I'm not trying to start a debate here. I'm not voicing my opinion on DD decks. Nor am I questioning that bad DD decks are bad. My problem is that DD decks, regardless if it's the EMT 950 or a cheapo plastic deck, often are discarded as a group. No matter how you look at it, this threat is about DD decks. It's not about decks in general. You don't see any, "Why is belt-drive bad" threads. And many people that look down at DD TTs for some reason have no problem with the technology when used in CDMs. It just doesn't make sense. I mean, how can a technology be good and bad at the same time? You only have to read this thread and stuff like "cogging," all which must exist in a CDM as well. And if the lack of inertia is a problem with DD decks it has to be so much bigger problem in a delicate CDM. Those rubber matts are said to improve the sound of CDs.
If the criticism only concerns bad DD TTs, why are we even talking DD? Shouldn't it then just be about good and bad TTs?
If the criticism only concerns bad DD TTs, why are we even talking DD? Shouldn't it then just be about good and bad TTs?
Ah, but the data (0s and 1s) coming off a CD goes through a time base corrector. It's written into memory complete with any off-disk speed variations, but presented to the DACs by being read out with a different (and hopefully perfectly stable) clock. That's why all CD players quote W&F as "quartz crystal accuracy." With LP, we have to fret about unwanted modifications to our analogue signal both in time and amplitude.
I started the thread, because I'm beginning my DIY TT project, and I wanted to make it DD. Like you, I had noticed that DD is looked down on, and I wanted to know if there was something inherent in the technology that I should be aware of.
What I got so far is that speed control with negative feedback implies some degree of W&F, and that massive platters are better. Neither of these problems is inherent in DD. Some folks (baggins) believe that the belt in a BD system makes it possible for the platter to have less speed variation than the motor. I don't know whether it's true, but if so, that would constitute something that's inherently bad about DD. (I think.)
In fact, I think you'll find that speed control in a CD player is much less precise than a good TT. It doesn't matter because, as EC8010 noted, the timing is corrected after the information comes off the disc.
Hi Pixpop,
Point is not that a DD is bad or whatever. But to let an electric motor running at very low speed (33 rpm) AND at a very constant speed without speed aberrations at the same time without a gearbox is rather complicated and most time expensive. But it is possible.
It is simply much easier archived with a belt driven platter with a motor that runs at a much higher speed. That’s all.
Cheers 😉
Point is not that a DD is bad or whatever. But to let an electric motor running at very low speed (33 rpm) AND at a very constant speed without speed aberrations at the same time without a gearbox is rather complicated and most time expensive. But it is possible.
It is simply much easier archived with a belt driven platter with a motor that runs at a much higher speed. That’s all.
Cheers 😉
pixpop, just to make sure there are no mistakes. I didn't mean you. Just as you say, it was about the fact that we are more or less taught that DD is inherently bad. We are on the same wavelength.
Yes Pjotr you hit the nail right on the spot. 33 1/3 is difficult for an drive to create right on the axle, with a gear its more easy to get an equal speed.
Btw Pjotr, do you this nickname also on the dutch diy site?
I could not check flutter with the paper strobe as i wanted, its not accurate enough, and to small. The original cast aluminium strobe on the outer of the platter is much larger, and on this strobe i see very small "jerks" in speed, in a non-lineair patter (not equal with totations) First i thougt of small differences of the strobe spots, but with some continous-toned music i can hear the flutter also. But still strage that the speed of both strobes is not the same, small difference but still.
I am still curious what mods could be done on the electronic to get rid of the "jerks"
Tubee
Btw Pjotr, do you this nickname also on the dutch diy site?
I could not check flutter with the paper strobe as i wanted, its not accurate enough, and to small. The original cast aluminium strobe on the outer of the platter is much larger, and on this strobe i see very small "jerks" in speed, in a non-lineair patter (not equal with totations) First i thougt of small differences of the strobe spots, but with some continous-toned music i can hear the flutter also. But still strage that the speed of both strobes is not the same, small difference but still.
I am still curious what mods could be done on the electronic to get rid of the "jerks"
Tubee
DD dissected...
Hello,
I recently dissected a dead Direct Drive unit from Marantz.
One thing which struck me was the way the platter was driven.
In effect the platter was a circular linear electro motor with 4 electro magnets on the pcb (fixed to chassis) with the controlling electronics and several hall elements probably to determine the rotation speed of the platter.
Hello,
I recently dissected a dead Direct Drive unit from Marantz.
One thing which struck me was the way the platter was driven.
In effect the platter was a circular linear electro motor with 4 electro magnets on the pcb (fixed to chassis) with the controlling electronics and several hall elements probably to determine the rotation speed of the platter.
I know, its a very fuzzy shot scanned with an antique scanner from photo, but this is the modded direct drive:
http://img217.imageshack.us/img217/1732/plspeler4vd.jpg
Plinth is plywood on spikes, rubber suspension, marble base, DD from Akai, self-made aluminium anodised arm. First pic right after it was finisched, about 1989.
http://img217.imageshack.us/img217/1732/plspeler4vd.jpg
Plinth is plywood on spikes, rubber suspension, marble base, DD from Akai, self-made aluminium anodised arm. First pic right after it was finisched, about 1989.
This is most likely a fairly standard brushless DC design. Are there magnets attached to the platter, or maybe a ring magnet?
The Hall sensors are usually there to tell the electronics when the magnets are in the correct position relative to the coils. When that happens, the electronics send current through the appropriate coils.
The torque that results has higher harmonic content when compared to a true sine-wave generator.
The Hall commutation is much simpler, and that's why it's common.
pixpop said:
Ah, the magic word that explains everything! I had been wondering about your comment regarding the phase relationship between the sensors and stator coils. Now I understand. Thank you.
Yes, that makes sense. I think this is why the better DD motors seem to have many poles. They can generate more constant torque that way.
Thinking about this, it seems to me that if you're going to add mass to a belt drive system, you're better off adding it to the motor than to the platter, because you'll store more energy for the same mass. Similarly, you'd want a motor with a large diameter, rather than a small one. A flywheel would work, if attached directly to the motor shaft.
The elasticity of the belt would help isolate the platter from speed variations in the motor, but if the platter speed varied because of stylus drag, then a stretchy belt would make it harder for the motor to correct the speed variation. I think you would want something that doesn't stretch much, like mylar tape. I've seen some (DIY) that use recording tape as a belt.
The elasticity of the belt would help isolate the platter from speed variations in the motor, but if the platter speed varied because of stylus drag, then a stretchy belt would make it harder for the motor to correct the speed variation. I think you would want something that doesn't stretch much, like mylar tape. I've seen some (DIY) that use recording tape as a belt.
Agreed, but consider that you could be storing up problems further up the (detectable) audio band. By running your flywheel at a significantly higher speed you will increase the effective mass. Nobody said that you had to use only one flywheel...
DD dissected...
Hi,
as I can remember there were no visible magnets attached to he platter.
One option might be that there was ferro magnetic material forced into the platter, by I do not recollect any signs of that.
Hi,
as I can remember there were no visible magnets attached to he platter.
One option might be that there was ferro magnetic material forced into the platter, by I do not recollect any signs of that.
I,d leave the high mass in the platter if I were you for the reasons
presented earlier in the thread :- i.e
the inertia of the platter resists speed changes due to fluctuating load
these therefore occur slowly (damped ) - the belt acting as the drive
and the negative feedback loop - stretches (either way) and therefore
as long as the motor has sufficient torque adding flywheel to its shaft is unnecsessary (never could spell that ****** word).some flywheel
effect here may not hurt but overall the mass is better in the platter.
Just apply Mr Newton's F=MA
it occurs to me that dd turntables may not actually be able to run without generating some error signal, is this true?
presented earlier in the thread :- i.e
the inertia of the platter resists speed changes due to fluctuating load
these therefore occur slowly (damped ) - the belt acting as the drive
and the negative feedback loop - stretches (either way) and therefore
as long as the motor has sufficient torque adding flywheel to its shaft is unnecsessary (never could spell that ****** word).some flywheel
effect here may not hurt but overall the mass is better in the platter.
Just apply Mr Newton's F=MA
it occurs to me that dd turntables may not actually be able to run without generating some error signal, is this true?
This is a fairly difficult one to articulate, but I'll take a shot.
We start off with the motor which inherently disperses noise at:
a) mains/psu oscillation
b) pole cogging (typically number of poles X rpm)
c) bearing noise
added to the above will be transmission noise imparted via belt adhesion & slip or similar via idler wheel if employed. Further more, any flywheel; no matter how perfectly executed will be slightly eccentric & impart at least one node of "wow" per revolution.
It makes sense to me that whilst in an ideal world many of these effects could be dissipated and thus irradicated as heat, some will inevitably be stored (being as it were continuously "topped-up") and transfered to the following stage, be it another flywheel or a weight. Anyone whose played with a multiple speed open reel for any length of time can see how whilst some of the stored energy will be changed in pitch by differing speed flywheels, some will be passed on "as is". You might also want to take account of the differing mechanical resonances of the flywheels themselves.
error
For feedback / closed loop, ther may be some kind of phase-locked-loop? Also a pll can be good / bad...and tuned or modified....
Arne K
For feedback / closed loop, ther may be some kind of phase-locked-loop? Also a pll can be good / bad...and tuned or modified....
Arne K
It depends. Any servo system will have error, however small. If the load is constant, and the servo includes an integral term (the 'I' in PID control) then over time, the average error will be zero. But whenever there's a disturbance such as dynamic drag, there will be an error signal generated. This applies equally to DD or BD systems that incorporate servo speed control.
- Status
- Not open for further replies.