Resurrecting A5000

Arc/RPCs, peripherals, RISCOS operating system & ARM kit eg GP2x, BeagleBoard
Pasty
Posts: 16
Joined: Tue Aug 22, 2017 5:34 pm

Re: Resurrecting A5000

Postby Pasty » Wed Aug 30, 2017 12:21 pm

I have tried tracing the IDE tracks with a multimeter. They all seem to check out OK except that my board is a little different - It does not have the chip just above the ROM5 socket and in this area I have 8 Roms (I think that is what they are) so I cannot check that last connection. I have put the board back but sadly I am now back to no boot or video output. Getting a bit disheartened! After I was getting empty drive I started to get disc not formatted. I tried several discs and a replacement FDD (from my RiscPC) but there was no change. When I did get to the desktop the floppy flashed and the motor was working - so I am somewhat lost? I also tried the other IDE socket. The HDD is a genuine harddisc which definitely worked but we are talking about 20 years ago.

User avatar
jgharston
Posts: 2661
Joined: Thu Sep 24, 2009 11:22 am
Location: Whitby/Sheffield

Re: Resurrecting A5000

Postby jgharston » Sun Sep 03, 2017 12:40 am

A few months ago the battery gave out on my A5000, so I resorted to saving the configuration settings and R-Power'ing and reloading them. Last week the hard drive stopped working, just as described above, which is why I've not been updating CoPro stuff much, having to transfer stuff between the PC and the Server via the Master instead of the A5000.

Today was the first chance I've had to dismantle it:
Image
Image

As a stop-gap I've dug out an A4000 and transferred the main drive to it. Geee, it's slow. ClockSp on the A5000 was giving me 125MHz, on the A4000 it's giving me 45MHz. I can almost see the desktop being rendered.

I've got a replacement battery. Will there be anybody at the November meetup with vinegar, a fine soldering iron, and better eyesight than me? ;)

Code: Select all

$ bbcbasic
PDP11 BBC BASIC IV Version 0.25
(C) Copyright J.G.Harston 1989,2005-2015
>_

Pasty
Posts: 16
Joined: Tue Aug 22, 2017 5:34 pm

Re: Resurrecting A5000

Postby Pasty » Mon Sep 04, 2017 8:00 am

Many thanks to all who have tried to help and for a time progress was made and I got to the desktop with a seemingly working CMOS. However having checked IDE tracks which all seemed OK and replaced the mobo in the case I have no POST and no video output. Very strange ! I have checked over several things many times but no joy. My main wish was to get access to some of the software on HDD which I was involved with 20 years ago. I have 1 HDD in a A5000 and 2 HDDs in a RiscPC700 (The RiscPC also has some damage and the A5000 looked better). I have read that you can use an Arc Emulator on a Windows machine. Can anyone tell me if I were to install on a Windows machine would I be able to read the existing HDDs and thus get access to those files and programs?

User avatar
jgharston
Posts: 2661
Joined: Thu Sep 24, 2009 11:22 am
Location: Whitby/Sheffield

Re: Resurrecting A5000

Postby jgharston » Tue Sep 12, 2017 12:48 am

With my A5000 temporarily (I hope) dead, I've got the drives in an A4000. I've been using it to back up the server. Yea gods, the A4000 is slow. I take it the A4000 is the Acorn Electron of the RISC OS world. A compatible machine but with all the memory access crippled.

Normally, I zip up the server to the A5000, I set it going, turn my back on it, glance back and it's halfway through. With the A4000 it was sitting at fighting its way along. I was converting some files with BasConv, which is notably slow and I want to get around to rewriting it, but on the A5000 it hasn't been so slow as to be annoying. On the A4000 I drop a file on it and sit there watching the hourglass count upwards.

Code: Select all

$ bbcbasic
PDP11 BBC BASIC IV Version 0.25
(C) Copyright J.G.Harston 1989,2005-2015
>_

SteveBagley
Posts: 128
Joined: Sun Mar 15, 2015 8:44 pm

Re: Resurrecting A5000

Postby SteveBagley » Tue Sep 12, 2017 7:01 am

jgharston wrote:With my A5000 temporarily (I hope) dead, I've got the drives in an A4000. I've been using it to back up the server. Yea gods, the A4000 is slow. I take it the A4000 is the Acorn Electron of the RISC OS world. A compatible machine but with all the memory access crippled.


The main difference is CPU speed and cache rather than memory access. THe A5000 is running at 25/33MHz while the A4000 is running at 12MHz (50% than the original Arc line mind). IIRC memory access is still at 12MHz in both the A5000 and A4000 -- but the ARM3's cache will help a lot. With the cache on, a well written compression algorithm will sit in the cache and so the only memory accesses needed will be to load/store data. This couple with the faster CPU clock will make a huge difference.

Steve

User avatar
jgharston
Posts: 2661
Joined: Thu Sep 24, 2009 11:22 am
Location: Whitby/Sheffield

Re: Resurrecting A5000

Postby jgharston » Mon Sep 18, 2017 2:09 pm

Well, that's the temporary replacement A4000 also up the futz. Yesterday I'd just finished zipping up the lastest website update when something went Fzzzzttttz and it died.

Code: Select all

$ bbcbasic
PDP11 BBC BASIC IV Version 0.25
(C) Copyright J.G.Harston 1989,2005-2015
>_

User avatar
danielj
Posts: 5148
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester

Re: Resurrecting A5000

Postby danielj » Mon Sep 18, 2017 2:11 pm

Jonathan - have you taken the batteries out of your MDFS? (before anything else gets zapped!)

d.

User avatar
jgharston
Posts: 2661
Joined: Thu Sep 24, 2009 11:22 am
Location: Whitby/Sheffield

Re: Resurrecting A5000

Postby jgharston » Mon Sep 18, 2017 7:03 pm

danielj wrote:Jonathan - have you taken the batteries out of your MDFS? (before anything else gets zapped!)

There's batteries in my MDFS?

Well, yes of course there's _a_ battery. When I last had everything dismantled and took the lid off I took some photos of the inside, and everything looks fine, and the battery is well clear of the other circuitry, and there's been no sign of dodgy battery state such as lost clock or settings.

Code: Select all

$ bbcbasic
PDP11 BBC BASIC IV Version 0.25
(C) Copyright J.G.Harston 1989,2005-2015
>_

User avatar
danielj
Posts: 5148
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester

Re: Resurrecting A5000

Postby danielj » Mon Sep 18, 2017 7:39 pm

:D

Apologies for mis-pluralisation. ;) They can leak whilst still holding charge. Certainly in my MDFS it had done a fair amount of damage :( And in beebmaster's :(

d.

philb
Posts: 93
Joined: Sat Aug 05, 2017 6:05 pm

Re: Resurrecting A5000

Postby philb » Mon Sep 18, 2017 7:59 pm

jgharston wrote:the battery is well clear of the other circuitry


It's a bit hard to see from the photo you linked to, but unless your MDFS has a significantly different PCB layout to mine the battery is actually quite close to some of the other parts including the R/C network that controls the timing of the DRAM address mux. On my unit, although the battery hadn't actually ruptured, it had obviously been leaking a bit because that whole area was covered in green fuzz and the capacitor had disintegrated completely.

User avatar
jgharston
Posts: 2661
Joined: Thu Sep 24, 2009 11:22 am
Location: Whitby/Sheffield

Re: Resurrecting A5000

Postby jgharston » Mon Sep 18, 2017 10:23 pm

Are batteries more likely to leak with the power on or the power off? The MDFS has only been powered down for a few hours since 2011 when I did some rewiring in the cellar.

Code: Select all

$ bbcbasic
PDP11 BBC BASIC IV Version 0.25
(C) Copyright J.G.Harston 1989,2005-2015
>_

User avatar
danielj
Posts: 5148
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester

Re: Resurrecting A5000

Postby danielj » Tue Sep 19, 2017 5:16 am

Age is the real enemy in the end, but flat batteries are more likely to leak, everything else being equal. A 30ish year old nicad is basically a time bomb whatever though :?

d.


Return to “hardware”

Who is online

Users browsing this forum: No registered users and 3 guests