Beogram 4000: Complete Functional Restoration

A while ago a customer from Texas bought a Beogram 4000 on ebay from a seller in New York. The unit was in a  decent shape. See here for my initial assessment after I received it. Due to the cracked hood corner we decided to buy a second 4000 and create one perfect unit from the two. This second unit arrived recently, and so I started the restoration process:

I selected the unit with the better overall cosmetic condition, and began with rebuilding the carriage transport and arm lowering systems. One of the linear bearings of the carriage had come loose during shipping. This is a frequent thing to happen with the 4000. The linear bearing is the white 'triangle' in this picture:

While the carriage was up, I also removed and cleaned the threaded rod and the polished shafts on which the carriage slides. I also removed the other linear bearing and the threaded 'nut' that connects the carriage on the threaded rod, and cleaned them in an ultrasonic cleaner:

These bearings seem to just be pressed into the carriage bottom. I usually put a dab of epoxy on them before I press them back in, hoping this will prevent another dislocation during return shipping. This shows the linear bearings back in their slots:

I reinstalled the carriage and then it was time to rebuild the arm lowering mechanism:

This shows the individual parts removed from the assembly:

After reinstalling everything, I turned the Beogram around to take the sensor arm out for a re-lubrication of the damper-to-arm linkage. This shows the arms from behind:

I removed the two screws that hold the sensor arm in place and took it out:

and removed the linkage:

Be careful not to loose the small spring that is under the circlip that holds the linkage in place, if you try this at home.

I cleaned the pivot point and put some new synthetic grease on it and installed the arm back onto the carriage, and then adjusted the arm parallelism. The next step was the installation of a LED based replacement for the light bulb of the tracking sensor. The black box in this picture is the original bulb housing:

This shows the new LED based part together with the original bulb:

I installed the new part:

The last item on the carriage was the replacement of the cracked plastic pulley that connects to the carriage motor:

Pretty! If you like to get one of these precision machined replica pulleys, send me a message. I will be happy to get you in touch with the B&O enthusiast in Vienna, who had them manufactured. Like all the other parts shown on this blog, this part is available to other enthusiasts.

The next step was rebuilding the electronics. I started out with the main PCB and replaced the electrolytic capacitors, the RPM relay and the RPM trimmers. This shows the rebuilt board:

And a detail view of the RPM trimmers:

I install the new 25 turn precision pots in a way that they are easily accessible from the solder side of the board:

There are two more electrolytic capacitors on the power supply board, but I forgot to take a picture. On to the main reservoir and motor capacitors assembly. This shows the original setup:

These capacitors are out of spec on most 4000s these days and it is crucial for any restoration that they are replaced. I removed them, and it was obvious that some had leaked already...messy!

While I had the capacitors out, I also removed the AC motor

and opened it up:

A beautifully simple synchronous 'can motor' design, which will probably run forever since there are no touching current carrying parts that can burn up over time, no brushes or commutators. Nice! and made in Switzerland! Those were the days! 

I immersed the enclosure with the embedded bearings in oil and pulled a vacuum and let them sit over night. This treatment usually quenches any knocking noises that these motors sometime make when they run (even when driven with a perfectly adjusted sine wave signal).

I put the motor back together and installed it (using 3D printed parts that hold M3 nuts in place to replace the threaded rivets that need to be drilled out to open the motors up).

I installed an assembly of new capacitors using a 3D printed fixture to hold the smaller modern capacitors in place:

Beolovely! (see here for a more detailed account on how to do all this).

The next 'section' of this restoration was the keypad cluster. It contains the control system of the Beogram 4000. It is based on early TTL chips, and can be regarded a hard wired digital logic system. Pretty advanced for the day, and B&O made a couple steps backward in this department when they downgraded the control system to an analog system for the 4002 and 4004 models. Probably to save a few Kroners while keeping up appearances...;-). This shows the keypad removed and in 'service position':

The logic chips are on the vertical PCB, while the keypad switches are on the PCB that is bolted behind the keypad surface. The first item to tackle was to clean and gold plate the switch terminals. These mechanical switches are probably the most crucial task in any Beogram 4000 restoration, since they are usually well oxidized, and if one does not work properly, the performance of the entire deck is compromised. This shows the nicely blackened switch terminals in more detail:

I removed the lower PCB and then extracted the switches:

It is pretty obvious that they probably were not that reliable anymore! I cleaned the oxide off with a 1500 grit sand paper and cleaned them in an ultrasonic cleaner. First with water+detergent and then in isopropyl alcohol. Metal plating requires very clean surfaces to work properly. The next step was to plate the terminals with nickel to generate a surface that enhances the bond with gold. I usually use a wand based plating process:

The greenish sheen of the cloth on the wand comes from the Ni salt that is the electrolyte in this process. After 2-3 minutes a solid Ni layer can be achieved. I did all the terminals with Ni and then followed up with a few minutes of gold on all the terminals. This shows the final result:

Beogolden!! Then I re-installed the terminals on the PCB:

The final tasks on the keypad was the replacement of the four light bulbs with LED fixtures:

Before putting the keypad back together I needed to glue the strobe mirror back onto place.

I have not seen a 4000 in a while that did not need this procedure:

After this I put the keypad back together. Before installing it back into the enclosure, it was time to replace the cracked plinth guidance washers since one of them is located under the keypad. This shows the new 3D printed washers:

The front center washer

should be replaced with a black one, since whit ones can be seen through the crack between aluminum plates and plinth:

The next step was the rebuilding of the carriage switches, that tell the control system where the carriage is:

And after installing them back onto the PCB that sits under the carriage:

Not done, yet with the gold coating process! Some more switches are next to the solenoid of the arm lowering system. someone already had 'messed' with them as was evident from their bent nature and the missing bolts that hold the PCBs of these switches down:

This is probably the most 'messy' switch restoration process in the 4000 since the switches are mounted on vertical PCBs that are soldered into the horizontal ones, while everything is connected with very flimsy wires that like to break off. So do not drink coffee and play some quiet music if you try this at home. I removed the vertical boards by flipping the boards up:

and unsoldering their soldered connections to the main boards.

Then I was able to remove the terminals. This shows the tracking enable switch and the output grounding switches (that prevent the 'thunk' in your speakers when the needle sets down on the record:

This shows the switch that reduces the current through the solenoid after the arm is down:

I plated the terminals and soldered them back on the vertical boards:

This shows everything back in place:

Ah...the satisfying sheen of gold!...;-) I hope these switches will never oxidize again!

After this I replaced the last incandescent bulb, the one in the sensor arm. One can access it by pulling out the sensor compartment from the aluminum profile:

This one is a bit tricky since there is so little space in the bulb compartment. A while ago I developed a replacement circuit on a flex PCB that folds into the space. One cannot directly replace the bulb with a LED and a resistor, since that would fool the bulb failure detection circuit into thinking the bulb is dead (it measures the current, and LEDs have a much lower current). Therefore, a bypass resistor needed to be included in the circuit to make this a drop-in replacement.

This shows the replacement in action:

Time to upgrade the signal path. I replaced the crappy plastic DIN5 that someone had installed at some point

with a quality all metal plug with gold plated contact pins:

The next step was to install a switch that allows connecting system and signal grounds in case there is a hum issue:

This is done conveniently at the terminals that connects the thin wires from the tonearm to the output cable. This switch is often helpful to stop humming when running the Beogram through an RCA input on a non-B&O amplifier. 

Now it was time to do the sub-chassis, aluminum plate alignment and the final adjustments. First, I properly aligned the aluminum panels relative to the keypad by bending the alignment pins until a good fit was achieved. Then I adjusted the sub-chassis and the platter to be horizontal and freely moving relative to the enclosure, and made sure the arms are horizontal and at the proper distance from the platter. Then I adjusted the arm lowering limit to about 1 mm above the lower rungs of the ribs on the platter:

This is a safeguard against ruining the cartridge should the record detection system ever malfunction and allow lowering the arm on the naked platter.

Then I calibrated the tracking weight adjustment. The first step is to get the counter weight locked in the right position. I usually replace the flimsy circlip on the adjustment screw with a M3 nut and a washer. This allows fixing the weight in place to survive the rigors of shipping:

Then I calibrated the scale of the spring based weight adjustment wheel with a digital tracking weight gauge:

Then I made the electrical adjustments, namely the slow carriage speeds, the RPM and the motor voltage. This shows the 33 RPM motor driver signals after adjustment:

The 45 signal looked similar but had a 10% smaller amplitude. This is normal, since the circuit only allows adjusting both at the same time, and one usually gives preference to the 33 RPM signal since it is more often used in most cases. Then I adjusted the end groove detection circuit:

This shows the carriage motor voltage right before the end groove started. The 'waves' on the left are the periodic carriage movement during the last few normal turns of the records before the arm entered the end groove. At that point the voltage shot up and that triggered the system to lift the arm and drive the carriage home (hence the large opposite voltage). The threshold for this to happen needs to be adjusted with 1VR4 to be above the 'waves' that occur during normal playback (turn VR4 a bit CCW if the arm lifts before the end groove is reached - this can be done by trial and error in a pinch).

The final adjustment was the tracking feedback, which was greatly facilitated by the intensity adjust potentiometer on my LED tracking sensor light source.

And now, finally, it was time for a test drive! I selected one of my favorite records, Milt Jackson's "Sunflower" (CTI 6024) where he plays together with Herbie Hancock, Freddie Hubbard, Ron Carter and Billy Cobham. How much better a line-up is possible? My favorite track is "People Make the World Go Round" on side 2:

Of course, this record had been ultrasonically cleaned on a CleanerVinyl ProXL System before play! The joy of vinyl! What a beauty the Beogram 4000 is! Only one step left: Restoring the hood. A nice workout in the garage is coming up!...;-)