exxos's Atari, Amiga & retro forum

Delving further into the manual for the HX-20, it seems that there are some interesting facts about the CPUs.

They are more like microcontrollers as they have ROM and RAM internally. The slave CPU has 4KB of ROM and the main CPU has no ROM and uses the external RAM/ROM, so it looks rather like they can't/shouldn't be swapped, so I need to be sure which is which.

Also the slave CPU can be asleep - completely asleep - when not dealing with it's assigned tasks, which are:

1. Audio cassette
2. Printer
3. Bar code reader
4. RS232
5. High speed serial (Between it and the main CPU)
6. Cartridges

It can be powered off independently of the main CPU.

So if it isn't running, then it isn't necessarily an issue.

According to some users reviews here, these ones can be quite tough to repair sometimes.

HigashiJun wrote: 05 May 2026 13:50 According to some users reviews here, these ones can be quite tough to repair sometimes.

Yeah, very true, complicated machine and not so easy to figure out.

As it goes the manual does have some troubleshooting some 300 odd pages in, and it gives individual IC pins to check, so I need to look at that.

Wish you a lot of luck !

I found where I bought my replacement battery...


Looks like they're still in stock: https://ledspares.co.uk/products/bst2x2 ... ery-ep3514

stephen_usher wrote: 20 May 2026 11:31 I found where I bought my replacement battery...


IMG_3720.jpeg


Looks like they're still in stock: https://ledspares.co.uk/products/bst2x2 ... ery-ep3514

Brilliant, thanks. I had some cells arrive just before I left for France, so will build a pack (I have enough for two) when I get back. I may end up buying a couple of these though if it doesn't pan out.

OK, so I finally got round to building a new battery pack and I have tested it in the 2nd HX-20 and I get this:


Considering I get weirdness on screen if I just power using my bench PSU connected to the battery connector, this suggests that it doesn't like being driven by anything other than a battery. OK, so I have a good battery now and cells to build another so that is positive. The battery seems to be charging.

What is slightly less positive is the corrupted display, so I think there is another issue here in that perhaps there is bad memory, or a ROM because every option I try drops you into the monitor.

However I am thinking that this is a good starting point because I can work on the missing lines in the display - that's likely just a connection issue, and I can then try at least probing the semi working machine.

Also It means I can swap the displays over and confirm if the one from the original machine works fine. I can potentially also swap over the CPUs and other ICs to the partially working machine and verify functionality.

I'm also wondering if the caps are bad on this one too.

Small amount of progress but I'll take it.

Machine 1 display works perfectly with machine 2.

Thats a good step forward, though machine 2 still has corruption of the menu items, it isn't missing any lines on item 1 of the menu.

So I need to figure out if there is a RAM or ROM issue with machine 2, but I'm happy with what I have so far.

OMG!!!

Machine 2 is working!!!

I followed the instructions buried deep in the service manual and it said to use CTRL+@ to initialise. I thought this only set the time and date, but it looks like it fully initialises the machine from scratch.

It works, so I now have a known good machine I can refer to to work on the other :)

I need to revisit the 2nd machines display that has a couple of missing lines, but that should be a quick easy fix (He said hopefully!!!) From a positive perspective I will at least end up with a single working machine after this, though I'm hoping for two :)

rubber_jonnie wrote: 26 May 2026 18:29 OMG!!!

Machine 2 is working!!!

:cheer: :girldance: :cheer: