New project: An ACT Apricot F1 "barn find"

Blogs & guides and tales of woo by forum members.
Post Reply
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

I'm getting nowhere diagnosing the floppy. I'm going to have to make up a Sony to Shugart cable but to do that I'll need components and it's too late to buy them and get them delivered before the weekend due to the delay in the 74LS21 arrival.

I do hate mail order sometimes.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

Seeing as I'm working from home today I decided to look into glitchiness on the display and crackle on the speaker over lunch.

The voltage on the 0V wasn't 0V relative to the cage ground and was wavering and glitching up to 1V!

At first I thought it was PSU capacitors. Anyway, I removed the PSU and checked as well as I could all the capacitors in circuit and they seemed good enough. Then I looked at the output pins and noticed that the 0V was linked directly to the cage ground, so it shouldn't be capable of having glitching at all, unless there's a bad connection on the connectors.

So, I took each of the wires out of the Molex connectors in turn and cleaned them, and cleaned (and sanded) the pins. There was now about half an ohm of resistance on the wire, which seems a bit high still. Anyway, I put the whole thing back together again and switched on. There's still about 200mV of noise and 600mV voltage drop between the two ends of the wire, which at 0.5 ohms would make about 1.5 amps of current, which sounds about right. Still, I don't like the relatively high resistance of those wires. I wonder if they're really corroded internally and hence the separate strands aren't in electrical contact, increasing the resistance.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

I've left the machine running all afternoon and after the lunchtime tarnish removal (a bit more after my last post, using a file) the average voltage drop on the cable is now a stable 360mV, as measured by the oscilloscope, with the peak-to-peak varying between 120mV and 160mV.

I'm still not totally happy as that will bring the effective Vcc down by a third of a volt and it was already only 4.96V (as measured on the PSU itself).
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
Tomswork
Posts: 775
Joined: Fri Nov 22, 2019 10:05 pm

Re: New project: An ACT Apricot F1 "barn find"

Post by Tomswork »

It could still be a dry capacitor

Tom
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

Not really. I'm measuring between two points on the same wire effectively, at the two ends, so it's really measuring resistance. The voltage drop is a proxy given the current should be the same or similar.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
Tomswork
Posts: 775
Joined: Fri Nov 22, 2019 10:05 pm

Re: New project: An ACT Apricot F1 "barn find"

Post by Tomswork »

Interesting if you replace the wire cut the insulation open see if the copper is oxidadized or or melted

Tom
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

Thankfully after all the cleaning the resistance is now down to about a third of an ohm and it's stable at that, so it's GoodEnough(tm). The Molex power connector is no-longer getting hot now either.

Filing off the tarnish on the power connectors (+5V, +12V, -12V) have helped with those levels a great deal too. The surface of the brass had really oxidised and that layer was really tough.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

Today's been one of those hitting your head against a brick wall days.

I've made a Heath Robinson converter cable for the floppy interface.

IMG_2291.jpg
IMG_2291.jpg (72.61 KiB) Viewed 2586 times

Real floppy drives (both 720K and 1.44MB set as DS0) are seen, sort of, but they don't seek or turn on the motor and are reported as 630K drives, but fail the BIOS test sequence, giving error 35.

A Gotek running FlashFloppy fares a little better. It still fails the BIOS test with error 35 but after a while, with a bootable disk image selected it gives error 38, which is not an error code in any of the documentation. (There is an error 38 for the Apricot Portable, but this is testing the voice synthesis module.) The drive access light also comes on. Somehow the machine detects the drive as a 1.44MB drive (though according to the docs I have the system doesn't actually support that size of drive!).

All very strange.

Anyway, I should be getting a proper IDC connector on Monday so I can look at making a decent cable, which may (or may not) help.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

I think the main is is that the header pins I'm using are too short for the IDC connector and so they're not getting a good connection.

By fiddling around yesterday evening and pushing the connectors together more I managed to get the BIOS to pass its diagnostics with the Gotek and then try to boot from the drive. At this point it gives error 8 (can't find sector), so the previous "error 38" was actually "error 35" (no floppy drive) followed by "error 8" without the display being updated.

Anyway, there's hope that in terms of getting the hardware fully functional to the point of booting to MSDOS is close. I've still got to fully rebuild the case and build a virtual infrared keyboard using a Raspberry Pi to generate the output. Then again, I have put a bid on an Apricot Portable on eBay, mostly for the keyboard!
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
stephen_usher
Posts: 5533
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: New project: An ACT Apricot F1 "barn find"

Post by stephen_usher »

I can now guarantee that I have connectivity on the pins and it's still not reading the floppy. However I have found some things of note:
  • If I have the /motor-on signal connected correctly then the Gotek fails the BIOS drive test. If I have it connected to 0V then the BIOS test passes. With real floppy drives the test never passes.
  • Gotek's don't pull up /CS0 and /CS1 (drive select) correctly. The lines are only being pulled to ~2V. This will possibly affect their use in combination with other drives on the same cable. A 10K resistor (should be 1K for the specs) will pull it up correctly.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
Post Reply

Return to “MEMBER BLOGS”