Remaking an Amiga 1000 keyboard - substitutes for tantalum caps?

[ncstebb]

Hi all,

New to this forum... please be gentle. Probably could have put this in Beginner or Vintage Computing but overall seemed better here.

I've been working on a project to remake keyboards for the Amiga 1000 computer. I'll add details below for those interested but for now I'm at the stage where I am looking for components to build the PCB.

I'm a noob when it comes to selecting and ordering components, but I'm getting the impression that tantalum caps are now hideously expensive. There is one tantalum capacitor on my original keyboard marked with 25 on the front and N2 on the back which seems to be a 22 uF 50V cap on the schematics. It runs from one leg of the timing crystal to ground. Unless I've misunderstood something it looks like these go for $6.62 each on digikey, whereas equivalent ceramics are around 0.40. So I'm wondering if I can do a swap? (but also after some confirmation of my suspicions that this is a known issue with tantalums these days.)

Thanks,

Nick (ncstebb)

More details if you're interested.
- specs I'm referring to are here: https://deskthority.net/wiki/images/3/32/Commodore_Amiga_1000_keyboard_schematic.jpg
- I've modelled the original keyboard case and after several prototypes have it working nicely when resin printed.
- I've modelled the keycaps using both mitsumi and cherry mounts
- I've traced the artwork and have produced engraved keycaps which I can resin print, paint and sand leaving something fairly close to the originals
- I've modified the pcb to use cherry compatible switches and have just got 5 of these back from JLCPCB
- here is a photo of where I'm at: https://drive.google.com/file/d/1_Vf5euy-9QX78yrR9vyTgbZR97zGnUdd

[Tomorokoshi]

From here it looks like a 22pF capacitor. Not tantalum. Use a C0G / NP0 22pF capacitor. I can look at my A1000 keyboard to make sure. A capacitor that large wouldn't be used with a crystal.

[Ian.M]

+1
In addition, its certainly *NOT* a Tantalum as its not polarised.  Compare with C8 which has a '+' by it to indicate its polarity.

[ncstebb]

Ahh, so sorry, I was looking at the wrong cap on the schematic. You're right, that one next to the crystal is not a tantalum.

Looking at an actual keyboard, the one I meant seems to connect pin 1 of the NE556N to ground just before a 510K resistor (labelled R1 on the board, but I think this doesn't match the labelling on the schematic). I don't think this cap is on the schematic. The only polarised one I see on the schematic seems to match up with an electrolytic.

Added a couple of photos to illustrate. It's the blue one and has a little + on it and a + marked on one contact on the PCB.

[Kleinstein]

The capacitor at the NE556 does not have to be high voltage. So a 16 V part would be OK. There is also not high current spikes. So the existing tatalum cap is OK in this and would no fail catastrophic. So there is the real option to just keep it as it is.  For 1 µF there would also be film caps as an alternative - overkill, but still not too large / expensive. A MLCC would propably also do, though with drift down in capacitance.

[Tomorokoshi]

What is the lead spacing of the 1uF capacitor on your new board?

[ncstebb]

Thanks for the help. Sorry for the late reply.

The specifications had been confusing, but it's making more sense now. Looks like the 1uF on the specs was replaced by the tantalum on the actual board. This is marked with a + on the Gerber's that I modified to take Cherry switches.
- so the lead spacing of this is 3mm
- if I understand the advice I might try a film cap first to replace the tantalum.

Thanks for the help. If I get it working my plan is to open source it :-)

[eugenenine]

How far are you in this project, I have a 1000 with no keyboard.