By popular request (really, I got quite a few emails about this!...;-), I finally completed my Beogram DC motor restoration video! It demon...
Saturday, May 27, 2017
Beogram 4004 (5526): Replacement of the Sensor Arm Bulb with an LED Assembly
The sensor arm light bulb is the most difficult to replace with an LED in the Beogram 4000 series. I recently developed a flex PCB based assembly based on a warm-white LED, which fits into the small bulb compartment of the sensor arm. This picture shows the original light bulb that was installed in the Beogram 4004 (5526) that I am restoring right now:
This shows the light bulb in comparison with the flex PCB based replacement:
The LED assembly folds precisely into the bulb compartment and is held in the right position by the 3D printed red wedge that is glued on top:
This shows the assembly in action:
The warm-white LED ensures that the B&O logo is lit up properly in red...amber LEDs do not contain any red photons, and so the logo would look grayish-amberish.
Once the LED is installed, the detection circuit needs to be checked and adjusted for maximum sensitivity. I wrote a detailed explanation in the above link, which explains how the circuit works. One of the issues of the circuit is the amplification of the sensor signal, which itself depends on the position and shape of the light source in the bulb compartment, as well as the biasing of TR3, which takes care of the initial amplification of the fairly weak sensor signal. The biasing depends on the current gain of the particular transistor installed as TR3. Current gains notoriously vary throughout a transistor production batch, which suggests that feedback-based amplification schemes are much more stable across a circuit production run. The circuit used here, however, is not feedback based and so the amplification can vary. The key is to adjust the working point of TR3 that 4V are at the collector when there is no sensor signal (record present). This involves adjusting R26 until 4V are measured. In the above post I simply implanted a 2M resistor and hoped for the best (it worked) but in the spirit of Kaizen, this time I used a 2M trimmer that I could adjust to achieve a precise 4V collector setting. To do the adjustment it is a good idea to first install the trimmer on the solder side of the board that it can be accessed while the board is installed and the turntable running:
After adjusting the trimmer to get 4V at the collector
I flipped the board up and installed the trimmer on the component side:
It fits snugly into the place of the original 1M resistor.
After this installation I measured the sensor signal at the collector while the platter was spinning (if you do this yourself, you would need to install a jumper cable on the board feeding out the signal that the platter can spin while doing this measurement):
The achieved 6.62V amplitude is more than sufficient to inform the record detection circuitry about the absence of a record, i.e. this installation was successful.