Magnetic turntable bearing

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have just built a magnetically repulsing bearing to retro fit to my turntable. The turntable itself is a home built and designed affair but the main bearing and platter are from a Linn LP12. The magnetic bearing consists of two opposing rings of small (6mm diameter) Neodymium magnets pres fitted into CNC machined acrylic discs. The intention of using many small magnets (40 in each half) was to provide an even magnetic field with no cogging effect. The magnets are placed very close to each other and surround the main bearing. A bit of ‘fancy’ machining was required but I am lucky enough to have easy access to a CNC router. I am yet to do final measurements but I think the magnets suspend the platter by as much as 5mm but I will investigate further tomorrow. Note, the main bearing shaft is still used for horizontal alignment.

To say that I am delighted with this modification is a huge understatement. I have read elsewhere on the forum that the edge of the platter can be seen to rise and fall with some magnetic bearings; this is not the case in this instance. Friction levels are substantially reduced and background noise is hugely reduced. Bass instruments are significantly easier to follow and the whole sound is substantially more open and natural. Literally every aspect of reproduction has been enhanced. In the past I have experimented extensively with turntable design and have investigated many materials and construction methods. My current design uses quite an elaborate, dual layer, suspension, for example, to reduce the effects of feedback. However, this bearing modification is the largest single improvement I have so far achieved.

I strongly urge anyone planning on building (or currently doing so) a turntable to investigate this type of bearing as they may also be as pleased with the results as I am.
 
Hello Graeme, I too live in Sheffield S11.

It may be possible to do as you suggest though it would require a significant re-working of my turntable and stability could well be an issue. In fact, thinking about, it as I type, I think the side force exertd by the belt would be a real problem. One of the magor advantages of my mod is that it is entirely retro fitable and can easily be removed if so desired.

I now believe the magnetic gap to be in the order of 3mm or so and as such there is very little 'bounce' from the magnets.
 
The cartridge seems to be completely un affected (a re-built Linn Troika) - still experimenting though. The platter is not particularly thick - about 12mm in the area where the magnets are placed. As I said before, the platter and inner platter/main bearing are from a Linn LP12. The platter is cast (post machined on all surfaces) from Mazak which is a lead, Zink alloy.
 
A few pictures would be nice...

BTW Mazak- or it's more common trade name Zamac, is a zink/aluminium alloy.......


The use of a spindle bearing handling the side load only, and magneitc repulsion in the vertical, I beleive was first seen in the "Platine Verdier" late 80s....
 
Yes, i can show you all a few pics - as I've just taken a few :)

I stand corrected as to the alloy used in my platter, I thought some lead was included but perhaps not.

I am aware that the "Platine Verdier" used a similiar principle. I don't claim to have invented anything particularly new, but it does seem to work remarcably well.

I'm gonna mess about with the photo thing now.
 
OK here are the pics:

This is the lower part of the bearing. White acrylic at the moment but I will re-make it next week in black. The 3 hex head bolts simply hold the three layers together. The layers are required to raise the magnets sufficiently.
An externally hosted image should be here but it was not working when we last tested it.


This is the back of the part pictured above. The three lobes are to create a defined location about the bearing housing.
An externally hosted image should be here but it was not working when we last tested it.


The same part in position on the subchasis. The magnetic bearing is not fixed in place in any way.
An externally hosted image should be here but it was not working when we last tested it.


The upper part of the bearing. The thee central lobes are press fitted to the centre boss and the bearing sits flat against the lower surface of the inner platter (as shown).
An externally hosted image should be here but it was not working when we last tested it.


I also made myself an acrylic platter matt (3mm). The matt is made from five individual components that fit together like a jig-saw. The platter has a small lip about the outside edge and this is sufficient to locate the mat in position. No glue is required and the matt simply does not extend beyond the edge of the record label.
An externally hosted image should be here but it was not working when we last tested it.



The subchasis shown is made from two layers of Fibrelam laminated together. Fibrelam consists of carbon fibre skins over a centre core of phenolic impregnated paper honey comb structure.

The corner elements just vissible in one of the pics are the lock nuts for the fine level ajustment. Here is a not very ggod pic of the whole deck. The stand is structural to the player and althoufgh it cant be seen in the pic the lower plinth is in fact split and contains another level of low frequency de-coupling. The upper section is in effect another subchasis. This result was arrived at after much experimentation. Another platter matt is shown in this pic where the record rests on five metal points that make contact with the record label. The black disc shown does not touch the platter or the record and is only ther to position the metal elements correctly.
An externally hosted image should be here but it was not working when we last tested it.
 
Very nice indeed!

I would like to complement you on the excellent execution of your design.

I have 3 questions for you:

1) Did you consider using ring magnets instead of a large number of smaller magnets?

2) How are you keeping the magnets captive or in position?

3) How did you calculate the necessary "repelling force" to suspend the platter assembly?
 
Looks very good. If i was using a linn bearing/platter i'd want to do the same. I might try something similar as an experiment if its working well for you.

So, theres three of us from sheffield on here. I was wondering, would you be prepared to demonstrate your TT to me? Ive been building/modifying my TT for a while now but have never heard another decent TT so dont really know how mine stacks up. It would be great to be able to compare it to some others, or at least get someone elses opinion on mine (i could bring it with me).

Im using a much modded lenco chassis with a rega bearing/platter and a ladegaard arm.

P.S. it looks like your building these to sell? If so, i think your onto a winner:D
 
Thank you everyone for your kind words it means a lot to me. I will try to address your questions:

I did consider using ring magnets but I was concerned that the cogging effect as the magnets moved across each other would be significant, Just two large ring magnets would have been possible but these are difficult to source.

The magnets are held in positiob by this means; each disk has 40 holes machined into it with a depth of 4mm. The magnets are a very tight fit and are pressed into place - each magnet has a height of 5mm, so 1mm protrudes beyond the face of the acrylic.

I would like to say that I calculated the force of the magnets required but unfortunately I did not. i did build a test rig with 20 magnets and from this I suspected that 40 would be about right - luckily my guess was pretty good.

It is indeed a Naim Aro in the picture. I am trying to develop an arm of my own but have yet to narrow my design goals sufficiently. I didn't actually pull an LP12 to pieces to build this. All the Linn parts were bought as spares. The deck is powered by a Naim Armegedon power supply too :)

I would be happy to demonstrate mt T, table to a fellow enthusiast (or two). E-mail me to arrange a good time. It would be great if you brought over your own deck.

I'm not building to sell though - I don't think there is much profit to be made these days for turntable manufacturers - not for me at any rate; very kind of you to say so though.

The number of magnets is exactly the same in the upper and lower sections.
 
YNWOAN said:
I did consider using ring magnets but I was concerned that the cogging effect as the magnets moved across each other would be significant, Just two large ring magnets would have been possible but these are difficult to source.

Actually I think that individual magnets would be more susceptible to cogging than 2 opposing ring magnets which offer a constant circular magnetic surface/field.
The challenge, as you mentioned, is to find large enough ring magnets. I did find a source recently that had magnets strong enough for my application but they are only 75mm in diameter. The inside diameter 25mm.

What do you think a ring magnet of those dimensions would do since the inside diameter of the ring magnet would be very close to the bearing shaft? In my case I use a 16mm shaft.
 
does the platter height vary at all during a full rotation? i'd be interested to measure this visually with a laser level.

if there is no verticle variation then their should be no cogging either, as any varience in repulsion will manifest itself as changes in height.

No doubt you know the answer to this already YNWOAN.

care to enlighten us.

it certainly looks like particulalry well executed,do you have access to CNC or laser cutting for the jigsaw pieces?
 
ps: re the fibrelam, if you lay this up yourself you might want to try backfilling the Nomex with sand loaded epoxy before you add the final skin.

nomex/carbon is very stiff, and has great mechanical damping to loads, but every little helps.

it adds an awful lot of mass to the composite, though it can be awkward for gassing out if you dont pin roller the carbon skins.

i can bring some samples of cast epoxy/mineral loaded blocks if you'd like to see a few.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.