Since I got the ZX81 booty I've on-and off been trying to get a Cheetah 16K RAM pack to work.
I desoldered and tested all the 4116 RAM... All fine. Desoldered the three 74LS chips... all fine. Replaced all the caps, which were most definitely not fine! The 10uF 16V caps were testing as ~24uF and with a high leakage percentage. The 220uF one was better, only being about 20% out of spec. After replacing this lot the -5V rail changed from about -3.8V down to the required -5V. Still not working.
IMG_4404.jpeg
Reflowed practically every joint. Checked with the oscilloscope and only A6 looked a little strange but not terrible.
Today I decided to test the machine I was using with the other RAM packs I have, first verifying them on my known good original ZX81. All good, so it's not the machine.
Seeing as the Memopak has a pass-through and you can daisy chain RAM packs to give more memory I tried an experiment. I removed all the chips from the Cheetah RAM pack. This should essentially make it inert. Daisy-chain that on the back of the Memopak and switch on... non-working machine!
Seeing as the lines from the edge connector don't actually go anywhere how can this be?
Checked for shorts, and nothing.
As a last ditched effort I reflowed the pins on the edge connector, making sure to bubble any flux out of the solder. That did the trick!
Obviously the flux they'd used had broken the connection of one or more pins but made a connection via capacitance, screwing up the bus(es).
Replacing all the chips and...
IMG_4407.jpeg
A working RAM pack!
You do not have the required permissions to view the files attached to this post.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.