A5000 - diving in at the deep end!

[rubber_jonnie]

Well, the replacement plug arrived and though I did have to modify the black shroud as it was too long, however the results are great and the mouse now works a treat :)

PXL_20230302_160724792.jpg

PXL_20230302_160732656.jpg

[stephen_usher]

Excellent!

[rubber_jonnie]

Well, blinkin' flip!!!

I pulled the lid off the A5000 because I thought the HDD, which is a SCSI device from an old Mac, was starting to fail, as it was refusing to spin up and occasionally when it did spin up, it would stutter sometimes and cause the machine to crash.

The intention was to fit a BlueSCSI (I may still do so), but it seems that isn't the problem. it's the PSU - again.

Since the drive spins up fine with an external PSU, I checked the voltages on power up and they aren't good - less than 10v on the 12v rail and less than 4v on the 5v rail.

Like the Mac Classic, the PSU slowly starts up and eventually the voltages to get somewhere useful

:pullhair: :pullhair: :pullhair:

Since the PSU has been recapped and had most important components (SCRs, transistors, opto coupler etc) I'm seriously thinking it's time for a modern replacement :(

[stephen_usher]

Given that, from what I understand, switch mode power supplies depend upon resonance I'm guessing that there are components which are out of spec in there, such as capacitors, which slowly get better as they "warm up".

[rubber_jonnie]

Given that, from what I understand, switch mode power supplies depend upon resonance I'm guessing that there are components which are out of spec in there, such as capacitors, which slowly get better as they "warm up".

Well everything I replaced was the same spec as was removed, so I don't know, though I get what you're saying. Also, no schematics don't help.

This afternoon I took out a slightly dodgy Molex connector and soldered in a new one, and also removed the Laser Direct card, which TBH is no good for anything as it needs a very specific laser printer. It has a lot of tantalums and I was wondering if it might be pulling down the machine.

So far since doing that it's actually been fine, but I need to let it sit and for the caps to discharge, see if that makes a difference too. I also want to see if I can set up the BlueSCSI, but like a dimwit I found I'd deleted the SCSI tools from the machine. It's fine, I have them to hand, it's just a bit of a delay.

[rubber_jonnie]

On a side note, whilst it was running fine I hooked up the BlueSCSI which has 2 drive images and got this:

PXL_20230624_172107653.jpg

Whilst by no means 'working' it's a start and is simply likely to be a result of the images being unsuitable for the Archie/Morley HBA.

So, a bit more digging to do on Stardot to see what sort of images they are using there, and perhaps have a look at the setup of the physical drive I have installed, which was originally from a Mac.

If I can make or download an image that is Mac suitable, I may be onto a winner.

[stephen_usher]

Is BlueSCSI known to work on an Arch?

BlueSCSI only implements a subset of the SCSI commands, the ones used by MacOS. I know there are quite a few systems that it doesn't work with because it doesn't support a lot of the status commands.

[rubber_jonnie]

Is BlueSCSI known to work on an Arch?

BlueSCSI only implements a subset of the SCSI commands, the ones used by MacOS. I know there are quite a few systems that it doesn't work with because it doesn't support a lot of the status commands.

Supposedly it does, at least according to the thread on stardot, unless I've read it wrong.

[stephen_usher]

Fair enough. It was just a thought.

[rubber_jonnie]

Fair enough. It was just a thought.

And certainly worth mentioning :)

As it goes, the PSU is now smelling fishy again :(

I'm wondering if there are more cap failures...