve]

Possibly. There might be some other reason that I can't think of, too.
As an aside, in terms of synchronisation, in theory I think it's possible to completely stop execution by freezing the clock. I've meant to try it for years, but never got around to it.

 

[Harald Kapp]

in theory I think it's possible to completely stop execution by freezing the clock

The 6800 had a minimum clock frequency of 100kHz, possibly becasue some internal logic circuits used dynamic logic, not static logic, to save precious sislicon real estate?

 

[Old Steve]

The 6800 had a minimum clock frequency of 100kHz, possibly becasue some internal logic circuits used dynamic logic, not static logic, to save precious sislicon real estate?

Thanks for that Harald. I've learned something new.
Never played with 6800s. Mostly PICS, with their 'static' logic. (And about to add ATMega328s in a week or two.)
Mind you, the 6800s are an antique now, (like Danno's old machine), so I doubt that I'll ever get to put that knowledge to use.
The ability to 'freeze' the clock on more modern devices could be handy, though.

 

[Danno]

Old thread I know but how would you work out the frequency of the system, the crystal is 4MHZ I believe.

Thanks