Page 12 of 57
Re: A5000 - diving in at the deep end!
Posted: 04 Jun 2022 14:31
by rubber_jonnie
Ok, so I traced the FDD socket and found puns 18 & 32 disconnected, battery goo having eaten through the topside traces for both.
I've now added a bodge wire for both and the FDD is working 100% having tested with all format options available :D
Time to trace the IDE socket!
Re: A5000 - diving in at the deep end!
Posted: 04 Jun 2022 14:56
by stephen_usher
Yeah, the normal problems with the A5000 battery damage then. :-)
Re: A5000 - diving in at the deep end!
Posted: 04 Jun 2022 15:47
by rubber_jonnie
stephen_usher wrote: 04 Jun 2022 14:56
Yeah, the normal problems with the A5000 battery damage then. :-)
It would seem so, yes.
So far IDE is looking quite bad, I have 6 bad traces, some easy to fix, some not so much, and I've only properly done the odd pins so far!!
A very quick test of the even pins is showing promise so far though, and it is becoming clear that the battery was removed just in time, it could have been a lot worse in that corner. My mind is now wandering back to SCSI drives, as this damage is potentially difficult to fix with bodge wires topside, and I'm not finding any vias under IC49 that I can pick off. More head scratching required.
The new battery is here though and now fitted:
PXL_20220604_133312364.jpg
Though I have to say the rechargeable CR2032 did well in the interim.
Re: A5000 - diving in at the deep end!
Posted: 06 Jun 2022 11:46
by stephen_usher
I was just thinking about your IDE header wiring problem... Have you thought about removing the header from the board and instead soldering a ribbon cable onto the motherboard, with longer wires for the broken traces which you can then solder elsewhere?
Alternatively, a bespoke IDC cable adapter which fits into the current socket and has fly-wires to connect to the places where the broken traces go to and then has a socket into which a normal cable would connect. This would have the advantage of not having to desolder the current socket.
Neither would be pretty, but no bodge would be.
Re: A5000 - diving in at the deep end!
Posted: 06 Jun 2022 13:59
by rubber_jonnie
stephen_usher wrote: 06 Jun 2022 11:46
I was just thinking about your IDE header wiring problem... Have you thought about removing the header from the board and instead soldering a ribbon cable onto the motherboard, with longer wires for the broken traces which you can then solder elsewhere?
Alternatively, a bespoke IDC cable adapter which fits into the current socket and has fly-wires to connect to the places where the broken traces go to and then has a socket into which a normal cable would connect. This would have the advantage of not having to desolder the current socket.
Neither would be pretty, but no bodge would be.
Well, the IDE repair is going to have to wait sadly :(
I left it running for a couple of hours after replacing the battery and it has died on me again. I came back to it and the screen had frozen
So I took some advice over at Stardot and pulled the battery I just fitted (It's quite different to the original 1.2v spec at 3.6v, though well within the spec for the CMOS/RTC chip, but out of spec for the charging circuit)
The battery was one that was recommended, but for now I've pulled it.
I have tried booting with DEL held down, but all I get is a plip from the speaker and a POST code on the floppy light.
Codes are as follows:
Screenshot 2022-06-06 at 13.49.31.png
I have had similar to this before and it was related to the CMOS being corrupted, time to have a delve into the TRM. Hope it's not the IOC dead.
This machine is definitely another Mega ST 4!!!
Re: A5000 - diving in at the deep end!
Posted: 06 Jun 2022 16:10
by rubber_jonnie
Thanks to Ian Jeffray over at Stardot, I'm up and running again.
The culprit? IC49, a 74HCT245T buffer chip in the IDE circuit and very close to the battery damaged area.
So, back to booting up now, but the proof of the pudding is going to be if it boots up tomorrow after an overnight break, which is when it seems to have a paddy!
Parts on order, I still need to finish buzzing out the IDE socket so when the bits get here I can look at resolving the IDE issues by rebuilding the bad tracks.
Phew!!
Re: A5000 - diving in at the deep end!
Posted: 06 Jun 2022 16:59
by rubber_jonnie
Just thinking if I should go SCSI with all the IDE issues I have, and one look at prices for small 50 pin drives is just ridiculous. Luckily I found what I think is an 80GB drive from an old apple machine. Not sure if/how it will work with standard SCSI, but In case it doesn't and wants to be plugged into a Mac Classic again,
So, I also have an 8GB SCSI 2 drive, so 68 pin plug and I've got an adaptor on the way (None of the ones in my parts bin were suitable... GRRR) to try as well. I know this exceeds the IDE partition size for the machine, but SCSI may work differently.
We'll see...
EDIT: Turns out there is an image of the Utilities disk and the manual for the Morley SCSI interface, so both are downloaded and I'll have a look at those next.
Re: A5000 - diving in at the deep end!
Posted: 06 Jun 2022 20:49
by rubber_jonnie
Ok, so taking a look at the FDD system a bit more, I now have 2 working drives connected, which is good news, plus I also managed to extract the SCSI utilities so am about to copy them over to floppy, see what's what :)
Re: A5000 - diving in at the deep end!
Posted: 07 Jun 2022 10:08
by alexh
If you're going to go SCSI you should get a bluescsi (£55) / rascsi (£70) / SCSI2SD v6 (£110).
https://bluescsi.flamelily.co.uk/
Re: A5000 - diving in at the deep end!
Posted: 07 Jun 2022 10:56
by stephen_usher
SCSI2SD would be more likely to work as it implements the full SCSI command set, unlike the others which only implement the sub-set used by MacOS.