• Please review our updated Terms and Rules here

Oric Atmos 48k KB repair advice request

M

Macshrike

Member
Joined
Feb 27, 2022
Messages
10
Ls,

I'm rebooting this old, but newish, Oric Atmos 48K.
At first I blew IC1 by using a wrongly polarised PSU. Replaced that.
Some keys were failing, opened it up, inspected and cleaned it. 1 key started repeating endlessly, now no keys work.
The board looks pristine, no breaks, bends, colorations or nothing. Haven't physically tested that but I will.
Before that I was wondering; the sound chip on there seems to be really hot.
That is linked to the keyboard as well.
Could that chip failing also cause these problems?
Regards,
Mac
 
Sure, it's in charge of handling the keyboard columns; but it can also be the 4051 present in the keyboard that handles the rows. Being far easier to replace the 4051, I would start with it...
Btw, if you need to replace the AY, beware of using a socket as then, it's probably too high to fit the PCB inside the case.
 
Hi jltursan,
Thank you very much for your reply!
Apologies for the late thank you, I have been very busy trying to repair and sell my old Amiga which seems to degrade every time I start it.
Nevertheless; thank you again.
And if I may be so bold;
I am very new to getting hands on with microelectronics, I'm more of a software guy.
Having bought this cheap 40Mhz analogue scope trying to find faulty chips, I find the learning curve steep.
I found the service manual for the Oric Atmos and can now understand some of it.
Could you perhaps give me a few pointers about checking the chips.
For 4051 I believe I can check this by pressing keys and scoping the address lines (very difficult with an open upside down keyboard =)
The AY; what would be the fastest way to find if it is faulty? I could replace both straight up but my soldering skills are still not up to par and I would like to remove myself from the error factors =)) Also, my budget has become quite limited.
Also: are these chips findable? I'm cleaning and repairing it as a last farewell to my father who passed away.
I could sell it as a brick but that feels un respectful.

Thank you and regards,
Mac
 
I consider myself a mechanic more than an engineer; but let's see:

1) Check this page about Oric's keyboard internals. There's a great schematic of the keyboard subsystem and explanation.
2) The keyboard depends on:
- The keys themselves. Can't be bad all of them...
- The AY-3-8912. Manages the column selection (check the schematic in the aformentioned page). If I understand how it works, you must ground a column pin to select it.
- The 4051 multiplexer selects the row using 3 bits (0-7)
- The VIA (6522) gets and process the 4051 output so it can also be bad; but if VIA is bad, more errors are expected: bad screen, random hangs, etc.; so seems yours must be working, isn't it?

So, now I'm educating guessing a testing procedure. With the keyboard unplugged (no 4051 output), if you ground one of the column pins of keyboard connector PL3, you must force a column selection into 8912 and then faking some key inputs. If you follow the matrix, you can even simulate an exact key press grounding the right column and row pins combination.
This way, if you are not able to get some feedback in screen, you can guess if the 4051 is faulty or not and leave the 8912 as the most probably faulty chip.

Take this with a grain of salt, I've never tried it by myself. I remember having a bad keyboard and first socketing the 8912 (bad move!) and then, replacing the 4051 to finally fix the Oric... :-(

Also, do you have a EPROM programmer?, if so, maybe you can try with the Oric diagnostics ROM. it doesn't need a keyboard and you can test the PSG/VIA with it, worth trying it.
 
Aha! Found it. Last pin on last Ic didn't have a clock. It looks like this machine has already been salvaged for that chip. It looks very battered.
Alas it's IC 23. The Prom.
Do you think this can be found somewhere?
 
I doubt it; but...what's your fault now?, I mean, does your PCB has an IC23 or is it empty?, if it's the last, as it's part of the video encoding circuit and not anything related to keyboard faults it doesn't matter.

About IC23, the PROM contents has been already dumped and I think that using a GAL you can replace it. Anyway, IC23 is only used to generate the signals that are being used by the RF modulator. If you're using the RGB output (I guess you're indeed using it), you don't need it to make the Oric work fine.

Does your keyboard works now?
 
Hi jltursan.

I took ic23 out.
Kb is totally unresponsive.
The 4051 scoped out normally.
I traced all the kb traces and they connect.
Resoldered the kb connector.
It gave a repeating u plus keyboard sound before it died. I think I had connected it with the wrong empty pin on the board.
Now it intermittently starts up.
Perhaps some rams are failing.
Nothing from the keyboard alas.
Didn't recap it but they look good.

Thx for your time.🙋🏻‍♂️
 
The Oric has a simple reset circuit and there's a cap that usually tends to fail, C21, If the Oric doesn't fully boots, try to force a reset grounding the RESET pin in the Z80 and check if it finally boots. If so you need to replace the cap even if it looks fine. Check this nice reading about some Oric quirks: Oric-1 Repair Part 1 - Reset circuitry

So, IC23 is not there, no harm then. The 4051 looks good, ok, also traces and connectivity; but...if you've connected keyboard in the wrong position, that's really bad as there's a 5V pin in there. You can have something fried now; hope not, if it boots ocasionally, maybe it's still ok. About the keyboard, the AY-8912 still looks suspicious to me...
 
I suspect, from my quick look at the schematics, that the keyboard works the opposite way to what I read in previous posts.

I am thinking that both the row and column signals go TO the keyboard, and the only signal that comes FROM they keyboard is on the keyboard connector pin 13. This goes (via transistor TR2) to IC6 pin 13.

The column lines select one column of keys. The row lines select one row of keys (i.e. the combination of the row and column signals select a unique key). This key state is converted via TR2 into a logic signal for the processor.

The 4051 can be used as an 8-way signal multiplexer or demultiplexer.

My first question would be: with the keyboard unplugged, do you observe signals on the row and column signals?

Dave
 
You must log in or register to reply here.
Back
Top