to Cobra2
to Cobra2:
I'd like to ask you a few questions regarding the Adyton amplifier you're so fond of so contact me, please, at:
sonaria@post.pl
skype: sonaria
+44-7725-124711 (also sms / text)
thank you very much in advance... talk to you soon !
Yours
Tomasz
to Cobra2:
I'd like to ask you a few questions regarding the Adyton amplifier you're so fond of so contact me, please, at:
sonaria@post.pl
skype: sonaria
+44-7725-124711 (also sms / text)
thank you very much in advance... talk to you soon !
Yours
Tomasz
Further to above post. Just read in the Technics Sp-10 Mark III brochure (found at www.vinylengine.com) :
"The 10kg platter both reduces wow and flutter and improves rumble"
and another section talking as a plus: "besides the huge moment of inertia of the heavy platter itself"
And in another area of the brochure: "1.1 tonne/ cm squared moment of inertia. Huge moment of inertia results from the 10kg platter made of diecast aluminium with a 15mm thick copper alloy surface layer. Therefore there is no change in speed with any fixed load up to 10kg .cm - equivalent to 1000 tonearms tracking at 2gm each. Of course the real benefit is that the platter keeps rotating at exactly rated speed even if you accidently touch it during play".
This suggests that a heavier than stock platter might be a good thing and might aid in speed stability.
A wooden platter might raise the height of the platter sufficiently that the magnet rfi/emi interference is that little bit further away from the cartridge and may also give the sp-10 a little more 'wetness' and tone that some people say is its main weakness. A wooden platter also would be pretty cheap to make and experiment with.
"The 10kg platter both reduces wow and flutter and improves rumble"
and another section talking as a plus: "besides the huge moment of inertia of the heavy platter itself"
And in another area of the brochure: "1.1 tonne/ cm squared moment of inertia. Huge moment of inertia results from the 10kg platter made of diecast aluminium with a 15mm thick copper alloy surface layer. Therefore there is no change in speed with any fixed load up to 10kg .cm - equivalent to 1000 tonearms tracking at 2gm each. Of course the real benefit is that the platter keeps rotating at exactly rated speed even if you accidently touch it during play".
This suggests that a heavier than stock platter might be a good thing and might aid in speed stability.
A wooden platter might raise the height of the platter sufficiently that the magnet rfi/emi interference is that little bit further away from the cartridge and may also give the sp-10 a little more 'wetness' and tone that some people say is its main weakness. A wooden platter also would be pretty cheap to make and experiment with.
What you are not understanding is that when the SP-10 electronics compensate for speed variation, they make the exact adjustment necessary to correct the fault based on the known value of inertia of the platter specific to the system. In other words, a different platter with the wrong inertia will result in the onboard "computer" making an incorrect adjustment. I'm sure that Panasonic redesigned the computer on the MarkIII so that it is able to keep up correctly with the heavier platter. If you put a heavy platter on an SP10 MarkII you can bypass the compensation circuit and adjust the speed manually.
As a side note, the "huge" 1.1 tonne/cm^2 inertia of the MarkIII platter pales next to the 26 tons/cm^2 value of the Micro SZ-1 motor-platter combination.
John
As a side note, the "huge" 1.1 tonne/cm^2 inertia of the MarkIII platter pales next to the 26 tons/cm^2 value of the Micro SZ-1 motor-platter combination.
John
<blockquote>they make the exact adjustment necessary to correct the fault based on the known value of inertia of the platter specific to the system</blockquote>
That's close but I think it can be explained thus:
A servo loop is a feedback system and like all such it must adhere to Nyquist's criterion for loop stability: the feedback must fall below unity before the phase shift in the loop reaches 180 degrees.
If you increase the moment of inertia of the platter from its present value of 0.1 kg m^2 (let's have no more of this ridiculous "tonne cm^2" when there's a perfectly respectable SI unit at hand) you increase the phase shift, as the inertia is the equivalent of inductance.
If you were to increase the inertia far enough you'd end up with the loop still feeding back when the phase shift has reached 180 degrees so your feedback is now positive and oscillation results.
The solution is obvious- slow down the servo loop to match the increased inertia. Better still, remove the servo loop and run the motor with a fixed frequency three phase drive with waveform compensation to remove the cogging torque.
That's close but I think it can be explained thus:
A servo loop is a feedback system and like all such it must adhere to Nyquist's criterion for loop stability: the feedback must fall below unity before the phase shift in the loop reaches 180 degrees.
If you increase the moment of inertia of the platter from its present value of 0.1 kg m^2 (let's have no more of this ridiculous "tonne cm^2" when there's a perfectly respectable SI unit at hand) you increase the phase shift, as the inertia is the equivalent of inductance.
If you were to increase the inertia far enough you'd end up with the loop still feeding back when the phase shift has reached 180 degrees so your feedback is now positive and oscillation results.
The solution is obvious- slow down the servo loop to match the increased inertia. Better still, remove the servo loop and run the motor with a fixed frequency three phase drive with waveform compensation to remove the cogging torque.
And speed in furlongs per fortnight?
My favourite unit is the mBTU which I always think of as the energy required to make a nice cup of tea*.
I trained as an engineer when Australia was just changing over to metric so we dealt with both imperial and SI units. Thermodynamics is sooo much easier in SI.
* yes I know it would be a very big cup. It's a joke.
My favourite unit is the mBTU which I always think of as the energy required to make a nice cup of tea*.
I trained as an engineer when Australia was just changing over to metric so we dealt with both imperial and SI units. Thermodynamics is sooo much easier in SI.
* yes I know it would be a very big cup. It's a joke.
Mark Kelly wrote (if you wish to use a heavier platter) : "The solution is obvious- slow down the servo loop to match the increased inertia. Better still, remove the servo loop and run the motor with a fixed frequency three phase drive with waveform compensation to remove the cogging torque."
How does one go about these two options? Are they fairly straightforward to implement?
What is the next main area to target with the sp-10 to improve its sound?
What sort of stand would best suit the sp-10 in a heavy, non-suspended plinth; lossy ala Ikea Lak table, suspended or heavy non suspended mass loaded?
How does one go about these two options? Are they fairly straightforward to implement?
What is the next main area to target with the sp-10 to improve its sound?
What sort of stand would best suit the sp-10 in a heavy, non-suspended plinth; lossy ala Ikea Lak table, suspended or heavy non suspended mass loaded?
Mark,
Thank you for your reply about the speed controller. I downloaded a circuit diagram from the service manual for the sp-10 at vinyl engine in the library section of the website.
You mentioned on a forum on Audiogon in response to posts on the Monaco Grand Prix direct drive turntable stereophile review:
"Grooves / Michael, I disagree that hall effect commutation has the effect you say it has, I think the problem is much simpler and was first "outed" by Sansui's engineers 20 years ago. It comes down to Newton's Third Law (you know the one, equal and opposite reactions etc)
The motor control electronics continually vary the drive to the motor to keep the platter speed constant. Any competently designed servo loop does this to an exquisite degree, virtually eliminating speed variation as a concern, but the variation in drive simultaneously torques the motor's mounting in the opposite sense.
By definition in a direct drive this counter torque is transmitted to the turntable chassis whence it proceeds to wreak havoc. How the designer copes with said havoc defines the success of any DD effort and it means that the designer is faced with a very different set of challenges from the designer of a belt or idler drive TT.
Mark Kelly"
I am assuming that this applies equally to the SP-10. What is the best way to deal with this? Does it mean that the plinth for the SP-10 and similar dd tables is more critical than usual? And does it mean a very heavy constrained layer damped plinth would be the best way to try and combat this?
Thank you for your reply about the speed controller. I downloaded a circuit diagram from the service manual for the sp-10 at vinyl engine in the library section of the website.
You mentioned on a forum on Audiogon in response to posts on the Monaco Grand Prix direct drive turntable stereophile review:
"Grooves / Michael, I disagree that hall effect commutation has the effect you say it has, I think the problem is much simpler and was first "outed" by Sansui's engineers 20 years ago. It comes down to Newton's Third Law (you know the one, equal and opposite reactions etc)
The motor control electronics continually vary the drive to the motor to keep the platter speed constant. Any competently designed servo loop does this to an exquisite degree, virtually eliminating speed variation as a concern, but the variation in drive simultaneously torques the motor's mounting in the opposite sense.
By definition in a direct drive this counter torque is transmitted to the turntable chassis whence it proceeds to wreak havoc. How the designer copes with said havoc defines the success of any DD effort and it means that the designer is faced with a very different set of challenges from the designer of a belt or idler drive TT.
Mark Kelly"
I am assuming that this applies equally to the SP-10. What is the best way to deal with this? Does it mean that the plinth for the SP-10 and similar dd tables is more critical than usual? And does it mean a very heavy constrained layer damped plinth would be the best way to try and combat this?
I wish I knew.
I am personally betting that removing the servo loop and going to cogless synchronous drive will reduce the effort required.
Beyond that, you are on your own.
IMO the OP's scheme of adding a mass of cast iron to the motor base works because it slows down the rate of torque transfer between the motor base and the rest of the assembly. The compliant base didn't work because nobody thought to design it specifically to cope with the torque transfer.
One thing I'd like to try is one of Jonathan Weiss's slate plinths on an SP10 with my electronics.
I am personally betting that removing the servo loop and going to cogless synchronous drive will reduce the effort required.
Beyond that, you are on your own.
IMO the OP's scheme of adding a mass of cast iron to the motor base works because it slows down the rate of torque transfer between the motor base and the rest of the assembly. The compliant base didn't work because nobody thought to design it specifically to cope with the torque transfer.
One thing I'd like to try is one of Jonathan Weiss's slate plinths on an SP10 with my electronics.
I'm not sure what you mean by this, but my belief was that the cogging torque of the motor manifested itself as vibration. The cast iron block was chosen not because of its mass, but because of its inherent vibration damping capacity. A block of magnesium of the same size but of much less mass would have been more effective whereas a similar block of brass would have been slightly heavier but much less effective.
These high damping capacity materials work independently of frequency, so (a large) mass isn't necessary - I don't know what the minimum weight of the cast iron could be and retain its effectiveness. The block I used is probably much larger than it needs to be because its ability to damp vibration is amplitude-dependent and the vibration of concern is relatively small.
John
EC8010 said:
I believe that what you propose is in fact how the SP10Mkii already works. Removing the platter interferes with the closed loop control, so the "cogging" you see is probably not a realistic indication of what happens in practice.
And beware of using old examples to evaluate the SP10's performance. A few models I have worked on did in fact cog with the platter in place (it was not visible, but could be identified by an audible drumming sound). But these were faulty units. Models in good condition do not display this trait (else how could it meet its excellent spec?).
Anyway, I gave up on those t/t's (they're now on ice), due to their use of now defunct components. I support any efforts at making a sinusoidal syncronous drive circuit using modern electronics. In fact, in another thread, I offered to do the PCB design if someone else comes up with a schematic.
A modern / updated control would be more interesting!
I think you will find most schematics at http://www.vinylengine.com/
(or, I do have some large PDF-files...)
Arne K
I think you will find most schematics at http://www.vinylengine.com/
(or, I do have some large PDF-files...)
Arne K
Shaun said:
As you say, removing the platter upsets the servo, but if you had a motor driven with sine waves, it would still rotate smoothly even with a disturbed servo. Obviously, the platter and servo together do not cog nearly as much, but it does show that the drive is not sinusoidal (I think it's trapezoidal). Faulty power supply electrolytics cause cogging with the platter in place.
EC8010 said:
EC8010
I don't intend to be argumentative; here I'm just saying what I know (about the SP10):
The 3-phase sine wave generation is done by the motor (by some additional 3-phase generator windings). This signal feeds the motor drive circuitry. The service manual shows the output of the drive circuit to be sinusoidal (at the least it is a distorted sine wave; at worst, a trapezoid with rounded corners- it's been hand drawn, so the only way to really tell is to stick on an oscilloscope).
Your assertion that sine wave drive results in smooth rotation anyway is true only if the 3-phase drive signal is constant. But above I stated that in the SP10 the sine wave is derived from the motor rotation. Removing the platter profoundly compromises this generator system, due the absence of the angular inertia provided by the rotating mass of the platter.
Agree about the faulty PSU capacitors. However, the PSU used in this case had new capacitors fitted. Also, the cogging was only evident on certain motor units; this also happened with another PSU that is switch-mode, and of a higher output current spec. To be honest, though, I put those t/t's away for when I can build the new circuitry. I felt that my time would be better spent this way.
What we need is a measurement...
I have no problem with differences of opinion. I see what you're saying, but I'm not convinced that the motor is driven by a sine wave. The tacho or frequency governing signal derived from the platter is more likely to be a square wave (hence the "teeth" in the magnet producing it, enabling nice sharp edges for phase comparison). It's many years since I connected a scope to the motor windings whilst it was rotating, but I would point out that as the frequency is derived from a divided-down crystal, you'd expect to see some rather elegant and extensive filtering if the motor was driven by a sine wave. Japanese servos of that era tended to be pulse driven (for efficiency) rather than sine wave driven.
The only way to prove any of this would be to measure, but I no longer have an SP10.
Shaun said:
I have no problem with differences of opinion. I see what you're saying, but I'm not convinced that the motor is driven by a sine wave. The tacho or frequency governing signal derived from the platter is more likely to be a square wave (hence the "teeth" in the magnet producing it, enabling nice sharp edges for phase comparison). It's many years since I connected a scope to the motor windings whilst it was rotating, but I would point out that as the frequency is derived from a divided-down crystal, you'd expect to see some rather elegant and extensive filtering if the motor was driven by a sine wave. Japanese servos of that era tended to be pulse driven (for efficiency) rather than sine wave driven.
The only way to prove any of this would be to measure, but I no longer have an SP10.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.