Trying to restart a 33 years old ZX81 [SOLVED]

[Moggy]

You're not wrong Ian, God bless the modern 5v switching regulator.

[RetroTechie]

I desoldered the original "ROM" (actually, and I'm pretty darn sure of it now that it is proved it lost at least part of its contents: an EPROM without an UV erasing window)

Without evidence one way or the other, that's about the poorest argument I've heard about why some IC would be a One Time Programmable EPROM.

For starters, OTP EPROMs are usually marked as such - with their regular type number. If you're buying in large enough numbers to have manufacturer put a custom marking on the IC's, chances are you're already at a point where mask ROMs become interesting. Mask ROMs often have 'random' type markings that offer little in the way of indicating size, speed etc unless you find a datasheet for the exact part - like this ZCM38818P. I've got a ZX81 with this ROM inside myself, and have seen it in ZX81 board pictures online. Which at least suggests it was used in a large number of ZX81's. Read: large enough number that a mask ROM would be used rather than a OTP EPROM.

We'd like to think of IC's as solid, chemically 100% stable, hermetically sealed devices, with the ones making it out of the factory tested and defect-free. But that's just the ideal case. Actual IC's may contain defects small enough to not cause errors under test - but big enough to fail later in the field. Package materials may be not 100%, but 99,99% sealed for one chemical, 99% sealed for another chemical, etc. Bond wires may not be fully attached. Maybe some impurity made it into the silicon die. And then IC's degrade over time by ESD events (even small ones!), oxygen / ozone / acids in the air, the odd cosmic ray or perhaps some UV wavelengths, thermal cycling, etc, etc, etc.

In short: even solid rocks and the pyramids of Egypt degrade over time, with electronics it just goes faster. 30+ years is a loooong time for any IC.

Anyway: congrats to bringing your Zeddy back to life! If you haven't already, I suggest putting a heatsink on the ULA. It runs pretty hot, so even a simple strip of aluminium to distribute heat away from the center of the chip, will help to prolong its life.

[dinosaur]

I desoldered the original "ROM" (actually, and I'm pretty darn sure of it now that it is proved it lost at least part of its contents: an EPROM without an UV erasing window)

Without evidence one way or the other, that's about the poorest argument I've heard about why some IC would be a One Time Programmable EPROM.

While, during my long career, did see *many* EPROMs (or OTP PROMs) degrading over time (only a few years, sometimes, depending on their working temperature, mostly), I never saw a mask-type ROM dying... On the other hand, the death of this very ROM could be due to something else than progressive erasing, so you might be right... like I could...

We'd like to think of IC's as solid, chemically 100% stable, hermetically sealed devices, with the ones making it out of the factory tested and defect-free. But that's just the ideal case. Actual IC's may contain defects small enough to not cause errors under test - but big enough to fail later in the field. Package materials may be not 100%, but 99,99% sealed for one chemical, 99% sealed for another chemical, etc. Bond wires may not be fully attached. Maybe some impurity made it into the silicon die. And then IC's degrade over time by ESD events (even small ones!), oxygen / ozone / acids in the air, the odd cosmic ray or perhaps some UV wavelengths, thermal cycling, etc, etc, etc.

Agreed... The evidences, here, show that even a Z80A (which can't be suspected of having lost its contents, unlike a so-called-ROM-that-could-actually-be-an-OTP-PROM ), can become non-functional while merely stored during 3 decades...

In short: even solid rocks and the pyramids of Egypt degrade over time, with electronics it just goes faster. 30+ years is a loooong time for any IC.

Sure thing !... Over the time it took for the pyramids to degrade between their building and today, any IC will probably turn into sand (it's silicon, after all)

I just didn't expect seeing such a thing happening in "just" 3 decades for components that were just stored during this time. On the other hand, even the oldest electronic components are very young when compared to the pyramids, and no one (not even their maker) can reliably predict how long they will last.

Anyway: congrats to bringing your Zeddy back to life! If you haven't already, I suggest putting a heatsink on the ULA. It runs pretty hot, so even a simple strip of aluminium to distribute heat away from the center of the chip, will help to prolong its life.

Mine runs around 55°C (not hot enough to burn to the touch), but yes, it's a good suggestion nonetheless. The cooler, the better.

[jmc]

Hi guys,

you made my day when I found your exchange on this forum about that piece of software I wrote so many years ago... HRG 7.0

jmc

[sirmorris]

You made our day when you wrote it..!