Beogram 4002 (5513): Further Improvement of the RPM Stability Through Replacement of the RPM Trimmers and the RPM Relay

After rebuilding the DC motor, which resulted in a dramatic improvement of the RPM stability of the Beogram 4002 (5513) that I am rebuilding right now, I replaced the main RPM trimmers and the RPM relay hoping this would further stabilize the RPM behavior and return this Beogram back to the 0.05% spec state din the service manual. While I had the main PCB up, I also replaced all the electrolytic capacitors with quality Japanese main manufacturer 105C grade types. This shows the board in its original state:

And this shows it after replacing the capacitors, RPM relay and trimmers:

This shows the RPM control setup in detail. The relay was replaced with a modern encapsulated unit that fits into the original solder points via a custom designed breakout board:

The relay is available to other B&O enthusiasts - just send me an email. The 25 turn trimmers are installed in a way that their adjustment screws protrude through the openings in the PCB to enable adjustment from the solder side with running platter:

In difference to the original setup I oriented the trimmers in a way that the RPM can be increased through a clockwise turn making the adjustment more intuitive.

After putting the PCB back in I also replaced the main reservoir capacitor. This Beogram has the separate power supply for a CD4 board. I like to keep this original, and so I replaced the original double capacitor with two modern units held in place with a custom designed 3D printed adapter. This shows the original setup:

The single contact in the back towards the platter bearing is the common ground of the 1000u and 4000u capacitors that are combined in the single can. On the front the can has the two positive contacts of these capacitors:

This shows the replacement assembly:

It is held in place by the clamp that held the original capacitor:

And then it was time for the next RPM stability test with my BeoloverRPM device:

After another day of running the Beogram I was able to add another curve to my graph:

The top curve shows the performance improvement after replacing the trimmers and the relay. For comparison the previous two spectra are also shown. It is obvious that the replacement of the relay and the trimmers resulted in a further significant improvement of the stability. There is virtually no more long term drift and the fluctuations are now in the 0.05% 'corridor', except that there are a few small (~0.1%) singular spikes that still exceed the specifications. I have a theory what could cause this...let's see what the next experiment will yield! How exciting!...;-)