Building a simple 6502 microcomputer

[GARTHWILSON]

Quote:

I started assembling my SBC. Only trouble is that the kynar wires often break while wrapping them around the posts. I've been trying to wrap very slowly but even then it happens occasionally. What am I doing wrong?

What tool are you using?

From the wire-wrap page of the 6502 primer:

Quote:

Good wire-wrapping relies a lot on the senses. What does it feel like is happening inside the tool? Is the tension consistent? Can you feel the end of the insulation landing in the right place? Is it binding? Can you feel the previous turn under where the wire is currently being laid? The Kynar and the cheaper copper wire have very different feels, but I like them both. When you're new at it, you may tend to have wraps whose layers are not together and wraps where layers overlap. Especially the overlap is no good and needs to be redone. Wire-wrap tools tend to have sharp areas in them when they're new, and you may accidentally cut the wire sometimes at first. Just keep at it. Both you and the tool will improve.

Perhaps I should also add that I think it is beneficial to hold and spin the WSU 30-M tool from the smaller-diameter portion too, not the fatter part.

[BigDumbDinosaur]

GARTHWILSON wrote:

Perhaps I should also add that I think it is beneficial to hold and spin the WSU 30-M tool from the smaller-diameter portion too, not the fatter part.

The average wire-wrap tool actually produces quite a bit of mechanical advantage, making it easy to apply too much tension to the wire. As Garth notes in his primer, a light touch, coupled with experience, will lead you to making good connections. When I first started with wire-wrap (in 1970) I busted my share before becoming proficient. It just takes a bit of patience.

[alkopop79]

Turns out, the wire I was using was brittle. Then I used a different wire and it all went fine. Now I can make a connection in seconds. Each and every connection is tested. I have already finished the data and address buses, still waiting for one more IC socket and white wire for the glue logic and the clock (data is green, address is yellow). I have to say wire wrapping as way faster and soldering and can be undone very easily. Haven't had this much fun for ages. Happy days!

[ElEctric_EyE]

alkopop79 wrote:

Turns out, the wire I was using was brittle. Then I used a different wire and it all went fine... I have to say wire wrapping as way faster and soldering and can be undone very easily. Haven't had this much fun for ages. Happy days!

You'll have even more fun when you use an auto WW stripper, and an electric WW gun... Less delay between point to point connections.

[Dajgoro]

Ah, i recently spent 3 hours soldering enameled wire for a total of ~140 connections(4x 64k ram and 32k rom in parallel). maybe one day i should get ww stuff as well.

[alkopop79]

It's superfast.

[alkopop79]

I realised that if I want to use an ftdi-USB cable to for serial connection I need to bypass the 232 and wire the ACIA's TX and RX straight to the connector. Is that correct?

http://www.sparkfun.com/datasheets/DevT ... e%205V.pdf

[8BIT]

alkopop79 wrote:

I realised that if I want to use an ftdi-USB cable to for serial connection I need to bypass the 232 and wire the ACIA's TX and RX straight to the connector. Is that correct?

http://www.sparkfun.com/datasheets/DevT ... e%205V.pdf

From this page:
https://www.sparkfun.com/products/9718

It says, "Description: The FTDI cable is a USB to Serial (TTL level) converter which allows for a simple way to connect TTL interface devices to USB. The I/O pins of this FTDI cable are configured to operate at 5V."

So yes, it would wire directly to the ACIA's TX & RX. Also, wire the GND pin to your board's GND. The 5V pin is an output (from the USB port), so if your board is already powered, do NOT connect this to your board.

Good luck!

Daryl

[alkopop79]

Thank you! Almost finished the wiring, now it's just the clock and the address decoding missing. All the power, address bus and data bus are wired.

[Dajgoro]

Since you already have a nmos 6502, you could try the nop test, where you hardwire the data bus to the nop opcode, and the result should be that cpu will step through each address, so if you poke a speaker/headphone on the address bus, starting from A15, you should hear a higher frequency on every next address pin, don't expect to hear more than 20KHz (no scope test). But you will need to wire the cpu clock first.

[alkopop79]

Being busy wiring up the clock circuit:)

[Dajgoro]

You can test the clock circuit by getting 1MHz out of it, and then connect a wire to the clock, and leave it hanging in the air. Then go to your stereo and turn the radio on and set it to AM and tune at 1000KHz and then toggle the power and you should be able to hear it on the radio. Or if you have a bigger counter, use the speaker/headphone trick(no oscilloscope test, tested).

[alkopop79]

I still haven't finished my UK101 but have already learnt lot. Here's a few tips I'd like to share with anyone being new building an SBC:

- wire wrap (WW) might be more expensive than soldering but fast, reliable and can connections can be easily un-done. The tool can be purchased from Farnell and way cheaper than on Ebay ( http://uk.farnell.com/jsp/search/produc ... f-00001000 ). Do use coloured wires! I had different coloured Kynar wire for address, data, data decode and logic, clock and power. Orange colour is not good choice, it looks almost exactly like red (VCC).
- the components are on a plain perforated board (no traces). I wish I have bought TriPad, could make distributing power easier.
- the sockets are glued with hot glue gun. It ain't pretty but holds the sockets strong enough and can be removed if necessary. I've placed a socket too close to an other one. Since I didn't use cyanoacrylate (superglue), I could remove the socket and place it somewhere else.
- worth checking the connection three times: with my multimeter set to continuity meter I checked each newly wrapped connection. Then when I finished wiring, I checked again on the component side (I found quite a few connections wrapped to wrong pins!). And finally, check connections with the ICs placed in the sockets. This way I found a shorted connection that could have cause lots of trouble. I do not recommend powering the board with the ICs in until all the connections are verified!
- label the pins/posts!
- next time I'll print out the layout, wiring and the schematics and will mark all the connections wired with highlighter. Instead I used a tablet or my laptop.

That's all top of my head.

[BigEd]

Good tips!
How have you labelled the pins/posts?
Cheers
Ed

[BigDumbDinosaur]

alkopop79 wrote:

- wire wrap (WW) might be more expensive than soldering but fast, reliable and can connections can be easily un-done. The tool can be purchased from Farnell and way cheaper than on Ebay ( http://uk.farnell.com/jsp/search/produc ... f-00001000 ).

That tool appears to be the same one I have, which I bought a long time ago.