This was done with an Apple //c upgraded to ROM 0 and tested with:
- Unidisk 5.25
- Unidisk 3.5 (Liron) - see Booting a Unidisk 3.5 below
- Floppy Emu in 5.25 mode - see Booting a Unidisk 3.5 below
- SD Smart Drive - caveat: see Booting a Unidisk 3.5 below
The intended audience
- Has an Apple //c (not a IIc Plus)
- The IIc Plus has a 3.5 internal drive. The IIc Plus boot order seems to be: internal drive; Slot 5, Drive 2; the first 5.25 drive.
- Is willing to make hardware changes to the //c
- Has or wants an Apple II 5.25 disk emulator that is not like a Nishida Radio UNISDISK Air c
- The UNISDISK Air c is said to have a great feature where the user can select whether Drive 1 is the internal drive or is emulated
Consequent ROM revisions include SmartPort (e.g. Unidisk 3.5 support) and using PR#7 to boot was dropped. (Well, PR#7 was first changed for the never released SmartPort based AppleTalk printer box. In time the "memory expansion" [RAM disk] //c needed a slot....)
Thirty years later....
The creation of solid state 5.25 inch floppy disk emulators for the Apple II family has been a great development.
One member of the Apple II family is at a disadvantage with the 5.25 floppy emulators. The Apple //c (not the IIc Plus) has an internal 5.25 floppy drive hardwired as Slot 6, Drive 1 and always tries to boot from it first.
Being able to select the whether the Apple //c internal drive is used or not would be really convenient. Being able to have two external 5.25 drives would be a bonus - Ian Kim's SD drives support two 5.25 drives.
This is how I added an Apple //c Drive Switch.
If you find any of the following disturbing, this is not for you
- Soldering jumper wires to pins
- Desoldering
- Completely disassembling an Apple //c
- Including motherboard removal
- Drilling a hole in the //c case and installing a switch
- Cutting traces on the motherboard
The SmartPort code gets stuck in a loop when there is no 5.25 Drive 1. This should not be a problem for emulators that support both 5.25 and SmartPort drives.
A SmartPort device with a daisy chain port can boot when a 5.25 only emulator is plugged in - assuming that the 5.25 emulator does not have a bootable disk image.
Vintage SmartPort users: I am not aware of any SmartPort emulator that works with a Unidisk 3.5.
Disk emulator time to start
Newer (not a 5.25 controller card) Apple II disk bootstraps assume drives are available when the power is turned on. Most disk emulators need some time to load images, so expect to:
- Turn on the Apple
- Wait for the emulator to get setup
- Reboot
Act like a normal //c
- or -
All disks are accessed via the DB-19 disk port - like a IIGS without Apple 3.5 Drive support
The implementation
- Switch the internal drive enable
- For the DB-19 drive port, switch #9 and #17 between:
- external interrupt + drive 2 enable
- drive 2 enable + drive 1 enable
The switch
A three pole double throw switch (3PDT) is required. I used a C&K 7301. Get a switch with solder contacts. The switch I have seems to be for quick connectors and the contacts are quite long.
Toggle switches are the easiest to mount on a case.
Tools
- Screwdrivers to completely dissemble a //c (not listed because I have a grey market //c)
- Continuity tester
- Some type of stiff conductive wire to insert into the DB-19 port for continuity testing. D subminiature pins are ideal; a bent paperclip usually works.
- Wire cutters
- Wire insulation strippers
- Soldering iron
- Solder wick and/or solder sucker
- Appropriate drill for mounting the selected switch
Bill of Materials
- 3PDT switch
- Wire - I used two conductor hookup wire often used in model railroading
- When using multi conductor wire, make sure you can tell the conductors apart
- Use small gauge wire. A number of the connections are to existing pins on the motherboard
- Epoxy or similar to hold wire to the motherboard
- I used Shoe Goo which was sitting around
- Solder
- Heat shrink tubing for the cautious
- Prudent for the R16 mod
- I used heat shrink tubing due to the length of the switch contacts
The switch location dictates the lengths of the wire needed, so it's best to first start with figuring out where to mount the switch.
I have a grey market //c so am writing with generalizations.
- Turn the //c upside down
- Remove the screws around the outside of the case - the inside screws hold the 5.25 drive
- Turn the //c right side up
- The top cover may or may not include the ports on the back. Start pulling up around the port cover and move towards the keyboard.
- There is a cover connection under the spacebar which I find difficult
- Remove the handle and set aside
- That's where the 65C02 is
- It's the only place with any space. See the photo below (um...yes, I need to clean the keyboard).
- The keyboard has a short ribbon cable connecting it to the motherboard. Move the keyboard straight away from the disk drive only enough for it to be clear of the drive. Rotate the spacebar end of the keyboard up. Disconnect the keyboard's ribbon cable from the motherboard and set the keyboard aside.
- It's probably okay to leave any expansion card plugged in
- Remove the screws from the power supply. Slide the power supply just a bit towards the power switch to unplug it. Lift up and set aside.
- Unplug the floppy drive cable at the motherboard
- Turn the //c upside down again
- Remove the screws holding the floppy drive
- Lift the motherboard up. The floppy drive should be laying on the work surface underneath. There may be metal donuts on the drive or still on the posts on the bottom of the case. Move the drive and donuts aside.
- Turn the //c right side up again
- Remove all of the screws around the edge of the motherboard
- Remove the motherboard and set aside
A reasonable (least bad) place to mount the switch on the case primarily depends on whether or not a //c carry case is being used. The //c has minimal protrusions. From the photo above, the most obvious places would be between the floppy drive and the back, or perhaps between the keyboard and the floppy drive. Neither of these works in a //c carry case.
Having a //c carry case, I stole an idea from the volume control knob: the area under the side of the keyboard. The left side with the volume knob and headphone jack is already crowded so I drilled a hole on the right side.
The hope is that in normal mode, the switch won't protrude. When the internal drive is disabled it would be okay to stick out.
The older //cs have a keyboard brace. The switch hit the brace.
I searched a bit for smaller 3PDT switches and they seemed to be about the same size as the one I already had.
Early //c expansion instructions say to remove the keyboard brace. For a majority of people, this is the best option. Having broken the arm of a Return key on a //e, I figured I needed to keep the brace.
The left side of the //c motherboard has RAM so since I was going to make another hole I raised it up to the corner where the case widens out.
Moving the switch closer to the drive also meant the switch hitting the post for one of the screws holding the top and bottom of the case together.
The test fit...wasn't acceptable.
This resulted in drilling a hole in the support. It was a bit low.
The result was tolerable.
Looking into that screw hole isn't pretty.
Caveat about pinouts
This document refers to a DB-19 connector, two pin headers (internal drive) and dual inline parallel (IWM controller). Each numbers pins in different ways. Even better, much of the work is on the bottom of the motherboard so is inverse!
Connections used
DB-19 #9 - external interrupt/drive 2 enable via one leg of R16
DB-19 #17 - drive 1 enable
Header 14 - internal drive enable
IWM 20 - drive 1 enable
IWM 19 - drive 2 enable
R16 is the pull up resistor used with DB-19 #9. The leg connected to DB-19 #9 should be desoldered and wires run to the through hole and the leg. On my motherboard the leg is closest to SER 1.
Switch wiring
I used three sets of two conductor wire. The sets are:
- IWM 19 & 20
- Drive enable 1: Internal header 14 & DB-19 #17
- R16 through hole (DB-19 #9) & resistor to 5V
Internal drive enabled | IWM 20 red | IWM 19 black | R16 +5V red |
---|---|---|---|
Outgoing signals | Internal drive 14 red | DB-19 #17 black | R16 to DB-19 #9 black |
Internal drive disabled | R16 +5V | IWM 20 | IWM 19 |
I used heat shrink on the enabled and disabled connectors mainly due to the length of the connectors.
Motherboard mods
The least painful way I could figure out how to do this mod was to cut the drive enable traces on the motherboard. The drive 1 enable lines for the internal drive and DB-19 are cut.
- DB-19 #17
- Header 14
Here IWM 20 (drive 1 enable) has a trace to the internal drive and DB-19 #17 (external drive 1 enable) has a trace.
For each trace cut, two wires were run. Do not run wires through the holes for the drive posts. Clearly I need to figure out the runs before soldering.
Since the wires are on the bottom of the board, they have adhesive in a couple of places.
R16 is the pull up resistor used for the eternal drive interrupt, DB-19 #9. One of its legs connects to DB-19 #9, the other to 5V and SER 1. Use a continuity tester to determine the leg that runs to DB-19 #9 and desolder it. Run wires to the through hole and another to the resistor leg; heat shrink is a good idea for covering the connection to the resistor leg.
Since I didn't trim the switch contacts, I bent them. The blue wires are the jumpers between the top and bottom contacts.
I forgot to shoot closeup of R16. Hopefully the entire board view will suffice.
This is normal node, internal drive enabled. (Yes, I know the keyboard needs a cleaning.)
Here the internal drive is disabled. (Look beneath the Delete key.)
How about changing the internal drive to Drive 2 and booting an emulator as Drive 1?
This actually works for 5.25 drives and is easier to wire - and it breaks SmartPort. A workaround is to cheat: disable the internal drive, boot an emulator and then enable the internal drive. (The emulator becomes Drive 2.)
The 5.25 drive number swap is different in that:
- A double pole double throw (DPDT) switch is needed instead of a 3PDT
- No need to do anything with R16