After getting the sphere to Karlsruhe and doing some more extensive test runs, we discovered how hard it actually is to tune the sphere. Not only are the trimmers weak and over-sensitive (because they were quite cheap), but above all the oscillator frequency was very much depending on the battery voltage. Some measuring procedures proved that the 555 chip is stable between 6V and 12V, unstable between 12V and 13V, and stable again up to 20V. So the problem was, that when the battery got empty, the voltage dropped down exactly in this area where the 555 chip was sensitive and changed its oscillating frequency.

The only way to solve this problem was to add some DC voltage stabilizers (one per hemisphere), which would limit the battery power to stable 9V, no matter how empty the battery would be drawn. We decided to apply this 9V power only to the 555 chip, not to the output stage transistors. This way, we would have the full unstabilized battery power for getting the maximum volume, but some stabilized 9V for the chip to make it oscillate more precisely, and independent from any battery deviation.

All we had to do was cutting one line on each circuit board and running an additional +9V line all around the sphere, from board to board. Plus, we had to etch, prepare and install two more simple circuit boards, each of them containing a voltage stabilizer. Together with Leho, this was done in half a day and half a night. The results were positive: when the battery gets empty, the oscillation frequencies will remain unchanged - only the volume will drop down a bit. That’s how it should be!