New project: An ACT Apricot F1 "barn find"

[stephen_usher]

Oh no, that's rubbish :(

I know. As a last resort I may try refreshing the solder on the pins, just in case it's a dry joint which comes away when things warm up slightly.

[rubber_jonnie]

Oh no, that's rubbish :(

I know. As a last resort I may try refreshing the solder on the pins, just in case it's a dry joint which comes away when things warm up slightly.

Definitely worth a go. I have an SIO2SD for my 8 bit Atari's and that suddenly stopped working and that was fixed by a re-flow.

[stephen_usher]

I tried reflowing, with plenty of flux and it made no difference. I can cool a specific set of pins, keeping the freeze spray away from the chip, and that can work too.

However, it's all for nothing anyway as I tested the motor board whilst freezing the main chip and the motor doesn't rotate, it twitches and goes hard into one set of positions, meaning that the stepper motor control chip on the board isn't working properly and isn't stepping. This isn't really surprising given that the chip was the most exposed to the elements of all the electronics.

To be honest I'm surprised that it managed to survive this well given its original condition!

This is a reminder of what the front looked like, with the motor control chip right at the front:

IMG_2245.jpg

[rubber_jonnie]

I tried reflowing, with plenty of flux and it made no difference. I can cool a specific set of pins, keeping the freeze spray away from the chip, and that can work too.

However, it's all for nothing anyway as I tested the motor board whilst freezing the main chip and the motor doesn't rotate, it twitches and goes hard into one set of positions, meaning that the stepper motor control chip on the board isn't working properly and isn't stepping. This isn't really surprising given that the chip was the most exposed to the elements of all the electronics.

To be honest I'm surprised that it managed to survive this well given its original condition!

Damn, that's sounding unrecoverable.

[Tomswork]

What type of motor is in the drive maybe a complete submersion in ipa or hiflow oil


Tom

[stephen_usher]

I’ll have to dismantle the machine again to get the number. However, the motor spins freely when not powered and at least one set of coils is working well as the armature twangs into position when it gets energised. The issue is that it’s not changing the set of coils to cause the armature to rotate, which would be caused by the controller not working.

Anyway, I’m going to have to put this project on hold at the end of this week as I need to start getting a set of machine prepared for display at Lavecon, assuming that it goes ahead.

[stephen_usher]

What type of motor is in the drive maybe a complete submersion in ipa or hiflow oil

The motor is a Sony BHC-2101A

[stephen_usher]

Well, the project is officially suspended now as this morning I've moved the Apricots into storage so I can start preparing machines for an exhibition in a month's time.

There's a lot of stuff in the loft which will be shifted over to the storage unit over the next week, including four Sun workstations, VT220 and VT340 terminals, a green screen monitor, a couple of Mac Pluses and one of the Apple SE/30s etc. etc. etc.

[stephen_usher]

Seeing as it looks like I will probably not be streaming old games next weekend after all I brought the F1 back out of storage and started looking at how to change the data rate the floppy controller expects from the floppy drive.

The Apricot floppies are odd in that they run at 600RPM and not the standard 300RPM that all later floppies ran at. This meant that the data rate was double (500 bps).

Skimming the data sheet on the WD2797A (a very unusual WD floppy controller chip which supports 8" disks as well as 5.25") it looked like pulling pin 17 (/5-8) to ground with a 220R resistor should enable the internal clock divider and hence allow it to read more slowly. All that actually happened was that it couldn't read anything at all.

So, a more in depth reading of the datasheet was in order. As it turns out the datasheet was for the whole family of four chips and it was only the single-sided versions which had an internal divider. For this particular model you need to change the input clock speed (2MHz for 500bps and 1MHz for 250bps). *sigh* So, it looks like I'm going to have to design a daughter board which plugs into the socket and adds a flip-flop clock divider. At the same time I can put an inverter on the /READY line so that it should work with the later drives which have /DSKCHG on that pin. This is in effect the opposite sense as when you change the disk it briefly not ready. The Apricot does actually use this as a way of determining if the disk has changed really.

Other than this I corrected the floppy adapter cable wiring. The machine does only see one drive now, which is correct and it seems to read faster now too.

[stephen_usher]

Oh, and the Apricot-to-Shugart floppy socket adapter boards have just been shipped by PCBway, so they should arrive either Tuesday or Wednesday.