turntable material choice.

Bibio

Member
2009-03-08 3:36 am
this has more than likely been done to death but i'm about to make a new top plate for my LPT. i could make one exactly like the original but this would involve water jet cutting which is a rare beast in Scotland these days.

so my plan is to make a solid (no cutout) top plate but have a few ideas i would like to run past you knowledgeable people first.

first idea is to use full Acrylic but in 2 pieces to make up the hight the top part would be 6mm and the bottom would be 15mm both bonded using Acrylic cement.

second idea is to use hexagonal alloy core sandwiched between two pieces of acrylic.

third idea is to use a full piece of foamed PVC

and final idea is to use a full piece of HDPE.

recommendations for other materials are more than welcome.

thanks
 

Bibio

Member
2009-03-08 3:36 am
Pink Triangle LPT

finish on the HDPE would be sand blasted.

TBH i'm sorta toying with the idea of the foamed PVC as i know the achromat is made out of it but it might not lend itself to being a top plate. when i say foamed PVC i actually mean Ex-Cel.
 
TBH i'm sorta toying with the idea of the foamed PVC as i know the achromat is made out of it but it might not lend itself to being a top plate. when i say foamed PVC i actually mean Ex-Cel.

Wouldn't you want something a little more "structural" than foamed PVC? Looks like the original top plate was perspex. I noticed there's a Pink Triangle forum over at the vinylengine.com. Why not search there for potential ideas/solutions.

jeff
 

Bibio

Member
2009-03-08 3:36 am
well i got some HDPE and all i can say is for those that think it might be a good material and by the sound (finger tap) its pretty dead sounding...don't use it even at 20mm its wayyy to sensitive to heat and bows all over the place.

so next up will be Acrylic.

the original is 18mm MDF with a 3mm Acrylic top with a cutout similar to the Roksan Xerxes. i don't want to use MDF as i'm not a fan of the material.
 

Bibio

Member
2009-03-08 3:36 am
thanks brian, i'm in Kirkcaldy and we have a Marble/Stone supplier here that has massive slabs of all kinds who supply the raw material for kitchen worktops and grave stone makers.

i actually had an idea of getting some ally hex core filler then bonding carbon or acrylic to one side then filling the webs with silicone rubber for mould making then bonding carbon or acrylic to the other side. my reasoning for this is that if you have ever held silicone rubber it's very 'flesh' like but not stiff so the hex core would keep it in check. ally/hex laminate is very very stiff and filling it with silicone would make it dense and non resonant.

i then have the other extreme where using a material that acts like a fine musical instrument in that the resonances start and stop quickly and do not linger.
 
calculations predict a slate/Al/slate composite to have poor material properties, with no 'mass damping' because the critical frequency is so low, as the stiffness is so high. And there is about 15dB amplification of vibration/resonance amplitudes around 440 Hz (near to where some platters ring).

I don't have any details on the hex-comb materials, and would like to get hold of a sample 100mm x 100mm would be ideal. Honeycombs are said to be very good structural materials, but I would use a plasticine-type material rather than a silicone elastomer, as the former has a better damping factor. Try the 'Newplast' brand.

I don't understand why you would use something that is used in a musical instrument, usually musical instruments rely on resonances and vibrations to produce sounds, quite the opposite of what is required here!
 
There's next to no correlation between how one material sounds in one deck and how it sounds in another deck when used in a different shape, thickness and with different mounting points and masses connected to it. You can draw some weak generalisations, like stiffer is better etc but that's about it. You need to build them you can't 'model' them in simulation, no matter what FEA you use.

There's too many variables at play, volume, stiffness, mass, transmissibility, absorption, and that's before you even get in to the physical construct of the thing, where the mounting points are, what they do to bending moments etc etc.

The two things you can be sure of are as follows, you need the motor, main bearing and arm mount to remain in an absolutely stable geometric relationship and you need them to be as mechanically isolated from one another as possible.
 
There's next to no correlation between how one material sounds in one deck and how it sounds in another deck when used in a different shape, thickness and with different mounting points and masses connected to it. You can draw some weak generalisations, like stiffer is better etc but that's about it. You need to build them you can't 'model' them in simulation, no matter what FEA you use.

There's too many variables at play, volume, stiffness, mass, transmissibility, absorption, and that's before you even get in to the physical construct of the thing, where the mounting points are, what they do to bending moments etc etc.

The two things you can be sure of are as follows, you need the motor, main bearing and arm mount to remain in an absolutely stable geometric relationship and you need them to be as mechanically isolated from one another as possible.


I am afraid that I agree with the above. [Slate has proved very successful as a plinth for idler Lenco's Garrard 301/401, and direct drive Technics SP 10 MK11& Mk111 along with many other premium decks.]

Using honeycomb was a fashion many years ago with one TT maker [Commercial aircraft flooring panels], but can't remember who. It had a certain marketing appeal, but I don't believe that it lasted that long - nor do I believe that it improved that TT's performance by any significant degree.

This is not meant to discourage you from trying that approach, as it is only by doing so that you will get an answer.

But the second section quoted in blue above is essential for any credible TT design.

Best Luck,

B
 
Last edited:
calculations can be made to provide a means of estimating the losses expected in simple structures, with as little as the dimensions of the plinth, Young's modulus, Poisson ratio, density and damping factor of the material/s. From these data, such parameters as fundamental, critical and dilatational frequencies can be calculated, along with bending stiffness. This can be applied to single materials, three layers with same facing, and 5 layers with same facings, and I'm working on a calculation for a three layer with different facings.

From these calculations, the losses in the panel can be estimated; this is for a simply supported structure, as could be used for simple stone, wood or plastic plinths.

You are correct that more complex structures would need to be modeled more appropriately, but most plinths only need a simple design..

There is no argument as to what is required, but isolation of components is neither easy to attain, nor necessary, as long as the material/s of choice are appropriate. On this planet, like electronic circuits, all parts will (hopefully) be tied to the same earth, and therefore, not isolated.

Lastly, honeycomb structures have been the subject of much academic research, recently, for the control of structural vibrations. It was used, although not too successfully, in the Celestion SL600 loudspeaker enclosure, where they used 'Aerolam'.

In a slightly tongue in cheek response, slate has proved very successful to those who have shed blood, sweat and tears over the construction of a plinth. Of course its going to sound good after all that. Many who have heard slate have gone onto more appropriate materials.
 
Nothing wrong with wood...

...although Pink Triangle had spent time, effort and money to get the sound they wanted. Maybe check out pinkfish for ideas as it seems to be a pro-PT site. There must be something out there.

I won't restate my material preferences, because I've stated them so often. Here's a place to start: consider the physical properties of the vinyl used to make records and go from there.