This section details the work done on the Beogram 4000’s oscillating solenoid, and what was learned about winding solenoid coils and comparing them to the original.

The arm lowering mechanism is shown here:

When the control logic initiates arm lowering, the solenoid pushes the solenoid lever forward. The lever opens the solenoid switch when fully extended, decreasing the current flowing through the solenoid to a level just enough to keep it engaged.

This electronic operation is depicted in the circuit diagram snippet below:

Lowering Q1’ from 6V to 0V activates the solenoid. This action turns off RT10 through 1R25 and 1D3, which in turn activates 1TR11. This pulls up 0TR4’s base, and maximum current flows through the solenoid because its lower end is directly connected to GND via the solenoid switch. The solenoid switch opens once the solenoid is fully activated, forcing current to flow solely through the solenoid power resistor 7R1. This significantly reduces the current passing through the solenoid.

Q1’ rises to 6V when it’s time to lift the arm, and the current through the solenoid stops, allowing it to return to its starting position due to the solenoid return spring. This also closes the solenoid switch, preparing the solenoid for the next time Q1’ goes low.

Therefore, a solenoid oscillation usually indicates a problem with the GND connection via the solenoid switch: Q1’ activates the solenoid, but insufficient current causes the return spring to drive it back. Because the spring is slack again, the solenoid extends again, and so on, causing the oscillation.

Based on previous experience, all connections and wires were checked, but the wiring and solenoid switch appeared to be fine. The intermittent nature of the oscillation suggested this was not a typical case. A broken wire usually produces this behavior reliably.

As a result, a more in-depth investigation was required. An oscilloscope was connected to 0TR4’s collector (yellow trace) and base (blue), and the ’normal behavior’ was measured when the solenoid was not oscillating.

As you can see, the base voltage struggles to rise during the initial current burst and only reaches full magnitude after the current settles.

At this point, the ‘brrrr’ behavior could be induced approximately every 50 arm lowering activations. After repeated pressing of the up and down buttons, the ‘brrrr’ sound was finally heard and measured:

After an initially normal activation, the solenoid begins to oscillate. Because 0TR4’s base appeared to follow a similar pattern as during the initial lowering, it was concluded that Q1’ itself, the control logic, may have been the source of the oscillation. The fact that the initial lowering was executed flawlessly and that the next lowering sequence was only carried out after maybe 20ms supported this theory.

As a result, the Q1’ signal (blue) and the 0TR4 collector (yellow) were measured. These are the traces for a typical arm lowering sequence:

When Q1’ goes low, the solenoid is activated, and the solenoid switch opens, lowering the voltage at the collector.

The following traces were measured during a ‘brrr’ moment after more ups and downs with the buttons:

This confirmed that the issue was not with the control system because Q1’ did not change during the ‘brrr’ event. This was a positive discovery because a faulty logic chip would have added a new layer of complexity to the project. This result at the very least indicated that a component or connection after Q1’ must have triggered the issue. TR10/11/0TR4, as well as the spark snubber diode connected across the solenoid leads, were replaced:

This, however, had no effect. After a series of ups and downs, the (now dreaded) ‘brrrr’ returned.

After reviewing all of the oscilloscope measurements taken at various points in the solenoid circuit, it became clear that 0TR4’s collector going to zero during the ‘brrr’ moments meant that the connection between 0TR4 collector and the top of the solenoid, which is hardwired to ~40V, must be open circuit! Because Q1’ remained constant during the oscillation events, 0TR4 remained on, and the collector was no longer pulled up.

In the absence of bad solder joints, the solenoid itself was the most likely source of the problem. The original solenoid was replaced with one borrowed from a Beogram 4004:

The oscillation could no longer be replicated with this new solenoid in place, despite numerous attempts. With this replacement solenoid, however, an intriguing curve was measured:

The yellow trace represents 0TR4’s collector, and we can observe a brief dip in collector voltage about 5 ms after the arm makes contact with the solenoid switch. This was interpreted as switch bounce caused by mechanical oscillation of the switch terminal. This is supported by the fact that the phenomenon did not always occur, and when it did, the timing varied slightly. Bouncing is a normal occurrence in mechanical switches. It could have been caused by the replacement solenoid’s slightly different actuation speed.

Because the replacement solenoid belonged to a perfectly functional Beogram that needed to be reinstalled soon, it was decided to build a replacement coil for the original solenoid. The first step was to disassemble it. It is depicted here after being removed from the arm lowering mechanism:

The first step is to remove the plunger. The bent extension is screwed into the magnetic field-driven plunger.

It can be easily unscrewed by holding the plunger with pliers (protect the plunger from being damaged by using tape for the jaws) and turning the extension:

The tapped plunger end and the separated extension are shown here:

Following the removal of the locking ring, the next step was to press out the metal tube from within the solenoid coil core. This was accomplished with an arbor press and a small screwdriver bit:

The coil can be removed once the tube is removed:

A replica of the plastic spool was designed and 3D printed:

Then, a stepper motor with two 3D-printed centers to hold the spool was set up,

and a basic program was written to control the stepper motor’s speed and count the turns it accumulates during spool winding. Then began the process of winding a new coil with 28 gauge ‘magnet wire,’ which appears to be very similar to the original wire. After a few attempts, sufficient skill was developed to wind the windings in a ‘homogeneous enough’ manner. By guiding the wire between fingers while the motor rotates the spool, the first three layers can be made nearly perfectly:

[![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEheLsEbBogzgicuMwAbPL9WT-nXy7jKFmVtaRxkIYViMCksWQ-6nIC4hsmieY9KbF-MeMv9pkTCkRNV0XB_dzZjOjtcUwrncy9ihHUlT6acKHX-uN_YA1TISc1rNneAHLr2ssla9f8HrzWAGPOfUUrWvMvgl1_UOx6Llwn1p7