What exactly does that mean? What makes the 14MHz clocks of the VIC20, for example, non-overlapping? Why aren't they just complements of each other? Is it because of the Xtal and cap, or due to the non-symmetry of the 7402 tPHL tPLH?

And what shapes would the two phases have? I have an idea, if they were perfect, but I can't be sure because my USB oscilloscope doesn't have the resolution needed to display that rate accurately.

Finally... why has checking "Log me on automatically each visit" NEVER WORKED FOR ME???

_________________
"I am endeavoring, ma'am, to create a mnemonic memory circuit... using stone knives and bearskins." -- Spock to Edith Keeler

Non-overlapping means that one or the other can be high at a time, but never both of them high at the same time. And what that means, electrically, depends on the input threshold of the device which cares - if the input threshold is 2.5V that's easier to arrange than if it is 1.0V.

But if the circuit works, and is reliable, than that's fine. As there's nothing clever about this circuit - nothing which senses whether the other signal is high before letting this one go high, nothing specific to bring the true and inverse signal into phase - I'm going to suppose it works because in most technologies pulling down is faster and stronger than pulling up.

It's also possible that 'non-overlapping' is the spec, it's a spec which this clock-driving circuit doesn't strictly meet, but the whole thing works anyway - because being in-spec should lead to a circuit which works, but being out-of-spec merely means that the circuit might not work. It's not a good idea to be out of spec all the time, but sometimes the true line of working vs not working lies somewhere slightly different than where the spec lies.

(As for the forum trouble, I'd suggest clearing cookies or trying a different browser - this is a feature which works for me. If you don't want this thread to be a mix of responses to two unrelated questions, better to start another thread for the other question.)

Here's the suggested circuit from the original ARM databook for converting a single phase clock into a two phase non overlapping clock. Note that each output cannot go high until the other one has gone low.

Somewhere I saw a similar circuit, and this one DID sense the opposing signals, using cross-connections like the ones I added below (in color). I seem to recall this was for the 6509 CPU in one of the Commodore CBM machines. Clearly they wanted to ensure the two output phases couldn't overlap.

But, Richard, I wouldn't worry about the circuit you've got. I suppose you could add the cross connections. But it does seem to work as-is (as Ed said).

Edit: I see George has mentioned much the same solution.

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

I wasn't looking to "improve" the circuit, just to understand how it could be what the spec calls for. But as Big E suggested, sometimes a spec is more of a suggestion.

I've seen clocks like what gfoot posted and wondered why VIC's wasn't like that. I found that in schematics of earlier versions the clocking did have the cross-connected NOR flip flop, so in updating the designers must have decided that wasn't necessary, and the circuit worked well enough without it.

As for how the two phases actually look, I notice now the waves my 'scope displayed seem to spend more time low than high, so what I drew is probably pretty close.

Thanks all.

_________________
"I am endeavoring, ma'am, to create a mnemonic memory circuit... using stone knives and bearskins." -- Spock to Edith Keeler

Is it possible that your browser needs to have its cookies or similar "paranoia settings" loosened up a bit?

_________________
Got a kilobyte lying fallow in your 65xx's memory map? Sprinkle some VTL02C on it and see how it grows on you!

Mike B. (about me) (learning how to github)

I'll have to look into that. The thing is, sites that don't have the automatic login feature keep me logged in as long as I don't delete their cookie. My browsers are all set to save logins & passwords, so logging in here isn't a big deal. It's just never been automatic.

3rd party cookies are blocked.

_________________
"I am endeavoring, ma'am, to create a mnemonic memory circuit... using stone knives and bearskins." -- Spock to Edith Keeler

The target waveform is shown in that attachment and the VIC designers interpretation of it allowable timing is shown in the table. As Ed said, non-overlapping means that the phases will not be in the active state (whether that is high or low) at the same time.

The 6502 clock outputs also represent a non-overlapping 2-phase clock.

_________________
Bill