A year ago I introduced Project Leoht, my take on an ultra-low mass turntable inspired by Rega Research. I learned a lot from that project and version 2.0 has been in development, with physical part prototypes in my hands and several new ideas in my head and on paper. I wanted to write this article in the interim to discuss some of the new developments and answer some frequently asked questions that I have received since publishing Leoht V1.
The original Leoht pushed my engineering capabilities and knowledge to the limit. Developing version 2 has meant a great deal of learning and a lot of late nights. There wasn’t anything ‘wrong’ per se with the original, but I am rarely one to call a project ‘good enough’ – I have to improve anything to the nth degree. I only ever intended to build one Leoht, to test the ultra-low mass turntable principle for myself. Yet here we are, in the midst of a second revision.
I’ve been working closely with my good friend Gary of Garp Designs on prototyping the plinth for version 2. The goal is to streamline it further, making it lighter, stiffer, smaller, less resonant and more accurate. Improving on the V1 plinth is no mean feat, though designs are in the works that will mean an almost 50% reduction in material and thus in mass, while at the same removing points of mechanical stress, streamlining the design aesthetically and improving the distribution of load with respect to the platter bearing centre of gravity.
Part of optimising load distribution involves moving the feet to achieve optimal balance. Improving balance will alleviate significant amounts of stress in the plinth and stiffen the crucial area of the structure between the platter bearing and arm. Excessive stress here due to an imbalanced support can increase bending modes, which will affect the precise relationship between the centre point of the record and the point about which the arm pivots, and thus, microscopically, the alignment of the stylus relative to the groove.
The brace between platter bearing and arm will also be a great deal stronger. To that end, I am hoping to use ceramic bracing in version 2. Version 1 used carbon fibre bracing which is plenty strong, but a ceramic top brace will increase rigidity in this crucial part of the structure.
I wanted to improve the way the plinths are produced. I’m still going to use Carbon Fibre and Rohacell, but I’ve invested in the equipment to vacuum bag the assembly. Vacuum bagging will provide a perfectly even clamping pressure across the entire panel surface, and with an ideal quantity of adhesive calculated and applied should result in the best possible bond between the foam core and the carbon fibre skins.
I am hoping that I might be able to obtain some graphene-impregnated carbon fibre prepress, or a graphene-impregnated epoxy adhesive. So far none of the companies who are involved in such things have taken the time to respond to my enquiries. If any readers are involved or knowledgable in the use of graphene in prepregs or adhesives, please do get in touch.
Other plinth modifications include the use of custom-made threaded sleeve bushings, which will be bonded into the plinth to support the feet, motor mount and arm bolts. These will eliminate compression from nuts and bolts as used in the current design to help mitigate mechanical stress.
The drive system is undergoing a rework. The principle is the same, an aluminium sub platter riding in a VHL polymer bearing and driven by a triple belt arrangement from a machined brass pulley. All of these parts have been remade to tighter tolerances. A slimmer subplatter will better optimise the centre of gravity and concentricity are even better than the already outstanding results we achieved with V1.
The bearing housing will screw directly into the plinth with a thread cut into the plinth itself, maximising surface contact and all but eliminating mechanical stress in either component.
I’m working on a new platter in manmade Quartz. Ceramic oxide is out of my league both financially and in terms of what I can manufacture. However quartz stone is similar in hardness and is easier to work with. Prototypes to optimise the platter size and shape will be produced in other materials first, before they are used to draw up the design in CAD from which the Quartz platter will be machined, once I am able to find someone with a CNC machine who is willing to cut a challenging material. The platter shape, in particular the underside profile, will be optimised to create a dynamically balanced flywheel.
One of the toughest challenges I’m facing is the motor. Manufacturers of belt-driven turntables use either a synchronous AC motor, with various driving techniques, or a DC (direct current) motor, typically a brushed design. Brushed DC motors are easy to control, as you need only vary voltage to adjust the speed, and current delivery (within reason) dictates torque. However their torque vs load curves are suboptimal, they’re mechanically noisy, the brushes have a finite life, and to achieve extreme accuracy requires some kind of external feedback system and a complex control algorithm to control the motor without causing excessive, rapid corrections.
Rega use AC synchronous motors, specifically Premotec 318-series motors. These were designed in the 1950s by Philips for use in turntables, and the design hasn’t really changed since; though the price certainly has. They’re similar to a stepper motor in construction, but the motors built for turntables have reduced magnetism to reduce torque ripple and the ‘bumps’ as the rotor passes the stator poles. Driven with a twin-phase sinusoidal waveform, they can operate very quietly and with minimal vibration.
But they’re large, heavy and expensive. The 31823 motor used in Leoht V1 was heavier than the entire plinth. They’re also not particularly well built, with cheap shaft bushings, no thrust bearing and limited means of mounting that don’t cause mechanical stress.
I had investigated, and discounted, the idea of driving an off-the-shelf stepper with a sinusoidal ‘microtepping’ technique, as while it works perfectly in a turntable application those motors are physically similar to the Premotec I wanted to replace. The promoted design has a 7.5-degree step angle, and a 7.5-degree stepper will operate at 250RPM if driven at 50Hz. Torque ripple and noise can be reduced using a motor with a finer step angle, 1.8 or 0.9 degrees for example, and the same nominal speed can be achieved by increasing the drive frequency. A higher speed motor could also be used if the pulley and belts were adjusted to suit, but I was keen to retain those parts of the design as I already have the tooling to produce my belts and initial pulleys have already been made.
The solution? Design and build a new motor, of course. Not a small undertaking by any means, but I felt it was the only way I was going to achieve the kind of performance needed to take Leoht a step further. The new three-phase, brushless motor will be powered by an advanced software controller that is currently in development. This project pushed not only my design skills but also my rusty programming skills beyond their limits, but with working prototypes a reality it is no-longer the pipe dream it once seemed.
The feet are also being re-designed. The originals were 6000-series aluminium with a nylon insert, and optimised as much as possible to reduce their mass and to prevent high-frequency self resonance. But I felt I could go further. Advancements in 3D printing technology allow not only extremely complex, thin-wall structures to be printed, but those structures can also be formed simultaneously in layers of different materials which become one as the layers are printed.
Combining materials, experimenting with in-fills and wall thickness and optimising the shape and face patterns has resulted in prototypes that are as close to the ideal turntable foot as I think it is possible to get, cutting mass more than 10-fold and exhibiting exceptional resonance characteristics. I can’t wait to try them on the turntable.
Infills will also be used to strengthen the motor mount, while reducing its mass and optimising resonance. This prototype motor mount was made for the Premotec motor so there is some revision to come yet.
The arm remains the same but for a new underhung counterweight system, more of which will be revealed when V2 is released. Tolerances in the arm mount will be vastly improved though, giving an even tighter interface and more surface area contact between the arm mount and the plinth. I’ll revisit the arm in version 3, once the rest of the design is optimised as far as possible. While it is important to consider the design holistically, designing an arm for V2 is unfeasible.
That’s all I have to share for now. I’m pleased to say Leoht is advancing further than I could have imagined with version 2. I’m excited to have the finished build up and running, whenever that may be. The feeling you get when you play the first record on a turntable you’ve spent years developing is indescribable. If you enjoy reading about my work and would like to stay yup-to-date with future developments, do subscribe to receive those updates as and when they are published.
Share Your Thoughts