I recently picked up an Apple II+ clone, the type with the built in language card and Z80 softcard, and it wouldn't boot at all, only displaying rubbish characters. I replaced some 74161 counters that were dead and installed a modern power supply, and the computer still wouldn't boot. After scrubbing the ROM card contacts and unplugging the keyboard, I heard a beep and saw the BASIC prompt on my television! However plugging the keyboard back in caused the speaker to beep rapidly and a different rubbish pattern appeared on the screen. Rebooting the computer and unplugging the keyboard returned me to the BASIC prompt. I tried spraying deoxit into the sockets for the keyboard and character ROM but it didn't help. The prior owner of the computer installed the shift-mod, but unplugging that didn't do anything either.To sum up, the symptoms when the keyboard is plugged in are:
- Rubbish graphics on screen (continuously scrolling)
- Speaker continually beeps
- ctrl+reset causes the beeps and scrolling to stop, but trying to type doesn't seem to do anything
Troubleshooting attempted:
- Keyboard and character ROM sockets treated with deoxit
- Shift-mod unplugged
- Holding ctrl+reset while booting
If it weren't for the numpad I would have bought the Cherry MX compatible replacement, since the SMK switches in it genuinely feel awful. Looking for some advice on how to diagnose the issue, thanks!
The issue is definitely not in the character ROM. It's either one or more stuck keys, a bad keyboard encoder (if it has one as a separate dedicated chip), or some other chip on the circuit board that is part of the keyboard. Can you post a picture of the keyboard's circuit board?
One thing to try is boot with the keyboard unplugged and then carefully plug it in while the computer is on and see that happens.
Hi CVT.I've attached a photo of the sole chip on the keyboard, and the output I'm getting out of the composite jack. I plugged the keyboard in after it had booted to BASIC but got the same rubbish printing to the screen.
Yep, you probably need a new encoder. Not sure where you would find one though. There is another user in this forum that has a keyboard with that chip: https://www.applefritter.com/content/trying-identify-my-apple-ii-keyboard
You could also contact Joe and ask him if he has mapped this chip, so that it can be programmed into his universal encoder:
https://jcm-1.com/product/jcm-universal-keyboard-encoder/
I've got a clone that uses that same case and keyboard. The one I have has a different mother board though because it is a more standards ][+ clone without the Z80.
It would be a good candidate for someone who was good with KiCAD to design a version of the URK (Universal Retro Keyboard) project that fit in that form factor with the numeric keypad.
Knowing that the keyboard outputs standard ASCII keycodes I'm tempted to try and build a new encoder using an arduino or Rpi pico, but I'm not sure about the timings tha the board needs. Unfortunately Joe said he'd not run across the controller before.
Using your oscilloscope you could check if the encoder's clock is running. It should be on one of the single digit pins, around 3, 4, 5, etc.
I agree with CVT. That little picture of the encoder chip shows what looks like a capacitor and a crystal. Those are probably part of the clock circuit CVT mentions. You can probably trace to the pin where the clock signal goes in.
I went to put my hook probe on the end of the crystal oscilator and it just fell off! One leg had already snapped and the other was mostly made of rust. I'll order a new 3.579mhz crystal and see if that solves the issue.
The chip will be a microcontroller, as you can see printed on the keys of the keyboard there are multiple fuctions per key. When an option is selected it would send the key presses making up the function selected.
Hopefully it's just the crystal.
If I read the post correctly, that isn't a photograph of the actual computer in question, just another A2 clone with a numeric keypad.
It would make more sense for BASIC "macros" to be generated from code running on the host 6502 and not the keyboard. Computers such as the Sinclair Spectrum did that.
That isn't the way those keyboards worked. They did it on their own. In fact you could buy the version w/o they keypad and install it in a genuine ][+ and it would generate the macros w/o any firmware changes on the Apple. I did that on my original ][+ back in the day.
One thing I noticed while looking at pictures of other keyboards is that the ribbon cable in mine appeared to be plugged in backwards. I don't know whether that's a quirk of the clone or if the prior owner plugged it in the wrong way. The keyboard has clearly been taken off before since it has mismatched screws.