Better to not look at the Teres drawings so there is no quibbling about IP. The Teres bearing appears to be a "bearing at the bottom" design, whereas it seems like "bearing at the top" designs are a lot more popular with high-end TT"s. What is the opinion here? If the top bearing is zirconium it may not need any lubercation and the sleeve could still be oiled via an archimedean screw. It looks like the bearing shaft is tapered? Is that right?
Exceps for the dim's , there's really nothing special about the Teres bearing. In principle it is similar to the vast majority of TT bearings, even of the so called very high ends. Given the low rotational speed of a TT and sufficient accuracy in manufacture, it should be plenty good for an extremely good TT, and by far simpler for any potential DIYers. The trick lies in the position of the platter total COG relative to the bearing point
Last edited:
That's one reason I like the bearing on top design. Easier to get the bearing point up inside the platter and in line with the belt. Of the top of my head I don't think the machining work would be any more difficult.
You also bring up a good point that the bearing design and the platter design should go together, i.e. be a single design.
You also bring up a good point that the bearing design and the platter design should go together, i.e. be a single design.
Pardon me for little aberrant thinking. Would a more mass of same material added to bearing drain noise ? Would little pockets in the metal at bearing end filled with sand improve it further ? Few threads at the top of the spindle design will be help full so one can tighten center weight on vinyl, so one can use lighter (Less load on bearing) center weight. In some other thread there was mention of little magnets which repels platter upwards so noise can be reduced*. But for stability Ball Bearing would touch the thrust plate though. If suitable material is available for bush which holds the shaft I guess we can get away with regular circular bush and replace it with some what reuleaux triangle** shape. Not as extreme as reuleaux but just so that shaft touches the bush at only three points.
*verdier turntables
** Connoisseur turntable with nylon bush.
Regards
*verdier turntables
** Connoisseur turntable with nylon bush.
Regards
Here's a drawing I made based on the pictures that AuroraB posted. Does this look right/workable for a Teres style bearing? I can machine and have access to a pretty decent lathe and will run off a prototype once I get some feedback from this group.https://dl.dropboxusercontent.com/u/4582999/BearingAtBottom.pdf The drawing is missing some relief around the long part of the bearing. It should only contact the shaft at the top and the bottom, not continuously as shown.
Sorry, I should have inserted an image... noob error 
An externally hosted image should be here but it was not working when we last tested it.
Thanks guys for all replies.
Finally I solved with the problem with the bearing.
I stopped developing my diy bearing due to imprecision of my lathe machine, so temporarily I'm using a VCR head, and seems work really good.
Next winter I will go ahead with a diy bearing.
Here the finished turntable
Here the video of the complete work
https://www.youtube.com/watch?v=UXvz3y3JXw8
Cheers
Luigi
Finally I solved with the problem with the bearing.
I stopped developing my diy bearing due to imprecision of my lathe machine, so temporarily I'm using a VCR head, and seems work really good.
Next winter I will go ahead with a diy bearing.
Here the finished turntable
An externally hosted image should be here but it was not working when we last tested it.
Here the video of the complete work
https://www.youtube.com/watch?v=UXvz3y3JXw8
Cheers
Luigi
my 2 cents ..
Hello all,
25ys ago, i built a DIY TT. This TT still runs fine and sounds great and the only reason i might replace it some day is that i have some design/concepts i would like to try out. But sonically, i have not the slightest pain pressure to replace it. Platter weight is 4kg and i recently updated the motor and motor control (see http://www.lencoheaven.net/forum/index.php?topic=18696.135 , post #139)
To obtain a premium platter bearing and platter, i went to a machinist/toolmaker capable of milling, turning and precision grinding. The machinist firm usually makes tooling like SK40 milling shafts for CNC machining.
Here is how my bearing looks like: the shaft measures 19.98mm, the bushing 20.00mm in diameter, i.e. the radial gap between bushing and shaft is exactly 0.01mm. Don't try this at home! Have an expert do this for you! You may obtain the dimension but you will NOT obtain sufficient cylindricity !
The oil used is equivalent to Exxon Spinesso 10 (viscosity only slightly higher than water: 32mm²/sec); Warning: any oil with higher viscosity will NOT cooperate with a bearing gap that small !
Both shaft and bushing are made of the same tooling steel and hardened to about 55 Rockwell C before grinding. As long as there is enough oil in the bearing, shaft and bushing will never touch during operation. And temperature is not an issue as materials of both bushing and shaft are the same.
The shaft's bearing surface is about 65mm long and has some pockets to keep the oil distributed; a spiral groove was discussed and discarded, due to manufacturing costs . The interface between platter and shaft is a shortened Morse taper; at the smaller end there is the 7.2mm record centering pin, having an M4 threaded blind hole for the record clamp.
The bushing's metal bottom holds the PTFE/Teflon trust plate, 3mm thick. The platter/shaft is resting on a 4mm ruby ball, seated in the shaft's conical centering hole. The ball is a precision calibration ball, not selected for its hardness but for its extraordinary smoothness and negligible deviation from sphericity.
The ball creates an indent in the PTFE trust plate. But the cold flow of the PTFE subsides as soon as the ball has sunken in by a certain percentage of its diameter. So the bearing is long-term stable -- and dead-silent. And as the turntable's PRaT, dynamic stability and low-end performance never disappointed me, i had no reason to try other materials for the trust plate (but POM/Delrin, Vesconite, IGUS bearing plastics could be worth a try). Hope this helps.
Hello all,
25ys ago, i built a DIY TT. This TT still runs fine and sounds great and the only reason i might replace it some day is that i have some design/concepts i would like to try out. But sonically, i have not the slightest pain pressure to replace it. Platter weight is 4kg and i recently updated the motor and motor control (see http://www.lencoheaven.net/forum/index.php?topic=18696.135 , post #139)
To obtain a premium platter bearing and platter, i went to a machinist/toolmaker capable of milling, turning and precision grinding. The machinist firm usually makes tooling like SK40 milling shafts for CNC machining.
Here is how my bearing looks like: the shaft measures 19.98mm, the bushing 20.00mm in diameter, i.e. the radial gap between bushing and shaft is exactly 0.01mm. Don't try this at home! Have an expert do this for you! You may obtain the dimension but you will NOT obtain sufficient cylindricity !
The oil used is equivalent to Exxon Spinesso 10 (viscosity only slightly higher than water: 32mm²/sec); Warning: any oil with higher viscosity will NOT cooperate with a bearing gap that small !
Both shaft and bushing are made of the same tooling steel and hardened to about 55 Rockwell C before grinding. As long as there is enough oil in the bearing, shaft and bushing will never touch during operation. And temperature is not an issue as materials of both bushing and shaft are the same.
The shaft's bearing surface is about 65mm long and has some pockets to keep the oil distributed; a spiral groove was discussed and discarded, due to manufacturing costs . The interface between platter and shaft is a shortened Morse taper; at the smaller end there is the 7.2mm record centering pin, having an M4 threaded blind hole for the record clamp.
The bushing's metal bottom holds the PTFE/Teflon trust plate, 3mm thick. The platter/shaft is resting on a 4mm ruby ball, seated in the shaft's conical centering hole. The ball is a precision calibration ball, not selected for its hardness but for its extraordinary smoothness and negligible deviation from sphericity.
The ball creates an indent in the PTFE trust plate. But the cold flow of the PTFE subsides as soon as the ball has sunken in by a certain percentage of its diameter. So the bearing is long-term stable -- and dead-silent. And as the turntable's PRaT, dynamic stability and low-end performance never disappointed me, i had no reason to try other materials for the trust plate (but POM/Delrin, Vesconite, IGUS bearing plastics could be worth a try). Hope this helps.
igus ...
Hello sq225917 an all,
I had the same experience as you with PTFE filled with particles or fibres for bearing purposes, it is no longer silent. IGUS has their materials filled with some fibres to make them stable ... and after all, they all are meant to be self-lubricating dry plain bearings. However, they have such a variety of different bearing materials ... something suitable should be among them.
Vesconite or Vesconite HiLube is definitely something i would like to try. It is a combination of a polymer matrix and embedded PTFE dry lubricant. But ... it is meant for use under unlubricated or water-lubricated conditions; it works under oil-lubricated conditions but not at its best. Anyway, worth a try.
Another thing: one can scale up the load capability of the ball-and-trust-plate structure simply by increasing the ball's size and the thickness of the PTFE trust plate. As the volume of PTFE displaced by the ball (cold flow) is proportional to force and as volume is proportional to ball radius squared, tripling the ball size increases the force (the platter weight to be carried) by a factor of 9.
BTW, i was mistaken with my bearing ball: i looked up my notes and a ball size of 3mm only would yield a stable bearing. Currently i use a ball of 3.5 mm diameter.
Ruby calbration balls of 12.7mm diameter can be bought for not too much money; that would be 3.62 times my current bearing ball. It would carry more than 52 kg and still have headroom ... should be enough for any platter.
My notes also told me that i tried out POM as trust plate: the bearing was stable and and the indent made by the ball was much smaller (hence the load carrying ability is higher).
I also tried out steel (soft and annealed/hardened), tungsten carbide, brass, bronze and copper as trust plate. The hard metals did not stay silent for more than a few hours, oh, and they were sooo easy to damage. The soft metals lasted longer, copper was best. However, they did not reach the complete absence of bearing noise that i achieved by using a PTFE trust plate.
My system back then did not have the detail resolution to reveal significant sonic differences between different trust plate materials. Might be different today .... my buddy Hartmut Quaschik reported that different trust plate materials can be used to tailor the TT's sonics.
Hello sq225917 an all,
I had the same experience as you with PTFE filled with particles or fibres for bearing purposes, it is no longer silent. IGUS has their materials filled with some fibres to make them stable ... and after all, they all are meant to be self-lubricating dry plain bearings. However, they have such a variety of different bearing materials ... something suitable should be among them.
Vesconite or Vesconite HiLube is definitely something i would like to try. It is a combination of a polymer matrix and embedded PTFE dry lubricant. But ... it is meant for use under unlubricated or water-lubricated conditions; it works under oil-lubricated conditions but not at its best. Anyway, worth a try.
Another thing: one can scale up the load capability of the ball-and-trust-plate structure simply by increasing the ball's size and the thickness of the PTFE trust plate. As the volume of PTFE displaced by the ball (cold flow) is proportional to force and as volume is proportional to ball radius squared, tripling the ball size increases the force (the platter weight to be carried) by a factor of 9.
BTW, i was mistaken with my bearing ball: i looked up my notes and a ball size of 3mm only would yield a stable bearing. Currently i use a ball of 3.5 mm diameter.
Ruby calbration balls of 12.7mm diameter can be bought for not too much money; that would be 3.62 times my current bearing ball. It would carry more than 52 kg and still have headroom ... should be enough for any platter.
My notes also told me that i tried out POM as trust plate: the bearing was stable and and the indent made by the ball was much smaller (hence the load carrying ability is higher).
I also tried out steel (soft and annealed/hardened), tungsten carbide, brass, bronze and copper as trust plate. The hard metals did not stay silent for more than a few hours, oh, and they were sooo easy to damage. The soft metals lasted longer, copper was best. However, they did not reach the complete absence of bearing noise that i achieved by using a PTFE trust plate.
My system back then did not have the detail resolution to reveal significant sonic differences between different trust plate materials. Might be different today .... my buddy Hartmut Quaschik reported that different trust plate materials can be used to tailor the TT's sonics.
Bernhard, my own deck is Kuzma Stabi, but it uses the bearing and subplatjer from an LP12, the bearing is supported on a PTFE at its tip and the whole thing is run in heavy silicon oil. The journal walls of the LP12 bearing are Vesconite (well the older white ones are). It works well, bearing tip creep into the pt. settled long ago and it runs amazingly smoothly.
I still hanker after making a much larger diameter bearing with a 20-25mm shaft diameter and vesconite journals top and bottom.
I performed some wavelet analysis of bearing noise with and without teflon pad, the difference is pronounced. Not only does bearing noise drop by a healthy 2-3db under 100hz but the distribution of the noise changes as well. It becomes more broadband in frequency and sharp spikes are all but removed.
A metal/metal contact appears to help highlight the high frequencies, whereas the ptfe pad has a more damped sound. I'm sure the is 'very' design dependant.
I still hanker after making a much larger diameter bearing with a 20-25mm shaft diameter and vesconite journals top and bottom.
I performed some wavelet analysis of bearing noise with and without teflon pad, the difference is pronounced. Not only does bearing noise drop by a healthy 2-3db under 100hz but the distribution of the noise changes as well. It becomes more broadband in frequency and sharp spikes are all but removed.
A metal/metal contact appears to help highlight the high frequencies, whereas the ptfe pad has a more damped sound. I'm sure the is 'very' design dependant.
Hello sq225917,
your measurement are interesting. I did not pay heed to how the trust bearing influences sonics so far. My main focus back then (and still is) to have a stable bearing i do not have to re-build every week or so to get rid of the noise.
Silicone oil: i found its damping properties to be very pronounced (depending on viscosity, of course) but its lubricating and load bearing properties to be inferior compared to mineral oils. For that reason, i mistrust silicone oils.
I made my bearing's gap intentionally small and kept my chosen oil viscosity intentionally low. I gambled on the concept that the lubricating oil film's stiffness is more important, precision-wise and sonically, than the bearing's absolute load carrying ability -- and i won that gamble.
My reasoning was that a platter (during playing a record) never is loaded so heavily that the oil film is disrupted and bushing and shaft come into hard metal2metal contact.
The differential stiffness of a thin oil in an appropriately narrow bearing gap however yields to much higher rotational precision under realistic load conditions.
The bearing also puts a smooth but considerable base load on the drive motor: If i spin the platter to 33rpm without a belt attached, the platter very smoothly comes to a rest after about 2 revolutions. This is due to the high "shear" in the oil film (it happens without ANY noise). High base load on the motor is a good thing: the higher it is, the more the stylus drag force variations drown in the base load and the easier it is for the motor control unit to keep the platter speed stable.
Several buddies have described my TT's sonics as comparable to a big MicroSeiki with a 16kg platter, particularly at the low end of the audible spectrum.
your measurement are interesting. I did not pay heed to how the trust bearing influences sonics so far. My main focus back then (and still is) to have a stable bearing i do not have to re-build every week or so to get rid of the noise.
Silicone oil: i found its damping properties to be very pronounced (depending on viscosity, of course) but its lubricating and load bearing properties to be inferior compared to mineral oils. For that reason, i mistrust silicone oils.
I made my bearing's gap intentionally small and kept my chosen oil viscosity intentionally low. I gambled on the concept that the lubricating oil film's stiffness is more important, precision-wise and sonically, than the bearing's absolute load carrying ability -- and i won that gamble.
My reasoning was that a platter (during playing a record) never is loaded so heavily that the oil film is disrupted and bushing and shaft come into hard metal2metal contact.
The differential stiffness of a thin oil in an appropriately narrow bearing gap however yields to much higher rotational precision under realistic load conditions.
The bearing also puts a smooth but considerable base load on the drive motor: If i spin the platter to 33rpm without a belt attached, the platter very smoothly comes to a rest after about 2 revolutions. This is due to the high "shear" in the oil film (it happens without ANY noise). High base load on the motor is a good thing: the higher it is, the more the stylus drag force variations drown in the base load and the easier it is for the motor control unit to keep the platter speed stable.
Several buddies have described my TT's sonics as comparable to a big MicroSeiki with a 16kg platter, particularly at the low end of the audible spectrum.
Your platter comes to rest in only 2 revolutions? I rebuilt my spindle bearings on my Pioneer PL-41D using delrin for the thrust pad and light hydraulic oil for the lubricant. The spindle and bronze bushing looked as new, the honing cross marks still fresh looking in the bore and the spindle looking like it had just come off the grinder. Clearance between the 2 pieces is well less than 0.0005". Spinning the 2.2Kg platter resulted in about 3.5 minutes of rotation before it finally came to rest.
BillWojo
BillWojo
Linn's arguement was that the point bearing they use has the smallest number of tollerences in it's concept ( mostly right ). 1/10000 inch stated. Rega for what was less than $1 at the time worked to that ideal. As best I know the Rega spindle is an element from an aircraft roller bearing ( SKF ? ). This is already to a high standard. The outer is plain bronze ( not phosphor ). The process CNC machined. Process one is to simply drill the bronze. The drill point will average the centre location for the ball to sit in. Next a carefully adjusted reamer ( not fixed type ) finishes the job. Many test samples are made to get it right which seems to be one weeks work. After that a large production run. When a bearing is ground on cenres with a ball it might be slightly less well centered. Not bad for $1. Roy was with Ford so learnt how the big boys do things. He designed gearboxes I was told.
I think to spin a bearing without belt above 33 1/3 with a stroble is a great idea. Start counting when it hits 33 1/3. If an optical system it could log data, that might be nice. Drag is a good thing and notches a bad thing. 45 is equally good. I have an easy stobe design if people need it.
I think to spin a bearing without belt above 33 1/3 with a stroble is a great idea. Start counting when it hits 33 1/3. If an optical system it could log data, that might be nice. Drag is a good thing and notches a bad thing. 45 is equally good. I have an easy stobe design if people need it.
Hello Bill and all,
sorry, , posting with with my eyes already bleary and 
not checking my spelling was sloppy. My Apologies! I meant to say 12 revolutions. So, to be sure, I just tried it out again: platter comes to rest after 10 revolutions 
. But coming to rest after only 10 revolutions still means a lot of drag.My bearing's bushing is cylindrical over the full length and the shaft (diameter 19.98mm) has no waist either. Maybe your oil has even lower viscosity than mine and my bearing gap is 25% smaller that yours. So, the shearing area is much bigger and the torque is increased by the shaft radius as well.
I mentioned before that higher oil viscosity won't work in my bearing. I suspect that the oil i use is (marginally) happy in a gap of 10µm but not in a gap of 9µm. And i remember that the relation between gap diameter and shearing force is highly progressive.
25ys when i designed and built my TT, my initial idea was to have grease instead of ball and trust plate under the shaft and to have thin oil in the bearing gap
rolleyes: was an apprentice back then, not an engineer). That the thick grease would mix up with the thin oil at their borderline and that this mixture had viscosity high enough to make the bearing jam completely, I did not forecast and how severe this effect was came to me as complete surprise. Over several hours, the bearing would go smooth at first and the motor would have to work harder and harder and would finally stall and to revolve the platter would require manual force.When i tested my new motor assembly and motor controller, I tried this out again. I put thicker oil (still thin) into the bearing (yielding in the platter coming to rest after less than 2 revolutions) to observe if the motor was still capable to drive it. It was, but was audibly strained, and the belt (a spliced leader tape) would crack at the splicing within a few minutes.
Still no grinding/screetching noises from the bearing so the oil film was not yet disrupted, but the thicker oil acted as brake rather than as lubrication.
But let me repeat:
smooth bearing drag is GOOD! Bernhard,
You've found the same as me.
Viscous drag evens out speed stability.
Drag swamps stylus drag effects.
Added motor load linearises the motor.
Silicon oil isn't a great lubricant- that's ok in my polymer bearing....
The one thing I really love about viscous drag is that it's self correcting. If the motor should increase its speed the increased drag holds it back. If the motor drops it's speed the reduced drag helps even it out. I think it might be the only free lunch in the whole of audio.....!
My deck comes to rest from 33 in about 5 revolutions.
You've found the same as me.
Viscous drag evens out speed stability.
Drag swamps stylus drag effects.
Added motor load linearises the motor.
Silicon oil isn't a great lubricant- that's ok in my polymer bearing....
The one thing I really love about viscous drag is that it's self correcting. If the motor should increase its speed the increased drag holds it back. If the motor drops it's speed the reduced drag helps even it out. I think it might be the only free lunch in the whole of audio.....!
My deck comes to rest from 33 in about 5 revolutions.
Last edited:
Congrats.
FYI (& statistics).
Mine (DIY air bearing TT) do it in:
Free run from 33,3 rpm to 0 rpm: total time 11minutes 45 seconds.
The "last" second is amazing ...
Karel
My toy:
It's a second second home made heavy model.
Some caractéristiques:
AIR BEARING TURNTABLE 16”
Concept/technical specifications:
Type: air bearing spinning platter
Nozzels: multiple air nozzels (home made)
Dynamic mass: 12kg
Turntable platter: C45 / Dia: 325mmx14,4mm
Motor management: “à la Lenco”, with micro fine tuning Turntable speed (fix): 33,3 RPM
Turntable: 16” Glasplatter 450mmx8mm
Stroboscope : 16” (à la EMT … custom home made )
Center: three points centering (near absolute center)
Air compressor: oil free air compressor
Air pressure turntable: 0,7bar (minimumminimorum)
Total weight: 45kg (appr)
Free run from 33,3 rpm to 0 rpm: total time 11minutes 45 seconds
AIR BEARING LINEAR TRACKING TONE ARM 16”
Concept/technical specifications (home made):
Type: air bearing linear tracker (à la “Terminator”). Effective lenght: 450mm (16”)
Tracking error: 0°
Pressure tone arm air pressure: 0,4bar (appr).
.... It's music .... i like it ....
Under the skirt (!. Sorry for this expression):
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.