Beogram 4000: Complete Functional Restoration

This post describes the work done on a Beogram 4000 that I recently received from Australia. See this post for my initial assessment of the unit.

This photo shows the unit with the aluminum panels and platter removed:

I usually start with rebuilding the carriage:

I removed all of the moving parts of the arm lowering mechanism, and the components of the drive system that moves the carriage:

While these parts are being cleaned, it is the perfect moment for restoring the solenoid activated switches to the left of the yellow solenoid. As usual they were bent out of shape and oxidized:

The only way to rebuild these switches is to remove the terminals. Two of the small boards are soldered to another PCB underneath them. It is best to unsolder these boards while the carriage is 'up' for removal of the switch terminals

The third board can be unscrewed and removed after un-soldering the three wires that are attached to the terminals. This shows the small boards after I removed them:

While the carriage is removed, it is also easy to get to the board that detects the carriage position with five switches. This shows it from above with the plastic switch actuators visible:

This shows the board flipped up, which reveals the switch terminals and the solenoid current limiting resistor:

I removed all the terminals from the switches:

While the carriage position board is up it is the best moment for replacing the TIP41 transistor that drives the solenoid:

This transistor often fails, and therefore a prophylactic replacement is a good idea when restoring a Beogram 4000. This shows a modern higher voltage version, a TIP41C implanted:

This shows the cleaned and gold coated switch terminals:

I usually coat them first with nickel and then with a cobalt-gold alloy. This type of coating is best for switches due to its hardness and resulting high mechanical stability. This shows the carriage position switches but back together and a new solenoid resistor installed:

Beo-golden-lovely!..;-). I also cleaned the plastic switch actuators with an ultrasonic cleaner before putting them back onto the board:

This shows the straightened out and coated solenoid switches put back together:

And reinstalled next to the solenoid:

While the carriage is removed it is also a good idea to replace the cartridge mount. The Beogram 4000 was unfortunately fitted with cartridge mounts that were made from a plastic that becomes brittle over the years. It tends to break off easily. In this particular Beogram it already had cracked, but was still attached to the flexible PCB that forms the actual contacts that connect to the cartridge coils:

For replacing the mount the arm needs to be removed and 'cooked' for about 30 min:

Then the assembly can be pushed out (after removing the screw that holds it in place if it is an 'old-style' cartridge mount like in this case) with a suitable rod that fits into the aluminum profile:

This shows the new Beolover cartridge mount part next to the original assembly:

When installing the part it is a good idea to use an old cartridge to push it into the right position that the cartridge fits flush with the aluminum profile:

This shows the installed mount:

The final item to replace on the carriage was the tracking sensor light bulb, which is located in the black housing under the arms:

I removed it and fitted a Beolover LED based replacement in place.

Then I re-installed the carriage and the mechanical parts, which were cleaned ultrasonically:

I also replaced the rubber gasket in the damper plunger:

It often fails and causes inconsistencies in the arm lowering speed.

This shows the carriage back in place:

There was one item left to take care on the carriage, but which is best done after it has been installed again: The removal and cleaning of the damper to arm linkage. It can be seen poking out in the back of the arm assembly, fitted into a small "V" cut in the arm that connects it to the tone arm back:

In order to get to it, the sensor arm needs to be removed. This shows the arm removed and the linkage extracted (make sure you do not loose the small spring that is under the circlip that holds the linkage down if you try this at home):

As usual the small copper plate that helps the lateral arm movement while the arm is up was loose:

It is held in place by usually deteriorated double sided tape. I cleaned it and then glued it back in place with a dab of epoxy:

Once the sensor arm assembly is back in place, the arm 'parallelism' needs to be re-adjusted:

Now it was time to rebuild the reservoir and motor capacitor section, and the platter motor itself. This shows the original setup:

I removed everything, cleaned the compartment from some capacitor leakage, and opened up the motor for oil infusion:

I put the motor housing with the embedded shaft bearings into motor oil and pulled a vacuum. This extracts air from the bearings and makes room for oil to interdiffuse:

While the oil infusion happened, I focused on the keypad cluster. This shows it removed from the enclosure and opened up into service position:

As usual, a light tug on the strobe mirror made it come off from the keypad housing. Closer inspection yielded that the diffusor of the 'light box' for the uniform illumination of the carriage position scale had also started coming loose:

The contact terminals of the keypad switches showed the usual oxidation:

I extracted the terminals for electroplating:

This shows them after nickel and gold plating:

I also removed the switch actuators:

They have different, longer 'pistons' defined by the separation between the actual keys and the PCB underneath them. The longer parts of the pistons go into the plastic fixtures, as shown here:

I cleaned the plastic parts with my ultrasonic cleaner. This shows the terminals re-installed: 

The second PCB under the keypad contains the light bulbs for illuminating the scale and the two red adjustment trimmers for 33 and 45 RPM:

I replaced the bulbs with LED fixtures:

Finally, I glued the separated parts. This shows the position scale cover removed from the keypad assembly, allowing me to get to the diffusor:

I completely removed the diffusor. It came off very easily.

I glued it back into place with a small amount of carefully applied epoxy around the 'box':

I did the same for the mirror:

That concluded the work on the keypad assembly. I moved on to restore the main PCB. This shows it filled up to reveal the component side:

While this board is up, it is the perfect moment to replace the complementary TIP31/32 power transistor pair that forms the push-pull stage that drives the platter motor. They are bolted to the chassis for heat dissipation purposes:

I replaced them with stronger TIP41/42C types. These transistors have a history of failing, i.e. I think it is a good idea to step up their current capability. This shows the new types installed:

Then I did the main PCB. I replaced all electrolytic capacitors, all power transistors, the sensor transistor, the RPM adjustment trimmers and the RPM relay with new components:

There is a second, smaller PCB in the Beogram 4000, which takes care of the power rails. This board has also two electrolytic capacitors that need to be replaced. They are the two brown units in the lower left quadrant:

Another item on this board is the 6V rail fuse, which is often deteriorated. It resides in the yellow insulating tube next to the reed relays. This shows it after pulling back the tube:

After a light tug, the fuse immediately separated. I replaced it with a new one:

This shows the rebuilt board:

In the meantime, the oil infusion had come to an end, and I reassembled the motor and installed new capacitors with the aid of a (red) 3D printed fixture that holds them orderly in place:

There was still one incandescent bulb that needed replacement: The one in the sensor arm that provides the light for the detection of records on the platter. This shows the extracted bulb compartment together with the Beolover LED assembly, that directly replaces the bulb: 

This shows the LED assembly implanted together with the original bulb:

This shows the LED in action. It uses a warm white LED, which emits enough red photons to light up the B&O logo in style:

This concluded pretty much the work on the electric system, and it was time to do some measurement and calibrations. First I checked the motor voltage, which needs to look like a perfect sine wave. This shows the 33 RPM signal:

Perfect! On to adjusting the bias of the sensor transistor. I installed a trimmer that allows to calibrate the bias precisely to yield the prescribed 1.8V at the collector of the transistor:

After this adjustment, I measured the sensor response:

The amplitude of the sensor signal was a spectacular 3.8V. Definitely a passing grade! Every dip represents the passing of a platter rib under the sensor.

At this point I realized that the threaded insert that allows the cover of the fuses compartment to be bolted down had broken out of its plastic sheath:

I located the broken off fragment and the insert and epoxied everything back on together with the treaded insert in place:

This shows the fuses box with the re-attached insert:

and with the cover bolted back in place:

It is generally a good idea to not over-tighten the screw of the fuse box. It likely broke due to too much torque. 

The next step was replacing the cracked plinth guidance washers

with 3D printed replacements:

Then it was finally time to do all the necessary mechanical adjustment. First I aligned arms and platter to be parallel and at the proper distance. Then I adjusted the floating chassis to get the platter level with the surrounding aluminum panels. 

After this process was completed, I adjusted the arm lowering limit to make sure the needle would miss the ribs on the platter should the arm ever be lowered without a record present due to a malfunction of the record detection circuit:

The next step was to calibrate the tracking weight to be accurate around 1.2g, which is the specified weight for most B&O cartridges:

It is usually best to not rely on the weight scale on the adjustment wheel. I always use a digital scale for getting the weight right.

This restoration was coming to a close, and I finally performed a 24 hrs RPM stability test of the platter motor with the BeoloverRPM device, which allows logging the RPM in 10s intervals over extended periods of time:

This is the spectrum I  measured after ~24 hrs:

This spectrum is as good as it gets. The AC platter motors pretty much never have RPM issues due to their brushless design and synchronous operation mode. 

Before taking a first 'victory spin', I still needed to replace the corroded original DIN5 plug:

I replaced it with a quality modern all-metal plug that has gold plated contact terminals:

And now it finally was time to play a first record on this restored Beogram 4000! I selected one of my favorite Quincy Jones records, "You've Got It Bad Girl" from 1973 (A&M Records SP3041). Of course, this record was thoroughly cleaned on a CleanerVinyl ProXL System combined with a UC-3360 multi-frequency ultrasonic cleaner before listening. This restored the sound of this record to its original glory:

A perfect combination! This record probably was played on a few Beograms back when it came out! I will now play this deck for a couple weeks to make sure there are no intermittent issues. Then it will be time to send it back to its owner!