Overclocking CMOS 68000 CPUs

[Iggy]

How overclockable are CMOS 68K based parts?
I have seen a lot of data on '030, '040, and '060 parts, but what about the
MC68HC000 or MC68HC001.

I could really use parts like those running in the 40MHz range.

[kolla]

Good question - 68SEC000 on Minimig runs at ca 50MHz and it is not even luke warm

[Iggy]

Good question - 68SEC000 on Minimig runs at ca 50MHz and it is not even luke warm

Great! Since I just logged in to edit the first message with something like "And what about the MC68SEC000?"

Can anyone point me to a source of MC68SEC000 chips that will run at at least 40MHz?

And does the MC68SEC000 still retain the E output of the  MC68HC001?

Edit: Scratch that second inquiry.
The MC68SEC000 does not appear to have the E output of the  MC68HC001.
Not a big deal, I just have to upgrade some of the components I have in my current design.

Still, high speed MC68SEC000s, anyone?

[Iggy]

The only 20 MHz 68SEC000 I can find is the MC68SEC000AA20.

Again, does anyone know what might be used in a Minimig?

[crasbe]

Everything except this 68SEC000FU rubbish will run over their rated speed

I'm pretty sure, that freescale actually meant "FU" to mean "%&$#?@!%&$#?@!%&$#?@!%&$#?@! YOU", because these CPUs aren't overclock friendly

And by the way: the CPUs run WAY cooler if you run them with 3.3V.

[RobertB]

How overclockable are CMOS 68K based parts?
I have seen a lot of data on '030, '040, and '060 parts, but what about the
MC68HC000 or MC68HC001.

The 68HC000's in my SupraTurbo 28's run at 28 MHz..

Truly,
Robert Bernardo
Fresno Commodore User Group
http://www.dickestel.com/fcug.htm

[ElPolloDiabl]

If you finish your project could you share with others?

[Iggy]

If you finish your project could you share with others?

Ah, it started out as a weird speculation that I could create the needed quadrature clock for an HD63C09E running at 3.58 from the E output of an MC68HC001 if I could get one to run at 35.80 MHz.

Just a weird project that would run dual 8/16 and 16/32 bit cpus in lock step.

It has mutated seriously into multiple idea so I have no idea what I might report.

As an Amiga accelerator, the '030 makes vastly more sense thanks it its much higher throughput at comparable clock speeds.

OK, so no "FU".

Any suggestions on possible replacements?

[kolla]

The only 20 MHz 68SEC000 I can find is the MC68SEC000AA20.

Again, does anyone know what might be used in a Minimig?

One that everyone was surprised to find working perfectly well at speeds close to 60MHz, hehe. I can have a look tomorrow, left mine at work.

[Iggy]

One that everyone was surprised to find working perfectly well at speeds close to 60MHz, hehe. I can have a look tomorrow, left mine at work.

I'd like to see that.
I am getting wildly varying reports on what these chips will do.
But 60MHz doesn't surprise me.
The '060 did it, so why not?

[psxphill]

But 60MHz doesn't surprise me.
The '060 did it, so why not?

Intel design a chip and then speed bin it to get the fastest parts and when yield increases they apply speed limits to parts. Motorola design a chip for a specific speed and then sell it with that rating. The only exception I can remember is the 68000P12 being rated at 16mhz (the P8 and P10 were rated at 8mhz and 10mhz).

The why and why not a chip can run at a speed is complex, the design tools generally leave a large margin but all the interactions can introduce problems.

The SEC was aimed at the low powered embedded market and because it's fully static you can stop the clock, the non static versions need to be clocked to refresh the internal state (similar to dram refreshing). You might find that running at ~60mhz introduces too many errors, which you probably don't care about when running amiga software but in a pace maker it might be more of an issue.