All the motherboard issues were resolved in my previous post in this series, now it is time to reassemble and see if the hard drive works.
The fan on the CPU of the accelerator board likely wasn’t great at the time, but it is quite noisy now. Time for a change. I removed the fan and have a Noiseblocker XM1 to replace it
The old fan was soldered directly to the 12v line on the Zorro II card edge. I definitely wasn’t going to do this on the replacement. There was a minor problem, though. The screws were slightly too short for the new fan. That’s OK, I thought, these look like M3s, let’s try some longer M3s.
You might be able to see the problem in that photo, for those who can’t, the thread is different on the original screw, so it wouldn’t go in the mount holes properly.
The card was made in America in 1991. That means it was highly likely using a different standard for screws. Sure enough, when measuring the thread, I found it was a UNC #4-40 thread, and the screw was 5/16″ long. I didn’t have UNC screws in stock, beyond what is typically used in computers. So, I ordered some #4-40 3/8″ screws. These fit quite nicely.
I used a Molex to fan cable to hook this up with power. This will make the fan replaceable in the future.
I wanted to mount the battery somewhere accessible if the lid is taken off. The drive bay won’t have much space if I add a 5.25″ drive in the future, so I decided to mount it to the right-hand side of the motherboard.
Now that the machine is part-assembled. I decided to run some tests. The 68040 card was working fine, and the RTC battery appears to be working too.
Now that we have a working machine, let’s try the SCSI card.
I tried booting it and just got a Kickstart “insert disk” screen. I popped in a SysInfo floppy disk to check the drives and SCSI card autoconfigure.
The card was not being detected at all, despite spinning up. I cleaned the Zorro card connector for the SCSI card. This made a bit of a difference, now I’m just seeing a grey screen, after a couple of minutes, there was no change.
I then removed the drive from the card, hooked it up with a USB SCSI device to my laptop and took an image of the drive. It looks a bit cobbled together, especially because I used my Greaseweazle to power the drive, but it works.
Once I had the drive image, I set FS-UAE as an approximation of what this machine is, and tried to boot it.
OK, well, that explains the grey screen I was seeing on the machine itself. It is the disk’s partition data at issue. I looked at the hex of the disk to see if I could figure anything out, and tried running tools on it.
It kinda looks like an Amiga disk image. But something is up with it. I’m not an expert on the RDB format to know why this isn’t working. So, I’m giving up on trying to restore the data for now.
I have an Oktagon 2008 SCSI card spare, which I prefer to the one that came with the A2000. So, I decided to put this one into the machine. I’ve also acquired a BlueSCSI v2 for it so I could put hard drive images on there. BlueSCSI can also provide WiFi based networking to the Amiga.
I 3D printed this design for the 2023.10a version of BlueSCSI, which is a slightly different layout to the previous versions, particularly where the Pi Pico is fitted. This is the end result.
Using FS-UAE on my computer, I created a hard drive image of an AmigaOS 3.2.2 installation. I then used my EPROM burner to create a Kickstart 3.2.2 ROM.
The SCSI card with BlueSCSI were inserted into a Zorro II slot, along with an SD card containing the Amiga OS image.
When hooking up the LED, it is important to know that colours for the A2000 HDD LED cable were meaningless to Commodore, in this case, red mean cathode. It can be random.
The HDD LED cable was soldered to directly to the old SCSI HDD. I cut it off and crimped a standard 2-pin DuPont connector. I could then connect it to the activity LED pins on the Oktagon.
This gave me working indicators on the case once I’d put the module back.
With thanks to Retrofied, I have a replacement for the Amiga 2000 case badge. I stuck this on, not perfectly straight, but good enough. I also removed the 040 sticker, it was peeling and damaged. I might replace it if I can find a copy of it somewhere.
I used my Greaseweazle to write the installation software for the 040 accelerator onto a disk. This is needed to get the onboard RAM above 2MB working. Once installed, we have over 7MB of the total of 9MB in the system available to play with. The rest taken by the OS and caching.
Excellent! I ran SysInfo again, just to make sure we are still getting good numbers.
This Amiga 2000 is done. It now runs almost completely silently, and is probably the fastest Amiga 2000 I’ve ever seen. Judging by the AIBB benchmark, the RAM speed is much faster than other 68040 cards. This is likely due to the Fast RAM above 2MB being 32bit instead of 16bit, which is quite rare for an Amiga 2000.
The final result is a quite nice-looking machine. It currently has my Amiga 3000 keyboard connected, as I need to clean and retrobright the one that came with it. It also has a mouse I prefer instead of the stock tank mouse connected. But it all works!
With this Amiga 2000, I previously got it into a state where it would boot…
Last time I worked on Jops, I left myself a lot of work to do.…
I recently acquired an Amiga 2000 for £350 which was in an unknown state, but…
I recently acquired an Amiga 1200 motherboard in a spares/repairs condition for about £100 recently.…
Whilst I do like working with STM32 development boards, some basic information I need can…
When I last left this blog series, the first of Stoo Cambridge's A4000s had gone…