6502 Setup [2x image - 40k]

Sentient · Post by **Sentient** » Mon Sep 15, 2003 1:40 am

After reading and reading, this is how I have figured a 6502 setup could work but as with anything new, I'd like some expert feedback before I blow anything up

I have no use for single-stepping the processor - are there any other uses for RDY?

This is what I have come up with:

6502.org wrote:

Image no longer available: http://daztro.no-ip.com/digital/6502r1.jpg

The MAX6711L is a reset IC which holds reset if VCC drops below ~4.6volts and releases it ~140ms after VCC recovers to ~+4.6

The 7404 setup for the clock I read is Harry the Bastard's setup. Is this a 2Mhz clock (as it would appear)? If so, what considerations need to be made to make it 1Mhz - Can you simply change the crystal?

Once this setup can reach a majority agreement, I think it might be nice to have it hosted somewhere on the site - for the sake of people wanting to get started with 6502's. I couldn't find a basic setup diagram anywhere - Are you all hiding them, or just past this basic stuff?

GARTHWILSON · Post by **GARTHWILSON** » Mon Sep 15, 2003 6:06 am

I don't see any problem with it, but can suggest a couple of things especially for if/when you go to a CMOS 6502.

For less board space and probably less money, you can replace the TTL 7404 crystal oscillator (which, BTW, I expect would work just as well with your 2MHz crystal swap-out) for a crystal can oscillator that goes into a 14-pin-DIP socket. If you get the half-can size, it'll fit in an 8-pin-DIP socket. These oscillators require no external components, and probably constitute the most reliable choice for crystal control. BTW, Fairchild Semiconductor does have an ap. note on HCMOS crystal oscillators at https://web.archive.org/web/20180311021 ... AN-340.pdf

The CMOS 6502 allows you to hang a crystal directly on its pins, at least at the lower frequencies. If exact speed is not critical, you can also put a single resistor and capacitor right on the pins, with no other parts.

The CMOS 6502 also has a schmitt-trigger reset input, so you can put an RC again directly on the pins with no reset IC or other timer if you want things simple. Keep the capacitor connections from the reset pin to a ground pin as short as possible. The CMOS 6502's reset timing is not critical since it does not have the heating problem that the NMOS 6502 had whereby the uP could be destroyed if you kept the reset line true for too long. IIRC, the maximum recommended reset time on those was 50 or 100 ms.

If you think you'll ever put a WDC 65c02 in there, you may want to have a jumper option to disconnect pin 1 from ground, since WDC's pin 1 is a VP (vector pull) output.

Quote:

I couldn't find a basic setup diagram anywhere - Are you all hiding
them, or just past this basic stuff?

I have a 6502 primer mostly written I'll give Mike to post on the website after he's done with some other stuff he's working on. Part of this primer does cover things like clocking, reset, common glue-logic errors, and what to do with all the mystery pins. Edit, 2023 ! : I finally got the 6502 primer up on my site in 2012, at http://wilsonminesco.com/6502primer/

Sentient · Post by **Sentient** » Mon Sep 15, 2003 1:36 pm

With your suggestion for supporting the Western Design chip as well, I see they have used a few of the NC pins from the original design for other purposes.

While most of these are outputs (which I guess can be left unconnected), one change is of interest. BE (Bus Enable) is an input - should it be tied high to keep the bus enabled?

Also, the WD65C02 is 14Mhz. Any changes to get a stable 14Mhz clock?

Sentient · Post by **Sentient** » Mon Sep 15, 2003 3:25 pm

*dupe*

Sentient · Post by **Sentient** » Mon Sep 15, 2003 3:25 pm

Here is the second revision of the basic setup.

I have replaced the Reset IC, since the MAX6711L held the reset too long.

Also, the clock generator is replaced with a package type (the xxx denotes the speed, although I can only find 1, 2, 4, 8, 20 Mhz - no 14Mhz)

I changed the decoupling capacitor for the CPU to 0.01uF based on Garth's suggestions (and my own preliminary investigation into the subject).

Also added support for the Western Design 6502 (WD65C02) with J1 & J2 (still need clarification on how BE & J2 should work).

6502.org wrote:

Image no longer available: http://daztro.no-ip.com/digital/6502r2.jpg

Good, bad or otherwise suggestions always welcome.

GARTHWILSON · Post by **GARTHWILSON** » Mon Sep 15, 2003 8:09 pm

> With your suggestion for supporting the Western Design chip as well, I
> see they have used a few of the NC pins from the original design for
> other purposes.
>
> While most of these are outputs (which I guess can be left
> unconnected), one change is of interest. BE (Bus Enable) is an input -
> should it be tied high to keep the bus enabled?

BE has an internal weak (20M) pull-up, so you should be fine leaving it disconnected. It wouldn't hurt to add a stronger pull-up for insurance against noise pick-up on a home-made board.

> Also, the WD65C02 is 14Mhz. Any changes to get a stable 14Mhz clock?

I'm no expert at oscillators. I can make oscillators all day long that work, but I don't trust myself to design one that will work reliably at the crystal's marked frequency under all reasonable circumstances, like temperature and voltage range. I leave that for the engineers who are better at the S-plane, and I go with the canned crystal oscillators. I may have a little different perspective though, knowing from having designs in aircraft all over the world that there will always be someone who will unwittingly subject the equipment to operating conditions that neither I nor any government-prescribed testing had completely anticipated. I don't have to be so cautious for home projects, but I'd still like to know I won't ever have to revisit the design just because something starts misbehaving sometime down the road.

> Also, the clock generator is replaced with a package type (the xxx
> denotes the speed, although I can only find 1, 2, 4, 8, 20 Mhz - no 14Mhz)

Jameco www.jameco.com has quite a few more. Then I went to Mouser www.mouser.com and typed in "crystal oscillators" and got 104 results. I looked at Digi-Key www.digi-key.com and found 18 frequencies of crystal oscillators between the 8 and 14 MHz you mention.

It might be better though to get your feet wet at lower, more-forgiving frequencies. When you want to move up, make sure you understand the timing spec.s. A WDC 65c02 marked for 14MHz will, with a clean design and the right support components, go quite a lot faster; but I personally don't know of any commonly available EPROMs that are guaranteed to be fast enough for 14MHz. You may have to load the boot-up code into faster RAM before releasing the processor. Also, use 74AC or ACT logic, not HC, HCT, or LS.

kc5tja · Post by **kc5tja** » Mon Sep 15, 2003 11:17 pm

Sentient wrote:

Also, the clock generator is replaced with a package type (the xxx denotes the speed, although I can only find 1, 2, 4, 8, 20 Mhz - no 14Mhz)

Look in DigiKey -- they have tons of oscillator "cans" that vary in frequency from 1 to 50MHz, including 3.579545MHz, 14.318181MHz, and 28.63636MHz.

Quote:

Also added support for the Western Design 6502 (WD65C02) with J1 & J2 (still need clarification on how BE & J2 should work).

Datasheets for the 65C02 and 65C816 processors are available on this website in PDF format. Just download, open it with a PDF reader (e.g., Acrobat Reader), and print them out. It will tell you precisely the details regarding the BE signal.

Sentient · Post by **Sentient** » Mon Jul 26, 2004 10:15 am

I have a weird question

Does anyone happen to have (either saved or cached) a copy of my images that I linked here. They are no longer on the web server, and I dont have an local copies.

Would really appreciate getting them back if possible.

Thanks