TTL 6502 Here I come

[Dr Jefyll]

Well, the C in 65C02 means CMOS -- it designates the chosen technology. Along those same lines, Drass's machine could be called 65TTL02.

TTL is a bit of a misnomer, though. The chips are CMOS; it's just that they implement the same Medium Scale Integration building blocks originally made popular by TTL. So... 65MSI02 ?? Nah! Doesn't have the same ring to it!

Nice to see Drass and BDD both preparing to breathe life into new designs!

Jeff

[BigDumbDinosaur]

Dr Jefyll wrote:

Nice to see Drass and BDD both preparing to breathe life into new designs!

I brushed my teeth and took a mint to make my breath fresh so my contraption responds in a favorable manner.

[Windfall]

ttlworks wrote:

Windfall wrote:

I'm thinking : 'The 6574'.

Hmm... finding a good "number" for that CPU ain't easy.

From the Motorola notation, 657x\687x almost sounds like sort of a clock generator.
Besides that... if the next project would be (hypothetically spoken) a TTL implementation of the 6526, how to name it ?

7402 already is taken (quad input NOR gate).
8500 almost sounds like (godforbid) 8085.

I'd vote for 7502.

Point taken. Considering the existing 65xxx pollution, it's difficult.

How about '65074'. Elegant, and pretty distinctive !

[Windfall]

Dr Jefyll wrote:

Well, the C in 65C02 means CMOS -- it designates the chosen technology. Along those same lines, Drass's machine could be called 65TTL02.

Excellent. I like that one too.

[Ax2013]

Windfall wrote:

Dr Jefyll wrote:

Well, the C in 65C02 means CMOS -- it designates the chosen technology. Along those same lines, Drass's machine could be called 65TTL02.

Excellent. I like that one too.

Isn't it 65T02 then? Sounds like a member of the same family (to me at least).

Axel.

[Dr Jefyll]

BigDumbDinosaur wrote:

Dr Jefyll wrote:

Nice to see Drass and BDD both preparing to breathe life into new designs!

I brushed my teeth and took a mint to make my breath fresh so my contraption responds in a favorable manner.

Newbies, listen and learn. This is how it's done!

[ttlworks]

I just remembered an old Userfriendly cartoon...
If you notice a penguin sitting on top of your monitor, out to start a discussion about operating systems
(he might be biased), maybe you had a few mints too many.

BigDumbDinosaur wrote:

What's wrong with 6502-???, where ??? is the number of chips it took to build the contraption?

It's not the final version, there is a danger that the ??? number might change after each revsion of the PCBs.

Hmm... lots of 74AC chips on those PCBs.
What about 65AC02 ?

[Drass]

Thanks all for the encouraging comments, and for the wonderful naming suggestions! Nothing like a proper part number to make things official . A favourite so far is “65AC02” - sure sounds like a member of the family and has a very nice ring to it. I thought of “65DC02” (as in “Discrete CMOS”) but that doesn't feel quite right.

Anyway, as ever, some problems to hunt down: a couple of ICs got more than a little warm when I ran some power through Card A, and there is altogether too much current flowing. Maybe some drivers are fighting each other somewhere. I’ll have some time this weekend to check it out, and report back. Lots of fun

Cheers for now,
Drass

[Drass]

Troubleshooting a short - here’s my rookie method:

1/. Set power supply to limit current at 500mA, apply power to the board
2/. Feel for a hotspot - “ouch”, that's a problem!
3/. Patch it, rinse and repeat

I gotta believe there are more sophisticated techniques, but this is all I got. It's worked reasonably well for my most common blunder - upside down ICs. Thankfully, it's easy enough to spot and fix these issues.

But I also saw something more curious on Card A. Current would spike momentarily and the settle back down for a few seconds before spiking again. What? That's strange … maybe drivers fighting each other on a bus intermittently? Sure enough, I had left floating the signals which are used to select one of two drivers for the microcode control lines. These signals are generated in CARD B, which is not yet connected. Tying them down fixed the issue and current flow fell to appropriate levels. Nice!

As a side note, it’s interesting to see first hand how dramatic a driver conflict can be. Those are some nasty spikes! I recall working to implement a scheme dr Jefyll suggested to avoid collisions on the main buses. Seeing this sure makes me wish I had addressed the problem on the microcode drivers as well. Live and learn!

At this point Card A is well behaved, but not so Card B. The hunt continues!

(Edit: fixed the link to point to the right post)

[Windfall]

Drass wrote:

A favourite so far is “65AC02” - sure sounds like a member of the family and has a very nice ring to it.

'65LS02' then. '65AC02' sounds more like the tube version of the 6502 (now there's an idea ... anyone ?)

[ttlworks]

Hmm... looks like 6516 already is taken.
http://www.tubecollector.org/cv4063.htm

[rwiker]

How about 74AC6502?

[Drass]

The boards are behaving with a measure of civility now, and are no longer given to tantrums and spikes - a small victory, yes, but even small steps feel like milestones at this stage. Nothing melted, nothing exploded, and current consumption is mercifully stable - what a relief!

So, onto a setup for initial debugging of the logic:

Connecting the PCBs together with jumpers allows access to the underside of the boards. Flipping them over is hardly comfortable even with this arrangement, but at least touching a logic probe to an IC on the bottom layer without loosing state is actually possible - and pretty much essential to sanity.

The ribbon cables on the left carry the CPU signals out to a breadboard, where a makeshift toggle-clock and control panel now reside. On display currently is the address bus, while the data bus is set to $EA (ready for the inaugural NOP test). Moving the RESET line manually to GND and toggling the clock puts the device into reset. Thus far the reset sequence makes it only to the third cycle before failing; it’s a long way to go yet, but it’s something. Honestly, the first time I saw the internal state counter increment, I was thrilled. It was the first reliable and correct logic operation this unwieldy beast has ever performed, and I could hardly believe it! There is hope yet.

Here is a closer look at the inter-connected boards:

[Dr Jefyll]

Lookin' good, Drass!

Can I presume your makeshift toggle-clock switch is debounced somehow? Just checking. Also, is it the ribbon cables which carry +5 and ground to the CPU boards? Personally I don't trust jumpers for that. I usually take the time to establish soldered power connections, even if it's just a temporary test setting. This is in the "essential to sanity" category you mentioned.

Thanks for sharing the pics, and good luck with the debug!

Jeff

[BigDumbDinosaur]

Drass wrote:

Here is a closer look at the inter-connected boards:

CPU Forest.jpg

For some reason, after looking at your contraption this tune is running through my head.