Beogram 4000: Full Functional Restoration and a Test Spin with Chico Hamilton

This post gives an account of the restoration process of the Beogram 4000 that I recently received from Australia. My initial assessment of the unit is discussed here.

As usual I started with the restoration of the carriage. This shows it in its as-received condition:

I removed all moving parts from the arm lowering and carriage translation systems for ultrasonic cleaning:

In most Beograms of this vintage the lubricants are hardened and the parts need to be cleaned and then re-lubricated for proper functioning. This shows the carriage after removing the parts:

It is a good idea to place the carriage on a soft pad to protect the delicate wiring at its bottom. 

Next I removed the plunger from the solenoid. They are often magnetized which can be tested with a small (non-stainless) steel set screw:

If the plunger is able to attract the screw then it needs to be de-magnetized. In this case it did attract the screw and so I put my tape head de-magnetizer to it. After the treatment the screw was utterly unimpressed by the plunger as it should be:

While the carriage is up it is the perfect moment for restoring the solenoid activated switches. They are mounted next to the solenoid so its connecting linkage can actuate them as the plunger moves:

Like in most Beogram 4000 these switches were oxidizes and also had been bent out of shape in previous attempts to fix operational issues. I turned the carriage around to access the solder points of the small boards that carry the switch terminals:

I removed these PCBs for better access to the terminals:

There are more such switches on the green carriage position PCB that detects the whereabouts of the carriage so the control system can make decisions where to drop the arm or when to return the carriage home:

This board can conveniently be pulled up when the carriage is removed. Here we see the terminals and also the original red solenoid current limiting resistor:

I un-soldered all the switch terminals from these boards

This shows their sorry state with lots of oxidation after about 50 years of service:

I usually also ultrasonically clean the plastic switch actuators of the carriage switches:

They are often caked up with lubricants incorrectly applied by inexperienced operators during attempts to fix issues that are rather caused by oxidized switches.

I cleaned all the terminals and then coated them first with an adhesion enhancing layer of nickel followed by a 24 karat gold/cobalt alloy, which will prevent oxidation in the future:

Then I reinstalled the terminals of the carriage position board:

I also put a new 3W solenoid resistor (blue) on the board. The original ones often fail since they are a bit under-dimensioned at 1W power handling capability. Then I soldered the solenoid switches back together:

and reinstalled them on the main PCB next to the solenoid:

Beogolden!

While the carriage is up it is also a good moment to address the usually brittle cartridge mounting tab. In this case the lower part that holds the ground tab in place had already lost a piece of plastic:

I removed the assembly after simmering the tonearm for a while in lightly boiling water to soften the glue:

This Beogram 4000 has the old style mount assembly, which connects the arm wiring via a spring mounted connector. This shows the Beolover replacement for the upper part of the assembly that carries the cartridge tab together with the original part:

And here the original fragmented lower part together with the new Beolover replacement part:

I transferred the original wiring to the new parts:

This shows the new parts together with the original spring loaded contact assembly on the left side:

Then it was time to install the parts in the arm tube. I use a defunct MMC20 CL cartridge (from a sad ebay shipping experience) into which I inserted the new mounting tab to press the parts into the tube:

This approach ensures that the mounting tab is located properly to ensure a flush fit of the cartridges while making proper contact.

Next I removed the (red) first-generation Beolover tracking sensor LED assembly shown in this picture:

Removal revealed the tracking sensor aperture:

Then I installed the current updated Beolover Tracking Sensor LED Light Source for Beogram 4000:

It has increased light output compared to the original version. This can help with some Beogram 4000 that have too much friction in the carriage translation system and/or carriage motors.

 

Next I re-installed the ultrasonically cleaned mechanical components:

I always install a new damper gasket for Beogram 4000, 4002, and 4004:

The original gaskets are usually hardened and/or deformed, which can lead to unhappy arm lowering experiences without proper damping. It can be a hair raising experience when the newly restored $800 cartridge bounces down onto the record occasionally!...;-). A new gasket ensures a consistent damper adjustment.

This Beogram 4000 was already equipped with a later style solenoid lever fitted with a plastic end piece for actuating the solenoid current limiter. As usual it had developed a crack:

I replaced the cracked part with a Beolover Solenoid Arm Extension for Beogram 4000 before re-installing the parts. This shows everything back in place:

Next I focused on rectifying the motor and reservoir capacitor mess that had been created in this unit:

I removed the capacitors and also the platter motor:

These motors need their Oilite bearings re-infused with oil. After many years of service the oil in these porous bearings is usually depleted and needs to replenished. This shows the extracted motor

I took it apart

and immersed the clamshell housing halves with their integrated bearings in synthetic oil and pulled a vacuum. Immediately bubbling started coming from the bearings:

After a few minutes the entire surface of the oil had foamed up:

This process usually takes 2-3 days until the bubbling stops. Then the bearings are refilled with oil again.

In the meantime I focused on the remaining restoration tasks. This shows the main PCB flipped out in its original condition:

I exchanged all the electrolytic capacitors and power transistors. These components are often out of spec. I also exchanged the original Siemens RPM relay with a drop-in ready Beolover replacement relay and the RPM adjustment trimmers with modern 25-turn potentiometers. This shows the restored board together with the extracted components:

While the main PCB is up it is the perfect moment for replacing the complementary TIP31/32 push-pull transistor pair that drives the AC platter motor. This shows the original packages in place:

I replaced them with a stronger TIP41/42 pair for enhanced longevity:

Next in line for restoration was the keypad cluster. I extracted it and put it into service position. This revealed the incandescent bulbs on the flipped up PCB that back illuminate the arm position indicator and the user-accessible RPM trimmers. The lower PCB that is bolted under the keypad carries the eight switches that are actuated by the keypad:

I extracted the switch terminals for gold plating. As usual they were also oxidized:

This shows them after my Ni/Au coating process:

Nice and shiny! I also removed the plastic actuators for cleaning:

Then I replaced the light bulbs with a set of Beolover Scale Illumination LEDs for Beogram 4000 and a set of RPM Trimmer Illumination LEDs for Beogram 4000. This shows the LED replacements in action:

At this point it was time to clean out the enclosure

and also remove the residue from previously leaked out reservoir capacitors. This shows the cleaned motor and capacitor area:

In the meantime the motor bearings had completed their oil infusion process and I broke the vacuum, extracted the motor housing from the oil, and re-assembled the motor:

Before I installed the motor again, I implanted a Beolover Efficient 24V Power Supply and Main Capacitors for Beogram 4000 board. This board replaces the entire capacitor setup with an easily installed PCB that has the solder pads in the approximate spots where the original capacitors where connected. Therefore, the board is a direct plugin that can be used with the original wiring setup. The board also replaces the linear voltage regulator setup of the 24V rail with a modern buck converter that wastes much less energy. The result is a cooler running Beogram and less power consumption.

I usually start with soldering the yellow and blue wires of the 6V system capacitors followed by their two black ground connections in the upper half of the board:

Next I solder the wires of the 24V rail and the thin blue wire that directly feeds the unregulated voltage from the rectifier to the solenoid to give it maximum power:

Then I bolted the board into place and installed the motor. The first step is to solder the wires of the motor phases to the remaining pads on the board:

Then the motor can be bolted in:

Next came replacing the carriage motor. The original motor ran fairly noisily which is a sign for dry bearings. Dry bearings cause higher current consumption which can fry the H-bridge transistors that control this motor and it also introduces vibrations into the system that can be audible when playing a record. This shows the extracted original motor:

I opened up the housing:

Then I implanted a new Beolover Carriage Motor for Beogram 4000, 4002, and 4004, which fits into he original plastic housing:

Since the motor is a bit shorter than the original ones the leads can be fed through the opening in the bottom of the enclosure:

I added the top part of the enclosure

and then installed it.

Before I could install the main board again I needed to adjust the bias of the sensor transistor to yield the service manual specified 1.8 DC at the collector. This is an important adjustment to get a proper sensor response if there is no record on the platter to prevent arm lowering: 

After adjusting the bias trimmer I mounted it on the component side:

Then I began to do the platter height and floating chassis adjustments. These can already be pretty tedious before a satisfactory result is achieved, but in this case a few more obstacles had to be overcome before the adjustments could be attempted. For a successful platter setup the aluminum panels need to be seated properly. In this case I soon discovered that one of the locking features that allow the plinth to hold down the plates when it is pushed back into locked position had come loose earlier and it was put back into place at an angle. Luckily a tug with some pliers liberated it so I could try again:

I removed the old glue from the recess as good as possible and then epoxied the nipple back into its place:

Then I realized that one of the alignment pins that position the top plates properly relative to the keypad cluster was only loosely inserted into its mounting hole:

I epoxied this one into its hole, too and spend a while to bend it into its correct position to get the aluminum plate seated in its proper position. Then I adjusted the platter height and the floating chassis to get the platter flush with the surrounding aluminum plates as well as the same platter arm height of 23mm across the entire travel of the carriage and between back and front of the arms.

Then it was finally time to do the remaining adjustments: First came the arm lowering limit. It is supposed to prevent the needle from hitting the platter ribs should the record detection mechanism ever fail:

The next step was calibrating the tracking weight. This is done by adjusting the counter weight position with a screw. Originally it is held in place with a flimsy retaining washer and a bit of paint:

I usually replace the washer with a nut so I can lock the calibration in place to make it more resilient against the rigors of shipping:

Then I adjusted the counter weight to get the proper 1.2g tracking weight when the small dial in the back is set to 1.2g. This scale on this dial is notoriously inaccurate across its range, so it is best to simply calibrate the weight with a digital scale and ignore the markings on the wheel:

Then I adjusted the tracking feedback:

And then it was time for a RPM stability measurement with the BeoloverRPM device. In its 'slow' mode it is set up to record a RPM measurement every 10 sec and sending the data to the serial port of a computer:

Using a terminal program these data can be recorded and then plotted. This is the curve that I measured after about 24 hrs:

The second operational mode of the BeoloverRPM, 'fast', records a measurement every time a black rib passes under the sensor of the device. This gives a high resolution measurement, which allows the determination of 'wow and flutter' (RPM fluctuations on a short time scale). The most noticeable feature of the curves measured in this mode are a sequence of patterns that repeat exactly every 24 measurement points:

Since the platters have 24 ribs these patterns can be identified as a measurement artifact stemming from small manufacturing inconsistencies causing the rib spacing to be slightly varying around the platter. It turns out that every platter has a distinct pattern, i.e. these patterns can be regarded a fingerprint of a particular platter.

This graph shows a measurement over 60 sec (corresponding to ~30 platter turns). The superimposed sine-wavy pattern represents the actual wow and flutter RPM variations. An evaluation yields that they are about 0.1%, approximately double what is stated in the specs section of the service manuals. This difference may well be a result of the entirely different way this was measured in the early 1970s when digital tools were not available yet.

At any rate the measured RPM fluctuations are much too small to be noticeable for humans who typically start noting pitch variations around the 0.7% mark.

And then it was finally time for a first test spin! I selected a recent acquisition to my collection a wonderful record by Chico Hamilton titled "Chico the master". It came out in 1974, i.e. could well have been played on this lovely Beogram 4000 when it was released. My favorite track is Gengis. 

What a beautiful combination!

The Beogram performed perfectly during this test. I will play it for another week or two and if nothing new comes up, it will be time to send it back to Australia!