The last days I was working on the final source code for the tape deck CPU whilst I noticed another problem – it was not possible to read the states of the write protection of side A & B and the event when the CD ends was not fired.
It took a while and I had to dismantle the machine again for three times but I was able to trace it down to an design flaw on my replacement board. So I fixed it and noticed the changes for the revision B of the PCB. Afterwards I was able to get the states and to finished the first release of the source code.
When I removed the front panel I also used the moment to replace the CDM2 with another working one. However, the original, which worked last time failed again – this thing drives me crazy. But the second CDM2 was fine and will now be used.
Nice side effect which cost me about two hours was a scratching noise in the left speaker when you touched the device somewhere. Only noticed the noise when running in CD mode. So I disassembled the machine again thinking of a problem of the CDM2 but was unable to find anything. Long story short – a capacitor on the amplifier part had a cold joint. I had to use a short wire to fix it.
When writing the code one challenge was to query the inputs from the CD-Player for auto sync recording. I ended up with handling the pause key completely without events and added an interrupt for rising signals when a new track starts. However, I recorded one CD which simply does not recognise if a new track starts. Usually when the CDM2 jumps to the next track it shows for a short period 2:00 before jumping to the chapter 2:01. At this time the signal goes from low to high. But this particular CD jumps immediately to 2:01. Not sure if this is a problem of the CDM2, the controller firmware version or the recorded CD (PS: it seems to be a problem with the CD since the event get fired for other, especially older, Audio-CDs).
The speed of the tape deck is almost equal to the original. Only some noises are different but this is because of the replacement gear which is not working quite well. There are also a lot of improvements to the original tape handling which I will explain later…
There are two small problems left:
- A segment on the LCD panel is always on. Was not able to find the issue until now and I do not want to tear down the whole machine again
- The replacement gear does not work perfect. Rewind is slow and might stop sometimes. Need to find a better method to print the gear. I cannot recommend a filament printer. Need to try a resin printer next time.
