DIY linear tonearm

nocdplz,

Bird Precision

Very good company with a large variety of of screw mounted jewel bearings. This is where I bought my bearings. The bearing are very light in weight so shipping should be fairly cheap.

Niffy,

I wound up buying about 100 jewel bearings (unmounted) on ebay. These bearings were used to repair Russian ship clocks !!!! They are about .111" OD small hole through center---If anybody needs, I will give some.
 
Niffy,

If possible and get some time, could you post some pictures of the bearings you are trying to improve ? I have a hard time picturing your turntable drive system. good luck with your research.


Hi Joe,

I hope these clarify my description.

IMG_20190220_162950-1280x960.jpg

IMG_20190220_163120-1280x960.jpg

IMG_20190220_163211-1280x960.jpg

The belt takes a 90° turn around the motor pulley and each idler wheel, touching the subplatter for 30° in between each. The photos are of the new idlers. The bearings are mounted in M4 grub screws. Any new bearings will need to also need to be M4 based which give a maximum diameter of about 2.5mm. I have a couple of ideas for how to get the bearings silent, just waiting on some bits to turn up in the post.

I have considered jeweled bearings of the type I linked. The problems are how to secure them within the grub screws (epoxy?) and what to use for the pivot/axle. Possibly round ended needle roller bearing. The smallest size readily available is 1mm diameter which would require that the hole in the jewel to be this size. Going much smaller in pivot size would make drilling the holes in the ends of the axles tricky.

Niffy
 
Hi Jim,

The idlers in the photos are actually the fourth version I've tried. The first set I made, with Teflon bushes, remains the best. The pin bearings come in a close second. My second attempt did use ballrace bearings. These were a high quality miniature type, the best I could find. They were horrible. Really noisy.

To be honest I'm being terribly OCD. It's the one area of the deck that I feel could be improved. It's more about reducing platter runout than improving the idlers themselves. If my current plan fails I'll just make some bearings the same as the originals and fit the new idler wheels to them. I will get the same very quiet bearing with the improved concentricity of the new wheels. I'll also be able to run the different belt setup with the new wheels.

Niffy
 
Niffy,

You put an awful lot of time and effort in your beautiful turntable. A picture is worth a 1,000 words.

You can definitely use epoxy to secure the jewels to any size grub screw you like. However (my opinion) you might be better off using piece of carbide rod that you can polish both ends and put into a U shaped, cup type bearing of Vespel or Torlon. You could drill and ream a hole through the length of yor pulley material and glue or force fit the rod in a slightly under sized hole in the pulley body.

I made my own carbise axles ( 1/16" ) using a Foredom tool running up against a diamond wheel ( on my surface grinder) at an 80 degree angle. These axles are polished on both ends and go through my magnesium carriage wheels, then enter the jewel bearings. I believe this way would produce the most robust and quiet bearing for your purpose but perhaps not as quite as jeweled bearings. Another possible advantage is that you could fill the bottom cup bearings with nano / synthetic clock oil. Hope this makes sense to you. Good luck.
 
Niffy,

I forgot to mention that you put a radius and polish both ends of the rod / axle using diamond stones (ebay) and then various grades of diamond polishing compounds. There is also the advantage of using "normal" sized materials, 2mm, 3mm, 4mm diameter carbide rods instead of micro / jewelers size components. Small carbide drill blank rods are available on ebay.
 
Hi Joe,

The previous bearings are very similar to what you describe. The shart is stainless steel rather than tungsten carbide. The extra hardness of carbide doesn't make that much difference when used against a softer plastic. The cup/bush in made of Teflon.

For the next attempt my plan is to use round ended needles from roller bearings, 1.5mm diameter, as the shafts. These will be set in a short M3 threaded axle. The shafts will fit into cups inserted into the ends of the fixed grub screws. The cups will be made of a short lengths of Teflon tube, 2.5mm od 1.5mm id, with 1/16" delrin balls at the bottom. These will be lubricated with petroleum jelly.
The wheels are acrylic and adjustable on the axle to allow them to be centered. Not have a precision lathe requires some lateral thinking. I have built a pair of jigs, one to measure how off centre the wheel is and one to then centre it. This bit of the new idlers works superbly.

Thanks for the input.

Niffy
 
Niffy,

As usual, you put a lot of time and thought into improving your turntables drive system performance. I am sure you will perfect whatever method you choose. I speak like everybody has a machine shop in their basement, like I do. Not having one, forces you to use off the shelf materials and think outside the box, in which you have done a stellar job.I could only imagine the work you would turn out if you had access to a lathe / mill etc. !!!! I will be watching, learning and contributing where I can.
 
Hi all,

Question:in order to have the right mass for the TA to match the compliance of he cartridge, itis probably needed to increase the weight- does it matter where to put this weight?

Best,
Coolerooney

Hi Coolerooney,

When building your carriage there is only so much mass of material that you can use. It is best to design your carriage for the cartridge you intend to use. Doing this allows you to determine the ideal mass for your carriage. Think of this as your mass budget. You spend a certain amount of this budget on each component of the carriage, eg the headshell, the bearings, the counterweight. You want to spend the mass budget as wisely as possible, the aim being to create a carriage that is as rigid as possible with the highest bending mode resonance. Adding weighs to tune the mass of the arm to the compliance of the cartridge is a waste of part of the budget. Design the arm so that it has the correct mass in the first place.

Niffy
 
Hi C., are you asking for linear trackers? for the Vertical or the Horizontal Eff. mass?
For the Horizontal one, usually there is the opposite problem; Since it simply coincides with the whole weight of the mobile masses (cartridge + head shell + shaft + carriage + counterweight) it is rather difficult to stay well under 25 - 30 grams, at the limit of a modern - high mass-low compliance - TA.
For the vertical one it is easy to properly increase all the weights of the moving masses, as Niffy suggests. Too bad that this will further increase the Horizontal Eff. mass.
To bypass the problem someone has simply moved the counterweight away from pivot: this reduces the weight needed (and so the Hor. mass), but increases the Vert. eff mass.
If I had to say that this convinces me, I would say the opposite of what I think, and always seen on any properly designed TA

carlo
 
Last edited:
Nailed it!

Hi all.

I've just spent today finishing off and testing my new idler wheels, bearings and belt configuration.

The old belt setup had a single belt that passed around the motor, both idlers and the sub-platter. With off balance records there was a slight wobble (runout) to the platter.
The new belt setup has three separate belts. One passes around the motor and the sub-platter. Each idler has its own belt that passes around itself and the sub-platter. The platter now remains level on all records (no runout). I placed a probe just above edge of the record and viewed the gap between the record and this probe using high magnification. I could not see any variation in the gap. This also means that my clamp is doing a fine job of flattening warps. I thought that fitting one belt was a PITA. Fitting three that all take different paths had me inventing several entirely new swear words. I finally found a method utilising bits of string and sticky tape.

I've also been swapping between the different bearings in the idler wheels. After much toing and frowing I have settled on the pin bearings as having the best sound. After a good run in they are much quieter than they were. The difference in sound between the different bearings is VERY slight.

The main reason for this tweek was to improve the runout of the platter which I have achieved. As my tonearm is very short it will be more susceptible to height differences in record surface than most arms. Variations in the height of the record will effect VTA errors . It will also cause something akin to warp wow. By reducing runout I should improve in both these areas.
It's different to be certain but I think that the deck now has a slightly better sense of timing than before.

Niffy
 
Congratulations Niffy, more and more sophisticated!
I'm almost tempted to put my hands back on my TT project, abandoned long ago with growing frustrations: so I would like to understand better.
In order to that, I've tried a scheme with my interpretations taken from your previous descriptions.

The old version had a single bearing, a sort of unipivot with a small diameter ball rotating on a conical seat, both made with the hardest materials. This avoids the problems of traditional or inversed bearings (noise, friction etc.) but leaves the platter free to tilt on three axes above the pivot. For sure this is limited by a relevant mass and a center of gravity much lower than the pivot.
Then there were three pulleys (1 motor + 2 idler wheels) with a single belt (flat, maybe) probably in axis with the unipivot, that stabilized the platter on three points, preventing the possible wobbling. The different traction on the three sides, due to the traction of the motor, could however allow some small residual wobbling.
The new one is the same in everything except the use of three separate belts to avoid wobbling. The motor traction is now balanced by the other two separate belts on idler wheels.

However here are my doubts: the three belts can not be co-planar, so the tractions generate different momentums on the pivot. (upper - aligned - lower)
How is it possible to have a correct and stable regulation of the three belts over time? (pulley shafts mounted on eccentric spring bases?)

carlo
 

Attachments

  • TT SCHEME.jpg
    TT SCHEME.jpg
    89.3 KB · Views: 405
Hi Carlo, Hiten,

One of my original designs had the belt run around the periphery of the platter level with the contact point of the bearing (gyrodeck style) . This setup offered very little stability, the platter would precess and would require some form of sleeve to stabilise it. This is because a large tilt of the platter will have very little effect on belt tension so there is little to prevent the platter from tilting.

In my design I have the bearing contact point as close to the surface of the record as possible and the center of mass of the platter as far below this as possible. The belt/belts are also positioned as far below the pivot as practically possible. In this way a small tilt of the platter will result in a larger change in belt tension. This also has the advantage of allowing a smaller sub-platter to be used than running the belts around the periphery of the platter, which in turn allows for a smaller sub-chassis as the idlers are mounted on the sub-chassis. In general the smaller the sub-chassis the better its resonance characteristics. Also a smaller sub-platter allows for shorter belts which gives the motor more direct control.

With a single belt the sub-platter can move around a bit without actually stretching the belt all that much. With three belts the same amount of movement of the sub-platter will result in a much larger amount of stretch in at least one of the belts. Although the belts do have to run at different heights they are still quite close to each other and a long way below the pivot. The aim of this approach to the way the bearing and drive systems interact was to be able to dispense with the need for any form of sleeve to the bearing. (a three internal contact point nylon washer is still a sleeve of sorts). The sleeve seems to be the part of the bearings that causes the biggest problems, has the largest negative impact on sound quality and is the most difficult and expensive to manufacture. The primary role of the sleeve is to keep the platter centred and level. The small diameter ball running in the tiny pit that my bearing uses keeps the platter well centered. The belt configuration and dynamic stability of an inverted bearing keep the platter level. In combination these points eliminate the need for any bearing sleeve.
Even with the old single belt setup the runout on the worst off balance record was still mild. Most platter mats, especially felt ones, will introduce more height variation than this system did. The improvement going to the three belt system is only slight as the maximum level of improvement possible was slight. It's very much a case of me being a perfectionist and having a slightest imperfection bugging me.

I got the idea for eliminating the sleeve from the days when I owned a pink triangle anniversary deck. This used an inverted bearing and a Teflon washer in very much the same way as Hiten suggested. I found that removing the washer gave a large improvement in sound quality even though the platter wobbled all over the place. The negative effects of this simple sleeve were still greater than a couple of millimetres of runout. Don't get me wrong, the anniversary is still one of the greatest decks ever made.

Attached is a photo showing the new belts.

IMG_20190227_120124_crop_372x318.jpg

Niffy