To be able to rewrite the code for the CPU I need to understand how the tape mechanics work. Unfortunately I was not able to get the service manual SER-1/5603 which contains the required information. Therefore, I it was necessary to do a reengineering of the mechanics. I surely disassembled the mechanics a thousand times to analyse its functions.
In a first step I checked the capstan motor which is working quite well. But it was not able to switch or move anything because it lacks of power. Of course, after more than 30 years the belt was week because it has not enough strenghness to move the slide servo up and down. At this time I did not had a replacement but I found a rubber band in which the asparagus I bought several days ago was bundled in which looked similar. In the meantime I ordered new belts from ebay and started some tests with the temporary rubber band.
Finally, I also figured out the rubber band prevented the mechanics from some of its movements because it was too tight and too thick. However, the new belts – shipped from Italy – arrived within a couple of days. I can tell the auto reverse mechanics are quite complicated. It took me about two weeks to get the head turning in case of a direction change – and this just by a random accident.
It was during writing the code for the fast forward/rewind function when the head suddenly moved when pressing rewind. Then I understand that the mechanics start to turn the head when moving to the right in full speed to have enough power to move one gear down which enables the lever for the head.
Another challenge was to understand how to move the lever for the record function. At this point I still assume its function since I need to assembly everything to get the lever in place for the tape audio PCB. Right now I have not added the handling of the REC signal too because it comes from the cassette audio PCB and can only be tested correctly when everything is assembled.
The music search is triggered by an input signal which causes an IRQ when it finds a silence of 4 seconds. The IRQ signal goes low in that case. By adding an IRQ routine in the Arduino I was finally able to develop the programming of 20 tracks for the tape deck. However, because I do not have the user manual, but only two pages of the service manual describing the “user interface”, I am not sure if it is 100% accurate.
Anyway, I am on a good way to get all the functions restored…