Trying to fix several issues:
- CPM65 works on the machine in 40-character mode (hacked), the Videx card is the next goal. The 40-character mode still is quite unreliable and crashes the machine, the screen TTY driver is still to be done for apple2.
- the 5V rail is on 4.5V now, causing issues with the keyboard (the length of the cable) and with the paddles. Not clear, whether this is the PSU (which is also showing 4.7V without load) or something on the motherboard. There are two +5V lines on the original PSU and only one on the replacement PSU, maybe that’s a problem?
- the keyboard beeps and issues ‘=’ and ‘/’ characters when connected to a longer serial cable, no problems when connecting the botched wire from the old connector on the keyboard PCB directly to the computer. Some shielding or capacitance issue? It also appeared only after playing with the paddle connectors.
- the new paddle connector with DIN-5 connectors works somewhat with paddle 0, though the nob is not totally up when on 255. This indicates again a capacitance issue, probably with the cable itself?
- two new teac drives turned out to have the very wrong connector for an Apple ][ disk controller, some adapted has to be designed and a PCB has to be printed.
- strange issues with floppy images not working in LinApple or also not on the real computer, always bailing out in the monitor in strange locations. Used Fluxengine and ADTPro to get images, they are identical. Checked the RAM of the machine, no errors. Must investigate if boot addresses are the issue (48k floppy masters not working on 64 or similar issues)
- HGR shows a line on row 0 only partially. This could be an adjustment issue of the CRT image or also bad RAM.
- the CRT shows some flickering and diagonal effects. Testing on an old TV shows the same issue (so I suspect it’s some grounding/shielding issue with the new PSU).
- the CRT is producing some flickering when turning the brightness nob, this should be fixable with some deoxit.
- the second floppy drive (the more used one) now sees floppies always as “write protected”. Diagnostic tests show alignment issues of the drive, this could be true, or the clone always had a different alignment than a default Apple 2 drive?
- the paddle wire is self-made on paddle 1 and quit a mess. Doing a rewire with DIN-5 connectors which then lead to a 9-pin DSUB-Apple ][ standard joystick connector, which will go to the 16-pin DIP socket on the motherboard. The original joystick connector has a completely weird layout and is not usable (it has no Apple 2 layout)
- really bad design to connect DIP 2x8 cables of the joystick and the paddles directly on the motherboard, you pull the cable a little bit too much and you tar out half of the motherboard.
- possibly much more
the paddle and the joystick connectors miss some legs for putting them into the sockets on the motherboard. No essential pins are missing, will rework this anyway as a 8x2 DIP to 9-DSUB, from there to an adapter where you can connect 2 DIN-5 cables, needs rewiring of both paddles. paddle 0 seems to be a little bit shacky (the button works unreliably). DeOxit in the button helped. Works much better now. the lever moving the contacts between 40-character and 80-character mode for the CRT output is corroded, doesn’t always make contact and produces a flickering image. The box is heavily sealed and there is no easy way to fix it. Using an audio switch box for switching modes now (the cables are standard audio-like cinch connectors). the floppy drives read floppies (both) but they don’t write to floppies (when copying in CP/M or creating a file), they work when formatting a drive on CP/M though.So, yes. DOS 3.3 and CP/M both use some ROM routines from ROM1 to write to disk. Works now. Could be a side-effect of the PSU not working properly.The PSU has been replaced, the voltages look ok, so I rule out the PSU. Could be some missing code in the Apple ROM (but that would be weird to call 6502 firmware code from the Z80 CPU). It could also be the write current being to low for writting (but why does formatting a drive work then?!) floppies written read/written with Fluxengine don’t really work either on the real hardware or the emulator. Suspecting a encoding/decoding issue.Updating Fluxengine solved the issue, also choosing 96dpi 5 1/4" for the real floppy drive. :-) the keyboard reset line cannot be taken to ground on the keyboard, works with shrtening ground and reset though just fine. Must be a bug on the ICs of the keyboard controller or a broken trace.Fixed with a botch to bridge the broken ribbon cable between keyboard matrix and the PCB with the controller logic. ROM 1 with BASIC is broken (and gets really hot). This explains the booting issues.. (will be tricky to get a replacement). The board layout doesn’t fit to the original ROM sizes too..Got an AT28C256, no more frying of chips, Apple Basic prompt appears, tested some commands and it seems fine. The 32k ROM has to be filled in starting from 0x6000 with #E000-$EFFF and from $7000 with $D000-$DFFF. If this ROM is ok we will see.
- Reseated and deoxied all ICs, some of them were quite corroded, at
least 3 more turned out to be bad.
Sadly it didn’t have an effect on ROM 1 not getting hot (must be something else).The new replacement ROM works fine and doesn’t get hot. So I actually blame it on the ROM itself causing the other ICs to fry and not vice-versa. power switch made some hissing noises, it was not the switch, now the original power is dead (and some transformers are humming). Ordered a Apple 2 replacement PSU, waiting for it..The new replacement power supply works like a charm, also replaced the power switch, the old one was too unreliable. sometimes I get into the monitor lower-nibble bit 3 of the key being pressed is missing, unclear where it is lost, I suspect a bad register on the motherboard holding the current ASCII value of the key pressed or something similar. Replaced a bad 74LS257N, works now just fine.
Not fixed issues:
The 220V fan no longer works.I just disconected it, I don’t think, it really helps to cool the system.
Known things to work:
- The keyboard works electronically (8039 MCS-48 like CPU with a 2K EPROM), has a new serial cable now and is fully functional
- the Rockwell R6502 CPU has a phase 0 clock and seems to go through address lines and output data (another strong indicator is the ‘Apple ][’ sign on the monitor when switching on which indicates 6502, 40-character ROM, video circuit seems to be just fine).
- Monitor is working just fine (which is sort of amazing for such an old CRT monitor).
- I can boot from a Apple ][ CP/M 56k floppy and I get a prompt, now as the keyboard works and the CP/M on the 80-character card I can also assume the Z80 is working just fine.
- the 80-columns card (a Videx Videoterm or rather a clone of it) works just fine.
- the highres graphic mode works just fine, I can run “Peter Anvin Chess” and play it.
- Applesoft BASIC works fine
- the new Super Serial Card II I got works fine, though it looses characters (without parity and adding some delay in minicom). Definitely usable.