Designing a Rega Rim Drive

Today's readouts differ greatly from those in the 1960s not so much in looks but in the technology used to derive the results.

Rather than circuit board complexity much of the rather busy circuitry has been shrunk down to die size is now called a PIC, which can do just about anything, because it is programmed by a microprocessor which relies on lots of instructions (routines and sub-routines).

A hardware solution to achieve the same would be necessarily mega-complicated just to carry out simple functions. One of those simple functions being the control of readout updating such that it isn't seen.

A hardware solution will always require to stop and count, and we of the older generation simply accepted a bit of flicker or subtle flashing of the digital readout, probably because it was "new technology".

The only problem with this hardware frequency counter is the update period, which is only a blink of the eye, but it exists, and is a blink. And people are picky, and will complain that it flashes, and that the speed isn't stable - why?

In choosing to indicate two decimal places it will indicate system accuracy, otherwise why bother with the decimal places? I could simply wire the seven segment displays by a matrix (one matrix per selected speed, or frequency indication) such that it would indicate a steady 33.33 for example, and the customer would probably believe the turntable was doing a constant 33.33 rpm. However, that depends on faith, and in this scenario faith in my abilities, and there are a few scorners left who like to rubbish my work online. Therefore I have to go for a display which is a tell-tale of what the speed controller is actually doing.

As a mock-up with long wires everywhere, and contact resistance down to the fragile connections you will always get from prototyping "breadboards", and less than ideal grounding and decoupling, the speed controller fluctuates by +/-0.09%, which I consider good, but some won't, and that's because a truthful readout will show as much as 33.30 to 33.36 deviation from one reading to the next - the readings being each second or thereabouts. The inertia of the platter will not let the speed change by that degree in that time, and will average out to something like 0.01%, which won't show up on a strobe disc.

Quartz crystal control is an option, but this gives a fixed speed and if a pulley (or idler) is less than accurate, then it isn't much good. The Technics quartz controllers offer pitch adjustment but this takes them out of crystal control, which is only valid when the pitch control switches out in its centre position.

Therefore the only option where the user wants to vary the speed or trim it to suit, is an RC oscillator, and these depend on the stability of the components used. Here the components will be better than 200 parts per million, and the oscillator chips will only drift in frequency if the power supply voltage changes (1% between 3V and 16V). But as with all these things, from new they require running-in, or burn-in (at least here the readout will offer proof that it does exist).

So, back to the display flash. It indicates that the readout is being updated, and if the last digit actually changes, it doesn't look like it flashes at all. Currently the display is running considerably brighter than what would be expected in the listening environment. It is having to compete with bright fluorescent lighting, and adjusted to a much lower brightness the on to off difference will be much less obvious, and to most will not be an issue. I suppose I am labouring the point a bit too much, but I can always refer a complainant to this text.

Richardl60 wrote: I know this thread has been running for some time now but have never been sure whether this is a design exercise or development of an intended commercial product? In respect of the display referred to is this to allow adjustment or prove in the development whether the speed is accurate?

Hi Richard, I'd like it to be a commercial product but for the time being I want it to be the solution to my woes regarding the Rega Planar 3 turntable I bought in 1979.

Just about everything I have designed has become part of a commercial product at some time or another, even if some of them were not sold by me. My early electronics engineering career was spent, self employed, mainly designing printed circuit boards for other peoples products, and ocassionally designing the products as well. Most of it was to do with industrial control or related to telecommunications, and I was eventually let loose on audio when fate intervened and I ended up working for a customer from Sheffield who broadcast to most of Yorkshire and the north Midlands. And then I had "work experience" in hi-fi... when I discovered something rather smelly

Anyway, enough of my history, the display is to enable me to set up the speed and also a strobe disc can be used to show if the turntable's pulley and belt, or idler, or direct rim drive - whatever form of transmission is used - is doing it accurately.

It is possible that a strobe disc could show a difference between what is being presented to the motor and the resulting speed after it transmits its energy to the platter. The machining of parts may not be sufficiently accurate that a 50Hz (60Hz on 120V) frequency leads to the strobe indicating exact speed.

An AC synchronous motor will always do its rated rpm provided it is given its rated frequency. The voltage can fluctuate somewhat, but so long as the frequency stays constant its rpm will be exactly right.

To compensate for mechanical tolerances there needs to be variable controls (a potentiometer - like a volume control) which can be adjusted to increase or decrease the motor's rpm such that a strobe disc indicates correct speed.

There are also other factors:

1. The availability of 50 (60) Hz frequency lighting - a strobe disc only works under artificial light which strobes at 50 (60) Hz. The new LED bulbs, I am told, give a constant output and a strobe disc is useless under these conditions. Therefore it would be ideal if the controller was equipped with a bright LED lamp strobed at 50 (60) Hz, which could be shone on the strobe disc.

2. The motor used on the Rega and many other turntables vibrates when used at rated voltage, and this is why in every application I have seen, these motors are run at a slightly lower voltage. A heavy platter will often not start to turn, and requires running up by hand. A new owner may not understand this and think the turntable is faulty. The speed controller should therefore apply full rated voltage to the motor to get the platter revolving and then drop its output to a lower voltage.

This last factor (I can't think of any more right now) is not set in stone: it may be that some other voltage will give least vibration, and that adds yet another control for adjustment.

As for the display, the user might not want to see it all the time, thus requiring a switch, or that the display brightness doesn't suit the ambient lighting, this requiring a dimmer instead of a switch. And therefore a bit of a blink which is a necessary evil - it has after all got to pause to count the frequency - won't be a problem if the display is dim or off.

So now all that is left to do is put it all together as a unit in a box with the connectors necessary to hook it up.

Hook it up to what? This is a good question. The unit reconstructs a 50 (60) Hz "sine wave". It is not pulse width modulation like some other speed boxes.

So basically it produces using logic components a stable in frequency square wave which is filtered to produce a sine wave, and this has to be amplified to a voltage suitable for the motor.

A 230V rms output amplifier is a tall order - it would cost a small fortune and would need a lot of protection circuitry - and it could be dangerous in the wrong hands.

All the Rega like motors are 110V ac, this is more easily met. Still an amplifier outputting 110V is going to be rather expensive to produce, and can still be dangerous - it could be easily damaged!

Therefore it would be better to make it on the lines of a "pure" sine wave inverter, which uses a smaller amplifier which drives a step-up transformer, and that is what I've done.

For safety it is 110 volts ac. This means you cannot directly hook it up to a 230V mains Rega or similar turntable. The motor is 110V but it is supplied by a mains dropper board which drops it below 110 volts to reduce motor vibration.

The dropper board needs to come out and be replaced by a direct connection which must also include the phasing capacitor which causes the motor to rotate in the right direction. This is not something most users will want to do, or are qualified to do, so I have to ask about its viability as a commercial product. It would probably fall to a dealer to do this work.

Therefore, at present I am treating this as a personal pet project, but I am researching it and building it as if it is for a commercial product.

As for the rim drive, that is something I intend to contine working on once I have the controller up and running. It can simply run the existing belt drive for the time being. At least I will be able to make use of the RP3 whilst I think out the future steps.