Featured Post

Beolover SyncDrive: DC Platter Motor Replacement for Beogram 4002 and 4004 (Type 551x and 552x)

Late Beogram 4002 and the 4004 (Types 551x and 552x), which have DC platter motors instead of the earlier synchronous AC motors usually suff...

Wednesday, September 9, 2015

Beogram 4000: Replacing the Tracking Sensor Incandescent Bulb with a Drop-In SMD LED Assembly and a Peek at the Opto-Resistors

The fourth light bulb that was broken in this Beogram 4000 was the tracking sensor bulb. I would have replaced it anyway. My recently developed SMD LED based replacement assembly works very well now in several Beogram 400x, and why risk a bulb failure down the road if this can be done with a LED, that most likely will survive the next 50 years? I use high efficiency white LEDs, which I run at a very low power level. This will make them last very long. Also it reduces the power consumption of these units, which is a concern due to the main transformers occasionally giving out, since they are not very over dimensioned, i.e. run close to their rated power. If all LEDs are replaced this way this can easily reduce the power intake by a few Watt, corresponding to a 10-20% current reduction in the transformer.
Anyway, here are a few impressions of this procedure. Details and instruction for the implementation of the drop-in ready assembly and the necessary adjustment of the tracking mechanism can be found in this video.
This shows the original light bulb assembly:

This shows the replacement design. The blue part matches the lower half of the original bulb housing shape to shield the opto-resistors in the sensor from ambient light. The PCB carries the LED and current limiting resistor, as well as a trimmer potentiometer to adjust the intensity of the LED. This is a great upgrade for fine tuning the tracking sensor performance to B&O spec (see video).

This shows the light source assembled from the bottom. The left 'window' contains the LED, which is located close to the original position of the bulb filament.


This shows the unit from the top. There is only one contact lead since the other contact is made to GND via the mounting screw. This simplifies the implementation of the assembly:


Here it is installed:
A test revealed that the LED performed as designed. 

Before I got it to work, however, this particular Beogram 4000 surprised me after the installation with a tracking mechanism that would only work in reverse direction (check the video for a demo - the 4000 can actually track a record in reverse direction, which makes it different from the later 4002 and 4004 models). I removed the bottom part of the tracking sensor unit, which contains the two sensors that control the carriage motor movement to check the two opto-resistors. My fear was that one of them had died. Here are a few pictures of how this assembly looks like.
This shows the top. You see the two slots under which each one sensor is placed. Depending on which one is brighter the carriage moves left or right. 

This shows the bottom:

Lovely old semiconductor devices. This is one of the reasons why I like this design so much. It is from a point in time where circuit design started taking quantum leaps due to the advent of integrated and other more advanced semiconductor devices.
Anyway, a test of the sensors showed that they worked. Here is what I measured: Several MOhms in darkness, and about 1 kOhm when blasting them with a quite bright white LED desk lamp. Both showed identical values indicating that I was searching at the wrong location for the root cause of the tracking issue. 
Of course, what I found was a cracked trace on PCB #6 that connects the sensor to the rest of the circuit. One of the pads that connect the 24V rail to the upper end of the sensors broke from the connecting trace. I fixed this by soldering a short extension to the sensor lead in question. You can see it in the above picture if you look closely.

Now the deck was ready for the tracking sensor adjustment procedure. I did pretty much exactly what I show in the video. First, I adjusted the sensor position mechanically using the excenter and then after being close to the prescribed two-record rotations, I used the trim pot to get a precise two rounds. I think I will implement this trim pot also in future Beogram 4002/4004 restorations. It just takes all the pain out of the adjustment process and makes it a quick affair.

Here you see it in action playing Somethin' Else by Cannonball Adderley on Blue Note:








No comments:

Post a Comment

Comments and suggestions are welcome!