SABC/SAUK "Dok Viljoen" Mark 3 broadcast turntable (from ±1974)

Officially badged as the "Mark 3", these turnables were the third and final version of a series of very high quality units used by the South African Broadcast Corporation, from the 1950s through to the 1980s. Casually they were called "Dok Viljoens", after a certain Dokter Viljoen, of whom I can find no inforomation. In the 1980s, they were gradually replaced by Technics SP10s, but some Mark 3s remained in remote studios until digital formats made them fully redundant.
The record shown above is a 17cm single, 7-inch. The black rubber mat is 30cm, 12-inches, giving an idea of the size of the 333mm platter. It can play a 16-inch transcription disc.

Radio-5 (Radio-5 since Oct 13, 1975, replacing LM Radio), around 1984.

Springbok Radio: May 1st 1950 - December 31st, 1985. This picture taken on the last day before Springbok was canned.

The SABC's administration "Piet Meyer" Building on the right, built in 1975 to cope with the expansion to television. Behind to the left, the Albert Hertzog Tower, built in 1962 specifically to provide antennae for FM radio transmission. The Hertzog tower is commonly called the Brixton tower because of the suburb in which it stands. In 1962, television was definitely not anticipated. Even today, the Hertzog tower remains an amazing oddity of RF Engineering for the large number of transmitters it houses.

The Dok Viljoen turntables were made by the SABC workshops, in collaboration with the CSIR (Council for Scientific and Industrial Research), and are exceptionally heavy (25kg) and durable - look at the size of the platter and bearing below. (The 21st century SABC is no more the capable, vibrant organization it once was, but now a mismanaged, bankrupt joke, with no such manufacturing facilities.)
The drive system is by three idlers, only one being put into service at a time by one of three solenoids. The main bearing incorporates a fourth solenoid, which can lower or raise the platter, connecting it or disengaging it from the rigid and oversized aluminium/rubber mat. When the platter is lowered, the blue mat rests in the raised bezel surrounding the platter, and is stationary. Raising the rotating platter causes it to connect with and lift up the mat, thus providing "instant start". The height change of the record is only around 1mm, so the bump doesn't dislodge the stylus.

The rights to manufacture this very turntable were given (orr sold) to the Australian Broadcasting Corporation, but I'm not aware that they ever put it into production.

The studios had multiple turntables because at that time all jingles and adverts were recorded on 10-inch records. An ad break or news break with jingles would have required several such records to have been previously cued up for instant playback - possibly by remote control if there was only a presenter and no assistant.

Mine was missing the motor, mat, cue control panel, electronic control box and power supply, and arm tube - but the azimuthal arm bearing was intact. Essentially all I really had was the plinth, chassis, platter, and idler mechanisms. But such a massive platter had great promise, it had to be put back into working order.
Since it was given to me, being thrown out in fact, I wanted to restore it without spending any big money. I like the idea of having one of the best turntables ever made, operating in perfect condition, and having got to that through buying only a few chap parts, and using stuff I had lying around.

The last two pictures are the speed control panel. I was missing the right-hand control panel that performs cue operations.

I was all ready to make a clone of the missing cue control panel, when in August 2010 "JD" acquired a Dok Viljoen of his own and some spare parts, from the Port Elizabeth SABC studio, and donated one of the control panels to me. Also a burned out Papst motor; this motor I wanted to re-wind, but the cost of getting that done was very high (over R1200). The original motor was a Bodine synchronous type intended for Ampex tape recorders. Later Dok Viljoen's used Papst hysteresis synchronous motors.
The button functions: 1: Cue (red). Lifts the platter without driving it - the idlers remain disengaged. This allows the track start to be cued up by hand, listening over headphones. 2: Ready (green). Engages the idlers, and lowers the platter while it rotates. The rubber mat rests in the bezel and cannot be rotated even by hand. 3: Start (orange). Lifts the platter and instantly start the record rotating, but audio is kept muted. 4: Broadcast (red). Switches the audio over to "on-air".
The black knob controls the audio gain, by means of some discrete FET VCA circuitry. Oddly, this changes the audio level of the broadcast out signal, and not that of the headphone monitor. I would have thought that broadcast level would be controlled only at some remote mixing desk.

Even more great things were to come: I acquired a complete control box from Shaun in CT (same one who sent me the SP10 platter) and audio electronics for a Mark 3. Below are photos of it as I received it - no restorative work done yet.

Click on an image to open it large in a new browser tab:


The Grässlin usage counter shows 32537.5 hours!

Front & top, cards pulled. Top panel flipped up. The IEC mains socket is non-original, the original mains input being the 3-pin Cannon adjacent to the fuse.	Top showing some I/O sockets. Cartridge input is via the 5-pin DIN.	Inside, showing output balancing transformer, and an additional small yellow transformer at the back - for the monitor headphones.
Back view, showing balancing transformer octal sockets, and a pair of XLR output sockets facing downwards.	More of the back: a resistor array - to provide 600R line loading, and on the far right the phase-shift capacitor for the platter motor, in front of the substantial mains transformer, and to the left, its start/stop relay. The add-on panel above with a single bolt is a mains suppressor block - also non-original.	The headphones sockets. "Transmissie"??? What kind of a hideous Anglicism is that? Wat het van Uitsaai geword?
Business cards! Left, voltage regulation, right audio amp, RIAA pre amp using a uA739 op amp, and transistor headphone & line amps.	Top left is the audio muting and DC-controlled volume circuit, below is the logic control card, and right the power transistor switching for the motor, lights and platter control solenoids.

In order to do necessary repairs and modifications, as well as figure out exactly how it all worked, I traced out all the circuitry. I haven't CADded them all yet.

Power suppy & solenoid driver stages.

Logic board.

RIAA - straight from the Fairchild data book!.

Audio fader and muting board.

Output line driver & headphone stages.

The control box was complete. A dead regulator (uA-723) I replaced with an LM-7805. The non-original IEC mains socket was turfed out, and I made a suitable mate for the Cannon XLR-LNE out of assorted bits of plastic & metal. XLR-LNE plugs and sockets are not permitted by the E-U nannies anymore, so buying them new is out. But who cares!? They are easy to make:

I also added a power on/off pushbutton to the control box, next to the hour counter, since as it was it had no method of local power off.

The biggest problem was the motor. "JD" had sent me the three-stepped pulley and its associated bearing, but the spare motor he had was burned out. Rand Armature Winders gave me a good re-wind quote on the phone, and then changed their price significantly when I arrived there with the motor. So, I strongly advise against doing any business with this company in Boksburg.

Finding a 1500RPM synchronous motor is harder than I imagined. Then I remembered that I had a NOS spare motor for a Grundig tape recorder - given to me by Jean of "Jeans Radio & TV Spares" when she retired and closed up shop a few years ago. This motor was electrically similar - also a Papst 4 pole 1500RPM synchronous hysteresis - but a little smaller than the Dok Viljoen's motor, and with completely the wrong shaft dimensions and mounting flanges.

Grundig motor attached to pulley bearing housing.

Shaft coupling.

Motor installed under platter

A motor transformer is now required (Ex Epson dot-matrix!).

Some aluminium tubing and threaded steel rod facilitated good stable mounting to the Dok Viljoen's vibration-isolation assembly. A plastic plumbing adapter, suitably turned in the lathe, allowed the Grundig motor shaft to mate correctly with the Dok Viljoen's drive shaft. The Grundig motor also differed in that it was rated at 39 volts! Another saviour from the past: the power transformer from a long-dead Epson LQ-500 dot-matrix printer - puts out... 38 Volts!

So now I had rotation. But the Dok Viljoen's instant start relies on a light disc of aluminium forming the platter 'mat'. This was cut from a sheet of 3mm aluminium sheet I had lying around. A custom-made wooden jig attached to a hand-held angle grinder allowed the roughly cut aluminium shape to be rotated against the grinder's cutting disc, thus grinding it to a perfect (to within 1mm anyway) circle. It's too large to anodize to the correct blue colour, so I have left it a brightly buffed raw aluminium finish. I might spray it with blue Aerolak one day.


The push-button lamps on the two control panels were originally supposed to be 36V 20mA: I could not find these anywhere, but 12V 80mA lamps are easy to get. I built an electrical interface using transistor drivers and a subsidiary power supply so I could install 12V lamps, without overloading the 36V supply (can't draw 80mA per lamp on that!). The original circuitry is thus unchanged, my board simply plugs in between the original wiring and the lamp terminals.
I tried LEDs in place of filament lamps for the buttons, but the LEDs looked very ugly and gave poor illumination.

The filament lamps shining through the clear coloured plastic pushbuttons have a very beautiful and serious look to them - so unlike the bright pinprick LED lighting on most audio equipment. I can never get the briliant colours of equipment displays to show up on photographs though.

Platter operation was now 100% The missing tone arm could be addressed.

The arm effective length is 11 inches, which is uncommon. No standard arm tube is long enough. Plus, most of the generic Japanese arm tubes are quite dense material and at this length would give an impossibly large effective mass. No supplier cold provide thin-walled aluminium tube of 10mm diameter; everyone had thick walled (1,7mm wall thickness). Thick-walled tubing when cut to the correct length weighed in at 25g - too high. I bound the tubing up on the outside with PVC adhesive tape an left it in sodium hydroxide solution. 48 hours later, the inner wall had been etched out providing a magically thin tube of only 12 grams.

I chose to use a straight arm, with an offset headshell. Precision bending of the arm tube seemed like too much work, and I prefer the look of a straight arm. Also, if I needed to change the offset angle, it could be done by replacing only the headshell. The headshell would have to be custom made too, since at this oddball arm length, the offset angle was going to be something non-standard too, no generic headshell would work. The headshell was made in exactly the same way as the one I made for my Marantz turntable, only this one has an offset angle designed in. After several hours of geometry and trying to interpret Baerwald's ramblings on tracking distortion, I gave up on the mathematics and went to the Vinyl Engine's web site, which provides an online Javascript app that instantly computes effective length, offset angle, overhang, etc. for any desired configuration. So easy! It even draws graphs of the tracking distortion for your chosen setup.

I already had the 'azimuthal' bearing, although someone had whacked it from the top, putting a ring of ball-bearing imprints into the bearing carrier. These were easy enough to polish out on the lathe. The 'elevation' bearing would have to be created. Luckily, the needle points were still intact.

The new bearing was made from 4mm diameter hard steel rod, cut to 10mm in length. The ends were countersunk to hold a circle of 6 ball bearings, into the center of which the needle bearing would project. The 1mm ball bearing I harvested from the swing-arm of a dead hard disc drive. The short steel rod passes through the aluminium arm tube - it is held in place by a brass slug - through which it also passes - inside the arm tube. Grub screws in threaded holes through the brass slug hold everything securely in position. The slug's origin is the earth pin of a 16 amp mains plug! The tone arm wires are the conductors from a fat 36-way screened cable I bought from A.P. They are marginally thicker than typical tone arm wire, but are easier to work with, and I could still get the necessary 5 strands through the 3mm hole.

On the matter of the final counterweight (145g), pictures tell the story best.

BEFORE.

AFTER.

And then I dropped it! = the dent on back edge, BAH Humbug!

The thumbscrew was later replaced with a grubscrew.

The cartridge, that was also 'lying around' in a manner of speaking. I had bought it for my Garrard 301 / Grace 640, but I think the Dok Viljoen is a better turntable and the new, unused Shure M97XE would get more use and enjoyment this way.

Trolley frame 1.

Trolley fame 2.

The trolley is made from around 8 metres of 25mm x 50mm x 2mm rectangular steel tubing, and some left-over 25mm equal-angle. The side panels and front skirt are veneered 16mm chipboard. The castors are attached by segments of 10mm threaded studding to allow for individual height adjustment. The trolley is now by far the most costly part of the revivification.

My welding is not great, and things always end up slighty out of alignment, so the frame is not perfect - from certain angles you might just see that certain steel sections are very slightly out of parallel. I did manage to ensure that the square on which the plinth sits is 100% flat - the plinth does not tilt or wobble at all.

I retained the original plinth (grey melamine veneered chipboad) because I want the 'look' of the old studios. It isn't very thick or heavy, but is now reinforced by being bolted around its periphery to a rectangle of 25mm x 25mm x 3mm equal-angle steel stock. This gives it the weight and rigidity of a substantially thicker wood or chipboard plinth.

These steel flanges are in turn riveted to the rectangular top section of the trolley, that being a square frame of 25mm x 50mm steel tubing. All-in-all then, the plinth is rigidly attached to quite a substantial mass of metal and chipboard.

The natural wood is entirely decorative trim - it gives no support or strength. It may add undesirable 'sounding board' characteristics and acoustic feedback though. In the radio studios, there were no loudspeakers so these turntables would never be subject to acoustic feedback. In a listening room, the situation is different, and may require attention.

Trolley view 1

Trolley view 2 (control electronics box)

Trolley view 3 (back interior)

Trolley view 4

The replacement motor (from Pabst for a Grundig tape recorder) is a switchable 4-pole or 8-pole 3 phase motor, giving the necessary 1500RPM, but also, if requireed, 750RPM. This will be useful for playing 16²/₃ RPM records of which I have several (speed set to 33¹/₃ but motor running at half speed). The new motor gets a dedicated control/interface board, not originally required. This board performs pole switching for the motor by means of 3 DPDT relays. An additional relay switches the new motor transformer's 220V primary. Also on the board are the 5 capacitors for the motor's phase-shift windings. Everything here came from my store, so nothing had to be bought.

The pole switching is activated by two pushbuttons on a custom-built front panel - which will be located just above the existing control box. Nice 'Rafi Lumotast' illuminated pushbuttons are very costly, but I used some doorbell buttons I had, which when lit from a pair of LEDS behind, give a very attractive coloured glow. The red "spot" is a neon AC-present indicator lamp. This panel also provides a location for the headphone sockets - for pre-listen and on-air monitoring, with their original bezel and labelling. The 'on air' socket gets is signal from some remote source further down the broadcast chain, so on this turntable it will likely remain silent.

Motor control circuit board

Motor pole select panel - 1500 RPM

Motor pole select panel - 750 RPM

Headphone monitor sockets (mono for now)

Motor contrrol board and transformer for 'new' motor.

The printed circuit boards on the back of each of the original speed and cue control panels (fourth photo above) are the two boards I made to convert the pushbutton lamps from 36V to 12V filament lamps. They each contain two transistors per lamp circuit, and an overall 9V regulator. The idea was to not modify the original circuitry in any way, so these two boards just slot in between the original bulbs' sockets and the new bulbs.

An annoyance I discovered with the original control box: The headphone driver amplifier is mono, but the PC board is laid out for stereo, so I only have to populate the second audio channel with components. Yet the metal inserts in the card-edge connector have not been installed in any of the unused contact locations! So I'll have to replace the edge connecter, or find an identical one from which to harvest contacts for the second channel headphone signal.

The empty spaces in the lower front section (below the control box, and above the motor pole select panel) I will fill in with either decorative wood, or aluminium sheet. The bottom space, it can be seen, has flanges for 19" rack ears, so I can install an amplifier or equaliser or any other component there if desired in future.

General view 1

General view 2

General view 3

General view 4