Thanks, I looked at your posts, and that makes sense.

Thankfully I soldered each pin to a wire regardless. I have some floating pins, but they are outputs from the 6502 and the CA and CB lines on the 6522.

When I have time this week I'll try to connect them to ground and see what happens, who knows.

Thank you for the advice!

Chad

EDIT: Correction, I did NOT actually solder a wire for each unconnected output on the 6502. Hm! Whelp, this will be interesting.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Just might. I was pretty careful along the way, testing with the multimeter against any two pins that were really close. Before connecting chips I double checked my address and data lines in particular. I was super strict on close calls, and used solder-wick very liberally.

BUT indeed there is dirt and grime in there that I could never get to. Especially after using solder-wick, or soldering a bit of the wire cover, crud starts flying!

Lots of learning to be had here.

Going off of the "floating pins" idea, I have a question: Could any 6502 outputs or 6522 CA or CB lines be left unconnected safely? I've seen countless schematics leaving some of these pins unconnected. All of the other "strange" input's on the 6502 were connected through pull-up resistors to VCC. This was standard procedure I figured, following Garth's and Floobydust's schematics.

About a month or so ago, I was testing out some 74LS' chips on the breadboard, full-adders and counters and such. I had left some pins floating, and it caused big problems, erratic behavior. It was like loony-land! Same seems to be happening here. Even with all of my double checking, I might have done something wrong. Who knows.

Thank you.

Chad

_________________
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?

Don't tie any outputs to ground - that would be asking for trouble.

When you have an unreliable self-built computer, there are several possibilities
- error in the circuit design (glue logic or clock logic is not safe)
- error in the construction of the circuit (you built something different than you intended to)
- poor quality in the construction of the circuit (dry joints, bad connectors)
- poor power supply quality (not enough bypass capacitors, too much resistance)
- bad component (perhaps damaged by static, or not the component it's labelled as)

The best approach is to be methodical, check everything carefully, and form a hypothesis before you make a change. Take notes of what you see, what you don't see, before and after a change.

Attached are some pictures of what my scope is telling me. When I try to connect the other channel to the clock (for comparisons), this channel goes dead. Sadly, I don't know enough about scopes as to why that is happening.

This is a pretty good representation of what I'm seeing on address and data lines. Last night I even saw it "both", then "high", then "low", then back to "both".

Again, this is happening... sometimes. When it does work, everything hums right along and the data and address lines that alternate look like the clock.

My clock has what looks like an overshoot, like a mini-spike past 5V when going up, and below 0V when going down. When I test Phi2Out or something, the clock signal is cleaned up a lot. Not sure if that has any effects on it.

I'll be pondering on this. Thank you all for the advice. I'm listening.

Chad

What Garth' s saying is, the stuff shown in the picture ( with your post here ) isn't appropriate.

The x10 setting on the scope itself is supposed to be set to match the actual probes you're using. Altering the setting doesn't alter the actual probes, unfortunately.

For the reasons Garth mentioned, your observations using the scope will be badly handicapped until you get proper probes. If you can't get proper probes, maybe you should be placing less reliance on the scope anyway, given that your scope skills are underdeveloped at this point. There are other useful approaches you can be taking.

(One thing you probably can 'scope successfully, even without proper probes, is the 5 VDC power supply. As you attach the connection, the trace should move from a straight line at zero to a straight line at 5V.)

-- Jeff

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

Thank you Jeff.

Yes, you are right. Heck, one of the clips I had to solder and then use electrical tape on because it was ripped. Not the best equipment.

The 5V is definitely coming up, that's no problem. Clock is ticking away as expected, so that's good.

Honestly I've been using my multimeter a LOT on this project. It helps tell me if something is connected, and it tells me voltages. These "both" readings I'm getting read at 2.5V, but so does the clock, and that's why I figured they were the same at first (but they are not).

Thank you again Jeff, I appreciate your experience and insight.

Chad,
First of all, you got an "A" for efforts and persistence which are extremely important to be successful in electronic.

Next time you do point-to-point breadboard, I'd suggest you use 30 gauge wire wrap wire. It is a smaller diameter wire so it is much easier to handle. They are used successfully in countless number of wire wrap boards, so they are good enough for manual wiring as well.

I suspect what you are seeing is oscillation due to positive feedback coupled into ground, an extreme case of ground bounce. Floating inputs can do that so can multiple data lines switching simultaneously. A solution is slowing down the circuit; I don't mean operating frequency, I mean the edge rate of components. Globally, you want to use the slowest device possible, have a solid ground, and lower the supply voltage. Locally, the clock signal and control signals can benefit from proper termination, but I don't think I'll attempt that given the existing mess of wires.

I've come to be AFRAID of W65C02. It is so fast that it needs to be tamed with good ground, good power supply (0.1uF bypass and 10uF filter), and operate at lower voltage (It'll work at 3V so don't be afraid of lower your global supply).

As others have mentioned, you don't have a 10x probe on your scope so the signals displayed will have distorted edge with overshoot and undershoot that are not real. Still, it is an useful tool and you can take advantage of the excessive capacitive loading by probing around and see if your board operate differently when certain signals are probed. That can be a strong indication of a localized problem.

Don't be afraid of lay out a pc board even if the design may not work. PC board is cheap, do reserve some prototype area. You clearly have the skill to patch your pc board if it does not work. It is a lot easier to cut & add a few gates than hand wire the entire board!
Bill

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Am I seeing that correctly? Do you have the sweep time set to 1 second and the trigger voltage set to 0V?

You might wan to set the sweep to something like 10x the clock period (say 10us for a 1MHz clock) to begin with then change as required to see what you need to see, and the trigger voltage to between 1V and 4V.

_________________
Bill

That's a great question. So when I hook my clock up to it, I get about 8 squarewave pulses on that screen. I'm currently running it at 250kHZ.

Now, is the scope changing it's setting automatically on me, and I just don't recognize it? Maybe! A "smart" version I guess.

I'll investigate that later, I'm at work now.

As for BDD and Plasmo, thank you. That's very very good advice. And it makes perfect sense as to why it works sometimes but not others. The capacitors change the 'habit' of it, which is why I've been thinking it's a capacitor problem.

I use the 0.01uF bypass capacitors on each IC, and a 100uF polarized capacitor near the power supply. Should I change those values?

Lastly, how do I change the speed of the rising and falling of these chips? Like you said, it's not my oscillating clock speed, but how fast it comes up and down that's the problem. How do I slow that down? Not just in this case, wire existing, but on a PCB or whatever. Is there something that *guarantees* a slower rise and fall time, besides just getting different chips entirely? With the W65C02 I'm kind of stuck with it, so would these capacitors be enough to 'slow' it down? I like the idea of trying lower voltage on this existing board, but like BDD says, 5V shouldn't be hard to manage. How then do I do that reliably?

Thank you very much.

Chad

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Thank you. I just ordered some 0.1uF capacitors, will replace those and see what happens. I also plan on bringing the VCC's 5V down to 3.3V with a LM1117 that I have laying around, and see if that helps.

I am using 74HC's along with the W65C02. My 62256 is 70 ns and my 28C256 is 150 ns I believe, nothing fancy as far as I can tell.

For the scope, because it's the college's, I don't want to invest a lot of money into it, if at all possible. But we will see going forward.

Thank you BDD.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!