Maybe there were more errors in all that
(*). Thank you for the correction.It is not alum alloy and I should have trusted the ‘signs’ (die-casting fine surface texture and feel of weight).
I have seen plastic sub platter only on TD-166. On all TD-160 I have seen, main and sub platter are metallic. Mine too (TD160 MkII, platter casting year 1978)
I measured the volume of the main platter. It is 300mL +/-5mL.
I weighed it. It weights 2.040kg+/-10g. Thus the alloy density is somewhere btn 6.650 and 6.950 kg/m^3 as you say.
George
(*) jacco when you’ll get high, have a look at the numbers in that 3rd revision of calculations. Do they seem OK to you?
Attachments
Yep this is caused by inadequate power supplies for the opamps in the active XO section of the system. At least in my various active speakers when I fix this issue, the HF sounds return to the place where they should be coming from, which is the same place the bass comes from.
Generally PSUs in such speakers are generated by LM7815/7915s which means the LF performance is fairly decent but HF impedance sucks. They're also generally rather distant from the opamps they're feeding, meaning series inductance rules. HF noise is created on the rails because the opamps run in classB and the local decoupling is inadequate. Opamps have sucky PSRR at HF too, and classB output stages generate noise for every switching transition meaning its 'tipped up' towards the HF.
Cleaning up the rails for the opamps is only the first step though, once this layer of the onion is peeled off, the issue of inadequate decoupling for the chipamps rears its head....
Have you tried biasing the op amp outputs with a current source or resistor to the negative rail?
Good point - yes that is indeed part of the solution to reducing the rail noise and makes a substantial difference. I forgot to mention it I also scale up the working impedances used in the active filters to reduce the loading on the OPS. JFET opamps can work with source impedances of 100's of k without breaking a sweat, this also results in physically smaller and cheaper caps.
I also scale up the working impedances used in the active filters to reduce the loading on the OPS. JFET opamps can work with source impedances of 100's of k without breaking a sweat, this also results in physically smaller and cheaper caps.
Good point - yes that is indeed part of the solution to reducing the rail noise and makes a substantial difference. I forgot to mention it
Yes, going to the source of the problem is usually the most effective solution.
Other serious companies of the day also used aging. For example, all reVox/Studer machines of the 70ies had their capstans aged, they felt it was necessasry because it would have to take God knows how many thumps from the pinch roller coming on line.
That's a good tip. However, sometimes it's not goping to help much, but other times, it can really transform the sound. Definitely worth a try, though.
Another point is adding a simple two transistor current booster. That works with each and every op amp I ever tried, although of course not to the same extent. It is most noticable in the bass region, where op amps can sometimes be thin and unconvincing.
You can even extend it to a pair of drivers and another pair of medium power transistors, such as say BD139/149, 2SC3503/2SA1381, etc. Perfect for driving low impedance loads, headphones, etc. If you're willing to accept the PSU line limitations, even a power amp.
There are so many ways of getting a system into a zone of "good sound' - I wouldn't be fussed about how it was done. Of course, if you do each and every refinement that's possible, , then amazing sound will be achieved, totally "blow your socks off" reproduction will occur - I've had glimpses of this at times when enough of the stars have aligned, , but it takes quite a bit of focused effort to pull it off ... not enough completely co-ordinated knowledge in place yet to guarantee such a result every time, but slowly getting there ...
At the moment I still go with the "worst" recording approach; play the track that exaggerates every remaining deficiency of the system, in the worst light possible - and use that to guide me to the next "job" to do. Listening for audible defects in the sound is an extremely powerful tool, hasn't let me down in 30 years,.
, then amazing sound will be achieved, totally "blow your socks off" reproduction will occur - I've had glimpses of this at times when enough of the stars have aligned, , but it takes quite a bit of focused effort to pull it off ... not enough completely co-ordinated knowledge in place yet to guarantee such a result every time, but slowly getting there ... At the moment I still go with the "worst" recording approach; play the track that exaggerates every remaining deficiency of the system, in the worst light possible - and use that to guide me to the next "job" to do. Listening for audible defects in the sound is an extremely powerful tool, hasn't let me down in 30 years,
.I have been thinking about the TD145 and how it was transformed by a new belt. Before that a very cheap direct drive beat it. Specifically Frank Sinatra was wrong. No air around the voice and distortion of the recording bleached out. I only knew this as my audio memory said no way was this right. There were so many other things that were "Audiophile" wrong about the TD145. If an Audiophile I would have spent big money to correct them first.
I conclude from this we mostly listen to the zero and maximum parts of a note or wave. Even when complex they have these things. What of digital ? The maximum and minimum is randomly found. This can only be solved by more samples? Then a further complication to playback reasonably as recorded. Obvious ? Perhaps. Often said ? I don't think so.
Bias on a tape recorder is the same problem. I suspect it is easier to get it right. It needs no decoding. Trust me the turntable effects are not subtle. Why should digital be easier? My Nakamichi BX 100 has a worn out drive idler. It has a very bad sound although not easy to say why ( I have tried to get the idler soft again ). The BX 100 on a good day is a perfectly good hi fi source. I never used Dolby as it was acceptable without. About like FM . For that the tapes needed pushing . Far better a bit of 1950's sound 3 % of the time than hiss 90% of the time. Dolby is very good on the BX100. If I had only ever tried it's version I would have been happy. Without Dolby suited my other tape decks better . The BX100 for master copies. I have used the Dragon, it is like the LP12 and TD145 comparison. The BX 100 is a top grade deck albeit very simple. Nakamichi used vibrational analysis as the advantage over competitors. This means they made more money because they thought things out. Yamaha ran them very close at half the price. I used to align Yamaha free of charge and still match London prices. It took me 5 minutes. Yamaha adopted my methods I am told!! It became a bit of a legend. I made a lot of money out of Yamaha tape decks. As much as LP12's. That is absolutely true. It came up in analysis of the accounts. Yamaha were bemused as we were a significant account that sold 90% cassette decks.
I can not agree that a stiff base board to an LP12 is better. What about the air from the speakers trapped inside the box. Also what is so bad about a 1 inch piece of solid wood around the LP12 ( and TD160 S ) that the base will brace? Come on, be an engineer. The LP12 isolates to about - 70dB except the air that hits the sub-chassis and arm tube. I will agree that if wanting to meet safety regulations a nice piece of 6 mm Tufnol is ideal ( paper phenolic ). It might well be better at > - 70 dB.
My LP12 was equipped with the Lingo PSU, sold off to raise some cash. Although a very fancy piece of engineering it seems by my measurements mostly to be a 90 V start and 66 V run device . This agrees with my ears. The LP12 I am using is a bit 1950's for want of a better way of saying it. Seems like second harmonic distortion. The TD 145 might be better on that. I doubt I would go to 66V. 75 V perhaps ? On the Lingo prototype it had a voltage control. All at Linn took it home that had LP12's. The Lingo as sold had the average setting of the staff. When I said why couldn't I have a control they said that would show lack of confidence on the part of Linn and that might be a very bad thing.
The Keel sub-chassis kit or whatever it is called. If a sheet of aluminium and some plywood of no greater mass than the LP12 chassis was made it might be a better chassis. Simple Q max hole punch, hack saw and file are the tools required. The original chassis and cutters should make centering the tools to get a perfect clone very easy. The master plan is to use boat grade epoxy to bond the pieces. Make the holes in the ply generous to allow adjustments to springs. An afternoons work? The idea is the ply not only controls resonance but does not have a high Q when bonded.
The weirdest thing about the LP12 is it sounds like an electronic noise gate is working. This fits well with Linn's beliefs. The downside is I feel slightly robbed of trailing edge detail. It is a better sound and that's all there is to say with a slightly wrong balance of priorities. My Naim Aro arm that seems to prefer living in Germany solved that. LP12 sounds like a Revox tape deck loop is in use with meters peaking. If I was to be critical it's like having too much sex. Sorry guys that 's the best way I can say it. And of all the bad things in life that's one I can cope with. Certainly messes up going shopping as does the LP12. The analogy is the LP12 is a pure minx and the TD145 sometimes likes a chat about art thrown in.
About TP 16. If you don't calculate the resonant point you will never know the arm at all. The TP16 would have had to work with Shure V15 mk 3 in it's day . It was joked that some cartridges of the time were needing - 5 g effective mass if trying to meet the supposed ideal of 12 Hz resonance. Load the TP 16 with enough mass and all is transformed. The Schroeder pick up arm is respected as a reference device much as the Breuer. I had a carbon-fiber one to try. It was about the most diabolical sound I every heard from a top grade product ( with Lyra Helikon I think ). Totally unhappy I mass loaded the arm. Transformed is an understatement of some magnitude. Perfection would be better. I had a very bad prejudice against carbon-fiber up to this point. It was simply the mass. I believe this Schroeder isn't made now? Shame as I feel it was valid. If you have a 1960's Pritchard ADC it might be a good match. Infinity Black Widow was the one people choose. Very high compliance cartridges work well with direct drives as they interfere less with the servos. Worth a thought.
I conclude from this we mostly listen to the zero and maximum parts of a note or wave. Even when complex they have these things. What of digital ? The maximum and minimum is randomly found. This can only be solved by more samples? Then a further complication to playback reasonably as recorded. Obvious ? Perhaps. Often said ? I don't think so.
Bias on a tape recorder is the same problem. I suspect it is easier to get it right. It needs no decoding. Trust me the turntable effects are not subtle. Why should digital be easier? My Nakamichi BX 100 has a worn out drive idler. It has a very bad sound although not easy to say why ( I have tried to get the idler soft again ). The BX 100 on a good day is a perfectly good hi fi source. I never used Dolby as it was acceptable without. About like FM . For that the tapes needed pushing . Far better a bit of 1950's sound 3 % of the time than hiss 90% of the time. Dolby is very good on the BX100. If I had only ever tried it's version I would have been happy. Without Dolby suited my other tape decks better . The BX100 for master copies. I have used the Dragon, it is like the LP12 and TD145 comparison. The BX 100 is a top grade deck albeit very simple. Nakamichi used vibrational analysis as the advantage over competitors. This means they made more money because they thought things out. Yamaha ran them very close at half the price. I used to align Yamaha free of charge and still match London prices. It took me 5 minutes. Yamaha adopted my methods I am told!! It became a bit of a legend. I made a lot of money out of Yamaha tape decks. As much as LP12's. That is absolutely true. It came up in analysis of the accounts. Yamaha were bemused as we were a significant account that sold 90% cassette decks.
I can not agree that a stiff base board to an LP12 is better. What about the air from the speakers trapped inside the box. Also what is so bad about a 1 inch piece of solid wood around the LP12 ( and TD160 S ) that the base will brace? Come on, be an engineer. The LP12 isolates to about - 70dB except the air that hits the sub-chassis and arm tube. I will agree that if wanting to meet safety regulations a nice piece of 6 mm Tufnol is ideal ( paper phenolic ). It might well be better at > - 70 dB.
My LP12 was equipped with the Lingo PSU, sold off to raise some cash. Although a very fancy piece of engineering it seems by my measurements mostly to be a 90 V start and 66 V run device . This agrees with my ears. The LP12 I am using is a bit 1950's for want of a better way of saying it. Seems like second harmonic distortion. The TD 145 might be better on that. I doubt I would go to 66V. 75 V perhaps ? On the Lingo prototype it had a voltage control. All at Linn took it home that had LP12's. The Lingo as sold had the average setting of the staff. When I said why couldn't I have a control they said that would show lack of confidence on the part of Linn and that might be a very bad thing.
The Keel sub-chassis kit or whatever it is called. If a sheet of aluminium and some plywood of no greater mass than the LP12 chassis was made it might be a better chassis. Simple Q max hole punch, hack saw and file are the tools required. The original chassis and cutters should make centering the tools to get a perfect clone very easy. The master plan is to use boat grade epoxy to bond the pieces. Make the holes in the ply generous to allow adjustments to springs. An afternoons work? The idea is the ply not only controls resonance but does not have a high Q when bonded.
The weirdest thing about the LP12 is it sounds like an electronic noise gate is working. This fits well with Linn's beliefs. The downside is I feel slightly robbed of trailing edge detail. It is a better sound and that's all there is to say with a slightly wrong balance of priorities. My Naim Aro arm that seems to prefer living in Germany solved that. LP12 sounds like a Revox tape deck loop is in use with meters peaking. If I was to be critical it's like having too much sex. Sorry guys that 's the best way I can say it. And of all the bad things in life that's one I can cope with. Certainly messes up going shopping as does the LP12. The analogy is the LP12 is a pure minx and the TD145 sometimes likes a chat about art thrown in.
About TP 16. If you don't calculate the resonant point you will never know the arm at all. The TP16 would have had to work with Shure V15 mk 3 in it's day . It was joked that some cartridges of the time were needing - 5 g effective mass if trying to meet the supposed ideal of 12 Hz resonance. Load the TP 16 with enough mass and all is transformed. The Schroeder pick up arm is respected as a reference device much as the Breuer. I had a carbon-fiber one to try. It was about the most diabolical sound I every heard from a top grade product ( with Lyra Helikon I think ). Totally unhappy I mass loaded the arm. Transformed is an understatement of some magnitude. Perfection would be better. I had a very bad prejudice against carbon-fiber up to this point. It was simply the mass. I believe this Schroeder isn't made now? Shame as I feel it was valid. If you have a 1960's Pritchard ADC it might be a good match. Infinity Black Widow was the one people choose. Very high compliance cartridges work well with direct drives as they interfere less with the servos. Worth a thought.
Sometimes it is worse. Always listen. Resistor can work. I used two op amps once. One to the positive and one to the negative and outputs summed. I told myself it was better. Complimentary errors. The next stage was 75 uS. It would have a little bit of statistical noise cancelling thrown in. The NPN idea was from LM741 . It seems to have become folk law. I can not remember if the op amp was NPN+ PNP. The load was 1 mA and the voltage swing gain of 62 with about 500 uV ( x 62 ) in 1 kHz and 5 mV ( x 62 ) 20 Hz , 50 uV 20 kHz ( x 62 ).
I think that we should try to make some sense out of turntable differences. There are technical factors, as well as subjective ones.
For example, why did Thorens and Linn forge ahead of the pack, back in the late '60's and '70's? This compared to the earlier AR, Grado, Rekokut (sp), most record changers, etc?
One first scientific measurement was introduced in the late '60's where the inertia of the turntable became most important. This put the Thorens TD150 ahead of the AR that was popular at the time. Higher inertia meant reduction in DYNAMIC NEEDLE DRAG, caused by changes in the program material. This is very hard to measure directly, but it seemed to make sense.
Later, the Linn had even more inertia, a better bearing and an improved suspension. This put it above most 'affordable' turntables at the time.
The counter problem is the introduction of RUMBLE, something that I have not heard for some time, but my first belt drive turntable had plenty! Apparently, the bigger the drive motor, the more trouble with rumble, all else being equal, so the turntables of the '50's had relatively large (and impressive) drive motors, and rumble seemed to be part of the equation.
I think it was AR who first thought to use 'clock motors' rather than larger AC motors, in order to reduce the rumble. This would appear to be why Linn dropped the drive voltage after getting the turntable started. I can see why Nigel found this problematic, IF the inertia of the table is not truly adequate.
For example, why did Thorens and Linn forge ahead of the pack, back in the late '60's and '70's? This compared to the earlier AR, Grado, Rekokut (sp), most record changers, etc?
One first scientific measurement was introduced in the late '60's where the inertia of the turntable became most important. This put the Thorens TD150 ahead of the AR that was popular at the time. Higher inertia meant reduction in DYNAMIC NEEDLE DRAG, caused by changes in the program material. This is very hard to measure directly, but it seemed to make sense.
Later, the Linn had even more inertia, a better bearing and an improved suspension. This put it above most 'affordable' turntables at the time.
The counter problem is the introduction of RUMBLE, something that I have not heard for some time, but my first belt drive turntable had plenty! Apparently, the bigger the drive motor, the more trouble with rumble, all else being equal, so the turntables of the '50's had relatively large (and impressive) drive motors, and rumble seemed to be part of the equation.
I think it was AR who first thought to use 'clock motors' rather than larger AC motors, in order to reduce the rumble. This would appear to be why Linn dropped the drive voltage after getting the turntable started. I can see why Nigel found this problematic, IF the inertia of the table is not truly adequate.
Last edited:
Iterations, I total : 0.03653 => 0.05304 => 0.01820
The main and sub platter both have a thickness of 4mm.
Keeping it simple :
-top disc of 300mm diameter and 4mm thickness : 1.8944 kg
-outer rim of 16,5 mm height and 7.85mm thickness : 0.7965 kg
-inner rim of 24mm height and 5mm thickness : 0.3915 kg
(total of 3.0824 kg, without the top disc added, or platter holes subtracted)
Moment of inertia top of platter : 1/2 x mass x (rm2)2
1/2 x 1.8944 x 0.15 x 0.15 = 0.021312 kgm2
Outer rim : 1/2 x m x [(rm2)2 + (rm1)2]
1/2 x 0.7965 x [(0.15)2 + (0.14215)2] = 0.016997 kgm2
(is fast & easy approximation, 1/2 x pi x rho x height x [(rm2)4 - (rm1)4] will be off by a very small amount for such a small rim thickness/diameter ratio)
Inner rim : 1/2 x m x [(rs2)2 + (rs1)2]
1/2 x 0.3915 x [(0.08)2 + (0.075)2] = 0.002354 kgm2
I tot : 0.04 kgm2 (plus something tiny)
(All-day birthday celebration, brunch, dinner, just home ten minutes, Now, let's see, where did I leave the amyl-nitrite and the XTC ?)
Last edited:
One of the most impressive features of the Linn is the use of a felt mat, instead of rubber or plastic, and a truly grounded turntable body. This removes all kinds of static that everybody complains about. The problem that it might cause, however, is a decoupling from the turntable due to its physical characteristics, being felt and cloth-like. Still, I find it successful in making many records easily playable without too much pre-cleaning, etc.
Taking into account what John just posted on rumble, I'd like to be told how is all this belt driving actully superior to direct drive from a Hall generator motor of decent power, such as the one used for example on Dual 701.
It's been decades, but if memory serves, the key advanatge of a Hall generator motor is that it has practically speaking an endless or totally continuous magnetic field, which assures contant torque and rotation speed such as no other combination can.
I realize that the question is theoretical as it disregards specific, concrete implementation, and we all know there are many ways to go wrong, hence sometimes large differences in performance of similarily constructed TTs,
It's been decades, but if memory serves, the key advanatge of a Hall generator motor is that it has practically speaking an endless or totally continuous magnetic field, which assures contant torque and rotation speed such as no other combination can.
I realize that the question is theoretical as it disregards specific, concrete implementation, and we all know there are many ways to go wrong, hence sometimes large differences in performance of similarily constructed TTs,
I think that the bigger issue is the acoustic impedance mismatch between the vinyl record and the felt mat- any acoustic wave propagated through the vinyl is reflected back with higher efficiency than when the platter has a matching impedance (and this assumes that the record is clamped down). Acrylic is a pretty good match for vinyl.
There probably are pretty good motors that are suitable for direct drive operation, but the belt acts like a low pass filter, I am pretty sure.
Back in 1970, having spent the previous 2 years designing servos at Ampex, I too thought that the direct drive would be best, but apparently I was wrong at the time. Direct drive worked great as an analog tape capstan motor, but the mechanical vibration from the motor itself appeared to be the problem. No servo can take that out. The Gale turntable was also direct drive, and it had lots of problems.
Back in 1970, having spent the previous 2 years designing servos at Ampex, I too thought that the direct drive would be best, but apparently I was wrong at the time. Direct drive worked great as an analog tape capstan motor, but the mechanical vibration from the motor itself appeared to be the problem. No servo can take that out. The Gale turntable was also direct drive, and it had lots of problems.
- Status
- Not open for further replies.