Acorn Electron - odd fault. Any suggestions?

[SAHBfan]

Hello,
I'm new to the board, so I hope I am posting this in the correct place.

I have taken my beloved Acorn Electron out of storage and powered it up for the first time in several decades. It almost works, but not quite...

Initial checks were to disconnect the power supply and check for +/- 5volts - all appeared well, voltages stable and within spec. I then blew out the dust and gave it a good clean. Everything looked OK, so I plugged it into the Scart socket of an old TV and powered it up. It beeps, which is very promising - but after about 30 seconds or so sometimes gives a second 'beep'. I don't remember that it was supposed to do that! There is no display, until I press break, when I get this:

Acorn Electron
Basic
>
>
>
>
>

The 'Greater than' sign cursor keeps scrolling down the page until the 'Acorn Electron' header scrolls off the top of the screen. It doesn't keep doing this, it produces maybe 20 or so cursors, then a few blank spaces, then stops. If 'Break' is pressed again, it repeats the same behavior, reasonably consistently.
It doesn't respond at all to any key press other than 'Break'.

It appears as if a key is stuck, probably the 'Return' key. Simple as that... except it isn't
If the Electron is powered up with the keyboard removed, it does the same thing! Obviously I can't press 'Break', but easy enough to initiate a break - I can even do it by earthing one hand and touching the keyboard connector with a finger - this initiates a reset and produces the above behavior.

To cut a long story short (I've spent a few hours with a logic probe, multi meter and schematic) I decided to replace the DRAM - for no really good reason other than RAM chips from this era are notoriously unreliable. I de-soldered the old chips, soldered in four sockets, fitted new RAM chips - and then spent a several hours fault finding and repairing all the new faults I'd put on the machine by damaging tracks - At this end of this exercise I was back at square one, with the electron behaving the same as when I started.

At this point I would welcome any suggestions or advice, before I do any more damage to the board!

Thank you

[maniacminer]

What revision of the board is it? Is the ULA (large square IC) soldered into the board or is it in a chip carrier/socket?

[1024MAK]

Electrons don't like having no keyboard connected. It confuses the OS ROM chip...

A poor connection on the ULA (on one or more of the pins used for the keyboard), broken PCB track, or faulty keyboard cable may also confuse the OS ROM chip.

Mark

[SAHBfan]

Hi,

Thanks for the replies.

It is an iss.6, with a 'blob' type ULA, soldered in.

I was hoping booting up without the keyboard and seeing the same fault might eliminate the keyboard as a possible cause.... perhaps not.

I did do a continuity check on the keyboard cable - it seems to be ok, at least as far as the plug. Maybe I'll check from the keyboard PCB to the ULA next?

Steve

[1024MAK]

This blog post may be helpful (not mine).

Mark

[SAHBfan]

That blog is very useful, thank you. I will have a rummage in my electronics bits box and see if I can cobble together one of these 'keyboard replacement' circuits he suggests. I'll report back.

[Bitstik]

I found simply spraying contact cleaner on the PCB pins and gently connecting and removing the keyboard connector solved my issues on an old elk.

[SAHBfan]

Hi,

I cobbled together a quick keyboard substitute circuit - just a few pull up resistors and LEDs for power and caps lock as recommended on the Tynemouth software blog. The switch is the 'Break' function.

The behavior with this board in place was.... exactly the same as with the keyboard. On power up it beeps. After pressing the break switch I get the 'Acorn Electron' logo, which then scrolls off the screen with > cursors.

I suppose that eliminates the keyboard - and the RAM has been replaced. Any advice as what to look at next gratefully received!

Steve

PS. Being an issue 6 board, there are very few connectors apart from the keyboard that I could clean (the ULS and the other chips were all soldered in) - and in my case cleaning the keyboard connector didn't help

[SAHBfan]

Here is a video of what happens on the screen when 'Break' is pressed.

[daveejhitchins]

Check the voltages on each of the keyboard lines (do you have another Electron to compare against*?) - Also check the NMI and IRQ lines. After that I'd be removing the ULA and fitting sockets for it then trying an alternative (I can "loan" you one if you think it would be useful?).

* I'm really busy at the moment getting ready for this weekend's ABUG - but I'll be taking my 'scope so I'll get you something to compare against whilst there - Send me a reminder (here) on Saturday else I will forget

Dave H.

[SAHBfan]

Hi,

I have been a bit busy with my 'other hobby' so haven't looked at the Electron for a couple of days, but I'm hoping to check out all the voltages on the ULA later [/hint @Dave H]

[daveejhitchins]

Hi,

I have been a bit busy with my 'other hobby' so haven't looked at the Electron for a couple of days, but I'm hoping to check out all the voltages on the ULA later [/hint @Dave H]

Had a quick look this evening. The 4 data inputs to the ULA ,from the keyboard, should all be high. Checked my ULA fault log - I've had 3 ULAs that exhibited keyboard faults - I suspect this in another!

Dave H.

[SAHBfan]

Hi,

Sorry for the hiatus - the Electron was shelved whilst I tackled some other pressing activities. Thanks for checking this.

I did find it quite tricky locating the right pins on the ULA. First, deciding which is pin 1 and then counting in the right sequence.
Once I'd worked it out it seemed logical - odd numbers on the outside and even on the inside - until you get to a corner and then it flips the other way around with a couple of the pins not being where I expected them to be

If you look at the tracks coming out of the blob it all makes prefect sense - but alas my eyesight is not what it once was. I made a diagram of the pinouts, as I couldn't find one anywhere - and a table of the signals and pins, just to make my own life easier.
I'll attach them in case they are of any use to anyone else.

What I discovered during this exercise was that every ULA pin was connected to what it was supposed to be connected too - no broken tracks or dry pins.

I'm off to check my keyboard data inputs

[paulb]

I made a diagram of the pinouts, as I couldn't find one anywhere - and a table of the signals and pins, just to make my own life easier.
I'll attach them in case they are of any use to anyone else.

Very useful, thanks! The pin numbers are, of course, provided by the circuit diagrams, but grouping them together in tables can be convenient. See the KiCad circuit diagram in case I didn't mention that here, too.

[daveejhitchins]

I used the pins on the keyboard connector - they are indicated on the Electron schematic!

Dave H.