New ZX80 boards

[Lee Hart]

Knightfire suggested looking at the ACE circuit. That's a good idea. I also looked at a few other ZX clones and related designs.

Unfortunately, they all seem to use the same design. Diodes from the Z80's address bus to the keyboard matrix, and either a 74LS365, 74LS244, or 74LS245 to read the other side of the keyboard matrix. Pullup resistors are as low as 10k, or in the case of the Jupiter ACE, *no* pullups!

When I looked at this problem earlier, it was apparent that this is not a "guaranteed to work" circuit. The diode drops already put the read threshold outside the worst-case limits for an LS gate. Lowering the pullup resistors increases the voltage drop in the diodes, which would seem to go in the wrong direction.

I tried germanium diodes, which have a far lower voltage drop. They make the low-state voltage meet specs, but didn't seem to help the shifted key problem.

More work needed...

[sirmorris]

Greetings
-oliver

Hi Oliver!

Welcome aboard - especially with such great input (Ha! no pun intended .. ;¬J ) about the keyboard problems here.

My board has been languishing recently - my project list is too long but I need to make progress. I'll try your resistor swap on my HC equipped board - I can't say when, however

C

[nathanallan]

Hi guys,

My poor TS is sick and this looks like the thing to do for it, replace the board!

So you have another interested person, though it will be a little bit before I can get together funds for it. Hopefully starting a new thing this week, but we will see.

Nathan

[Lee Hart]

Hi nathanallan,

I've been busy with the holidays, and haven't had time to work on my ZX80 Legacy board. But there will definitely be another "run" of boards. I just want to make sure I fix the keyboard problem before I do. (Some work, some don't, and I want to know why, and do something about it)!

PS: What have I been busy with? Search eBay for my "Electronic Christmas Tree Kit". These are little circuit boards with blinking LEDs that I'm selling for $9.95. It's a good project to introduce kids to building. .<:-) (smiley face with a Santa hat)

[Paul]

Hi Lee,
today I had a lot problems with my keyboard, specially with the row holding keys 6 to 0.
I started searching why pressing a 0 would result in a 1 on the screen.
I found my problem between diode 6 and diode 7 (or to be exact my soldering was the problem ). There should be no contact but there was (depending on temperature and the bending of the pcb: allways as today, or only once in a while). The rest of the wire of Diode 6 was too long and could touch diode 7.
Could you leave a bit more space between them on the next batch of PCBs to make it more idiote-save?

[zx81jens]

hi paul!

i had no problems at all

greetings
jens

[Paul]

i had no problems at all

I didn't think you had
You had your ZX80 built with "feet" under the PCB right from the start.
I used it a few weeks on a hard surface before i put feet under the PCB. By that time the wires were already bended.
To be honest, the reason why I put the feet underneath were my sporadic keyboard problems.
I don't think it is a big problem, I only wanted to ask wether there could be a change just in case, in order to prevent problems.
Or otherwise there could be an extra hint in the documentation to make shure the diodes don't touch (which should be clear to anybody, I know ), and not to operate the keyboard when the PCB is lying on a hard surface, and to make shure to use feet under the PCB). The current distance isn't to small, its just as big as necessary, its just that there is some space left that could be used, and the diodes are put in two lines, 4 in the bottom line, two a bit up (to stay clear of P3) and another three in the bottom line. If they were one in the bottom line, one up, another down, next up...... then even stupid fellows like me could do it right.
And of course I also wanted to state that my keyboard problems are now all solved.

[zx81jens]

hi paul!

you´re right - i had some problems - and without you and olli they would be still there

but: i NEVER had problems with the keyboard. the keys worked from the first powerup... that´s what i meant

greetings
jens

[Paul]

Another post about the Keyboard:
1. I did not have the HC-Chip Problem although I had replaced all chips against HC-Type.
2. I wondered why there was no reply to Olivers statement about 10K resistors against 47K resistors.

So I kept searching for a 74HC244 of a different Batch that could reproduce the problem with the Shift keys.
Thats not as easy as I thought. From 3 different suppliers, I got chips from the same producer, most probably from the same batch.
Today I found a chip that could reproduce the problem (Backspace always worked, but ="()+- just gave their equivalents as if the shift key wasn't pressed) when used as U10.
Glad to be able to reproduce the problem, I changed the resistors to 10K and voila, everything works as expected.

So I can definitly say that Olivers suggestion is a workaround

So it seems to be worth trying to change the resistors.

[Lee Hart]

Happy New Year to all!

On the spacing between D6 and D7. I tweaked the layout a bit and was able to move D7 a bit farther right. But I worry that the really tight spacing is to the trace on the bottom side of the board below D6 and D7 (between them and the keyboard connector). I could not see a way to get more space on that one.

On the 10k vs. 47k pullup resistors. I think the 10k would help, especially with the 74HC parts. 74LS parts already have a weak internal pullup resistor -- maybe that is what Sinclair was depending on?

I haven't had time to get back into the basement to fool around with the keyboard issue on mine. But I hope to as soon as I can get some other things cleared up. I want to have a solid fix for it before I order more boards. If all we have to do is make resistor changes or use certain types of chips, great! The layout won't change. But I want to be sure. There is nothing more maddening than building something that *almost* works.