ZX80 Gerber files

[MaFli]

I'm in the process of redrawing the schematics and PCB layout of the ZX80 in my CAD program. I know this has been done before and I have seen the great job that Grant Searle did, creating foils so you can build your own. But I've thrown away my UV exposure box and chemicals as soon as professional PCB manufacturing became available .

So I'd like to make a set of gerber files as close to the original as my CAD program allows me to. (Those darn curved traces on the top )... But I am getting there!

I don't have a ZX80 yet (one is coming my way, but it is still in transit. Woohoo!) So I have a few questions some of you might be able to answer.

One of the things I noticed is that the 74LS93 (IC21, char row counter) has an additional pin (#CKA, pin 14) which seems to be unconnected. Is this correct? I know the first part of the counter is not used, but I guess a floating input is not the best solution.

According to the schematic, A6 is connected to #INT. So why does the edge connector still has both of these signals available? Any idea? Btw, my CAD program does not allow two nets with different names to be connected without merging both nets. So what I had to do is rename it to A6/#INT. So this signal is present twice on the edge connector now.

There are some unpopulated parts on the PCB, TR1, R36, R37 etc. I haven't seen any schematic yet that explains what these were intended to do. Any idea?

I hope someone can help me out. For now I still have to rely on the documents that are available. Hopefully my ZX80 will arrive soon...!

[msellan]

I'm in the process of redrawing the schematics and PCB layout of the ZX80 in my CAD program. I know this has been done before and I have seen the great job that Grant Searle did, creating foils so you can build your own. But I've thrown away my UV exposure box and chemicals as soon as professional PCB manufacturing became available .

Hello, that's awesome! I've been hoping someone would come along and make gerbers for the 80. I'm in way over my head in terms of projects to even think about it. Unfortunately, I don't have answers to your questions. But there's a lot of experience here, so I'm sure someone will chime in.

Best,

-mark

[MaFli]

Hi Mark,

I just left a comment about your vacuum forming idea for the case. What a coincidence Great idea btw.

But yes, I figured this would be the right forum! Already found quite some info that helped me to complete the schematic. I was hoping Grant Searle is still reading this forum. I tried to send him a message, but the email address on his website seems dead...
I could use his foil design (as a vector, not as a bitmap the way it is now) and place that on a separate layer in my CAD program so that I can make sure I'm not missing any tracks.

But I am getting there. I'm just wondering how accurate I should try to make it. Components placement is as close as I can measure it on Grants foils. The only thing I am contemplating is how accurate I should make the tracks. 45/90 degree routing is a lot easier... But let's see if I can get it working first. Once I know my netlist is flawless I can still tweak the PCB to make it closer to the original.

[MaFli]

Found some more minor inconsistencies. According to the Assembly Instructions both RAM chips were different. Found the same thing for the 1-3K RAM pack. The chips were referred to as 2114L and 2114L-1? But with the additional note that "*In case of supply difficulty may be same type as IC4".
Does anyone know what the differences were about? Speed grade?




Also C12 is 47nF according to the schematic and 100nF in the Assembly Instructions. Not a big deal I think, since it is used for the EAR socket. Should receive my ZX80 soon so I can check. Unless someone wants to have a peek (no pun intended) on their board

[1024MAK]

Hi MaFli, welcome

One of the things I noticed is that the 74LS93 (IC21, char row counter) has an additional pin (#CKA, pin 14) which seems to be unconnected. Is this correct? I know the first part of the counter is not used, but I guess a floating input is not the best solution.

74xxx and 74LSxxx logic is not normally too bothered about unconnected input pins. Due to the design of the chips internal circuitry, the input floats to a logic high. It's not recommended to leave inputs unconnected though.

[According to the schematic, A6 is connected to #INT. So why does the edge connector still has both of these signals available?

Sinclair has provided all the Z80 CPU signals on the edge connector. Yes, those two are linked, that is required in order for the display system to work. I suppose if a user did not need a TV display (custom ROM), the relevant track could be cut, and so seperating A6 from /INT. Then there would be two separate signals...

Mark

[msellan]

Hi Mark,

I just left a comment about your vacuum forming idea for the case. What a coincidence Great idea btw.

Hi again! This is good timing.

I just got back from another class - this time on injection molding. It was cool to actually see and experience the difference in the manufacturing techniques between the two (vacuum forming being the other).

I see why Sinclair used vacuum forming to make the first ZX80 case - way simpler and cheaper to get started. I wish I knew more about how they did the final milling and finish work. That part is still a mystery and something I still need to learn about.

Best,

-mark

[MaFli]

Tedious work, but getting there... Vectorised bitmaps of the original ZX80 PCB and used that as an overlay to verify my netlist. So far so good...

[1024MAK]

Keep going...

Mark

[MaFli]

Keep going...

Mark

Haha... Trust me It is 6:35 in the morning here and still going! Nearly finished. Just a few more larger copper fills and I should be ready.
And still not quite happy with the key pad footprint. Will redraw that too I think.
Below some renderings of the top and the bottom.

Top.PNG (156.83 KiB) Viewed 3013 times

Bot.PNG (95.68 KiB) Viewed 3013 times

I've also traced down the components that are not populated (R36, R37, C3, TR1, D1, D2). Looks like they tried to fix the back porch of the video signal. Does anyone have any idea? Even more weird is the D1, D2 divider. Were these intended to be zeners or other clamping devices in case something went wrong with the power rails? And why is the center point between the two connected to a dummy piece of copper in the middle of the top layer? That is the long rectangular bar in the middle. Clue anyone....????
I did route these tracks anyway. So it is as close as possible as I can match it to the original ZX80.

CompVid.PNG (22.15 KiB) Viewed 3013 times

[sirmorris]

ESD?