Hi All
I've build a small prototype debug board for diagnosing KIM-1 problems.
During my use of it, I found that my 6530-002 was dead.
OK, some have used a 6532 to replace the 6530 on pinball machines. I
intend to make a PC board for the debug board and two small boards that would
hide under the 6532 to cross connect wires and have a small EEPROM to
emulate the 6530's ROM. The circuit and wiring is different for the KIM.
In order to reduce cost I've looking for others that would be interested in buying
the board set from me.
Some assembly would be required but I'd presolder the surface mount chips
to go under the 6532. The rest would be up to you.
You can contact me at dkelvey at hotmail dot com
You can follow progress at:
http://www.vcfed.org/forum/showthread.p ... 1-computer
Dwight

Depending on your requirements, you can probably have them made by DirtyPCBs.com/ or pcbshopper.com/ or pcbway.com/ extremely inexpensively, so you can afford to do it even if no one shares the cost with you. I've just started using DirtyPCBs and I tell of my experience in recent posts on the Cheap PCB stories... topic. You can get your entire order as low as $14, including shipping, for a small order of small boards!

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

Your the second one to recommend dirtyPCB. I still have to get
a cad program. Is Eagle the way to go? How much is that going
to cost.
Dwight

When I was shopping for CAD over 20 years ago, I gathered quite a few demos. Eagle said theirs absolutely would not crash. Well, the demo itself crashed on me. I'm sure they've made huge progress since then though, and probably have a fine product. There are lots of free CADs out there though. https://en.wikipedia.org/wiki/Compariso ... A_software might help (or it might be TMI).

I would advise against getting one that severely limits the size and complexity of the board. Some CADs will entice you with something free but then you find that you need more, and you can't get it with the free version, and by that time you've spent time learning to use it so you're reluctant to have to move to another one. (I'm sure that's what they're counting on!) I would also advise against getting a CAD that does not output industry-standard gerber 274X and Excellon files. (Gerber is for the photoplotter, ie, the graphics data, and Excellon is for the drilling.) There are board services that give you free software but it's made to only be good for their particular service. If for any reason you want to have another board house make the board for you later, you're stuck unless you can find a file converter.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

I'm familiar with Gerber. I used to have to hand edit these when the deisgners
would mess up stuff. I didn't route wire, usually just line width sizes and stuff like that.
That was years ago ( more than 20 for me ).
Dwight

The gerber back then was RS-274D. What's in use now is 274X. I had to learn to convert, but once you have your templates, it's a very quick process to do it by hand in a text editor. After you massage the main aperture list, it can be copy and pasted into most layers' files and they just take just a few seconds each to convert. Plane layers take more time, but if you do it frequently, it's still pretty straight forward. If I haven't done it in a while, it takes me longer since I'm not fresh at it.

I'm still using an old DOS-based CAD which puts out 274D. One reason I stay with it is because I have hundreds of custom component footprints in it which I don't want to have to re-make. Yeah, good CADs come with thousands of footprints, standard, and actually mine came with at least hundreds; but there are always lots more for special parts they don't have, and I re-worked the standard ones anyway for better density.

Years ago, with 274D, when I had complex layers with ground planes, chainsaw lines, text in the planes, shaving pads, etc. I had to do multiple gerber files for a layer, with multiple aperture files, and give very detailed instructions in the readme.txt file. The CAM people tended to mess things up and then call me saying the data were unusable, and I'd have to point out they didn't follow the instructions, that they used the wrong aperture file, etc.. The newer 274X lets you do a lot more, and get all the info for a layer into a single file that the CAM people don't have a chance to mess up. It really made things easier. It's also one of the reasons these board houses now can make small quantities so cheaply. They require a lot less human attention.

For a simple board, you could conceivably lay it out by hand on quadrille paper and take the coordinates and type up a gerber file without any CAD at all. Unless it's super simple though, it might be a question of whether you'd live long enough to finish. There are gerber viewers available to check the output. I use gerbv to double-check my work. gerbv is free. Edit, 12/1/16: I see there's a free 3D online gerber viewer at http://mayhewlabs.com/3dpcb .

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

Hackaday is running a series of posts trying various PCB tools at
http://hackaday.com/tag/creating-a-pcb-in-everything/

KiCad is well-regarded, and free.

A well timed tutorial.
Dwight

I second Kicad. I've found it really easy to learn. It has a few annoyances but I imagine any large application does.

_________________
Cat; the other white meat.

My hand wired adapter seems to be working. I still
have to do some more checkout but the I/O is working as expected.
I have to wait until my new laptop gets in before I can start the KiCad.
My previous machine has gotten sick.
Dwight

My prototype debug test board has successfully run test code on the -002.
I've created a prototype 6532 to 6530 adapter and finally got it
to work.
I was having issues getting the EEPROM to program. It has been
frustrating. I had a wiring error and some bad code. Now it is
doing what it should. Yaaa!
I now have a working KIM-1.
Next step is to make the PC boards.
Dwight

Sweet! As a KIM-1 guy from way back, I get a sentimental thrill from news like this. Are there any photos of your test setup you can share? Good luck with the PCB version, and keep us posted, please.

-- Jeff

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

The prototype is ugly but functional. I was having problems
with the loading of the EEPROM. It is a 28C64 but I only use 1K of its 8K space.
It is tricky. I use an image of the -002 with a modified NMI. The debug board
loads some code into RAM to provide the programming operation.
It switches to the RAM and waits for the NMI.
One then switches the debug board to the 1K -002 image.
Then you type the ST button.
If all goes good, the status LED on the debug board flashes.
If it then goes off, the EEPROM has been loaded including the patch of the
NMI vector, to the KIM's correct vector.
It is all tricky code as it has to all be done in sequence.
I'd moved the code around some and it stopped working. I'd forgotten
that the NMI vector in the image needed to be updated as well. I just couldn't
understand why adding a subroutine could cause it to fail. It finally dawned on
me when I added just a single byte offset that it wasn't the subroutine
itself it was the movement of the NMI. Originally I thought I could use it in
the code that loaded the RAM with the programming code. That wasn't
going to work. A chicken and egg problem. I'd patched the
images NMI vector and then forgot about.
3/4s of a day blowing test code on EPROMs to try and figure what was
wrong.
Dwight

Greetings, Dwight:

Congrats' on getting everything up-n-runnin'. I really like your logic for decoding the 6532 I/O and RAM address spaces.

Just for fun... I noticed that if you fiddle with the address decoding logic a bit I think you could fit it into the pair of Configurable Logic Cells in a little 8-pin PIC microcontroller (see below).

I look forward to your progress reports, Sir.

Cheerful regards, Mike

Hi Mike
Oops I take it back, I need to look again at what you have.
This would work fine for the 16L8.
I was intending to stack two SOIC chips and the NAND/INVERT pair
worked out well for cross connecting pins, when stacked.
Also, I wanted to keep selects to one or two gate levels deep.
I've since changed and will use a 16L8 on the PC boards.
By configuring the 16L8, one can do any 6530 applications.
At first I thought you'd made the same mistake that is on Ruud's
schematic for the KIM. It treated RS\ as a separate select as the
KIM does on the 6530. This won't work on the 6532.
Dwight