6502, Φ2, etc.

[BigDumbDinosaur]

ArnoldLayne wrote:

BigDumbDinosaur wrote:

ArnoldLayne wrote:

Hi, this is basically what the WDC 65c02's datasheet is stating. So, you are saying that this not only goes for WDC but for 65c02 in general?
I'm asking because I did not find anything about this in the Rockwell datasheet.

This is design advice from WDC...

Hm, then I should be able to leave that jumper always in WDC position.

Yes.

[Hobbit1972]

BigDumbDinosaur wrote:

Incidentally, here is the Ø2 clock generator circuit I use.

clock_gen.gif

Hmm - at least datasheet of my NMOS 6509 says it wants to have a full rail clock. (It actually seems to work w/ TTL level clock.) Is your circuit only for CMOS CPUs or is it also intended for NMOS w/ full rail clock?

[BigEd]

(I note that the 6502 launched at a 20% price premium for including the on-chip clock generator. The 8080 needs a 9V clock... clocking was hugely inconvenient prior to the 6502.)

[BigDumbDinosaur]

Hobbit1972 wrote:

BigDumbDinosaur wrote:

Incidentally, here is the Ø2 clock generator circuit I use.

clock_gen.gif

Hmm - at least datasheet of my NMOS 6509 says it wants to have a full rail clock. (It actually seems to work w/ TTL level clock.) Is your circuit only for CMOS CPUs or is it also intended for NMOS w/ full rail clock?

The 74AC74 flip-flop can swing Q and /Q from Vcc to ground and source or sink up to 24 ma. It should work fine with the 6509. You could also use a 74ABT74 instead of the 'AC74 to get more drive, but that's probably not necessary in your application.

Incidentally, if you need a Ø1 output along with Ø2 you can derive the former from the /Q pin on the flop. Ø1 and Ø2 will be exactly 180 degrees out of phase with each other.

[Dr Jefyll]

BigDumbDinosaur wrote:

The 74AC74 flip-flop can swing Q and /Q from Vcc to ground and source or sink up to 24 ma. It should work fine with the 6509.

I agree with BDD that AC series parts are an excellent choice for a clock circuit. But the 6509 is an odd bird, unlike other 65xx CPUs in that it requires two separate, non-overlapping clocks -- at least that's what the datasheet says. (The 6800 is like that, too, IIRC).

Somewhere I saw the clock circuit for the CBM-II (?), which is a 6509 machine. It used two TTL logic gates cross-coupled in a manner that avoids overlap, and low-value pullup resistors to ensure adequate logic-high voltage levels. I suggest you adopt the same circuit, but using 74AC (and omitting the pullups).

-- Jeff

[BigDumbDinosaur]

Dr Jefyll wrote:

I agree with BDD that AC series parts are an excellent choice for a clock circuit. But the 6509 is an odd bird, unlike other 65xx CPUs in that it requires two separate, non-overlapping clocks -- at least that's what the datasheet says.

Right, forgot about the weird clock requirements, which are clearly shown on the timing diagram (attached).

MOS 6509 Read Timing Diagram

MOS 6509 Write Timing Diagram

MOS 6509 AC Characteristics

[Hobbit1972]

Dr Jefyll wrote:

I agree with BDD that AC series parts are an excellent choice for a clock circuit.

Till now I use a SN74ALS02N... let's see what's in my storage box... *rummaging through arsenal*... hmmm... ...I even don't have another '02 of any kind... so the ALS will have to do.... and I should add the pull-ups then.

Dr Jefyll wrote:

Somewhere I saw the clock circuit for the CBM-II (?), which is a 6509 machine. It used two TTL logic gates cross-coupled in a manner that avoids overlap, and low-value pullup resistors to ensure adequate logic-high voltage levels. I suggest you adopt the same circuit, but using 74AC (and omitting the pullups).

You probably mean CBM-500 and CBM-6xx or "b-model": http://www.zimmers.net/anonftp/pub/cbm/ ... index.html
It uses 2xNOR to generate the non-overlapping clock. I copied my clock circuit from there (but somehow the resistors didn't make it to my breadboard...)

[Dr Jefyll]

Right, thanks -- this is the part I remember (from this page).

Pin 13 and pin 10 of the 'S02 NOR drive Φ1 and Φ2. And the cross coupling guarantees the "non-overlapping" behavior. Pin 13 must finish going low before pin 10 can go high, and vice versa.