Hi from Germany!
#16 Re: Hi from Germany!
FYI, that was a DC Supply
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#17 Re: Hi from Germany!
Yep, I should split this off into its own thread.
Phase 1 = sin
Phase 2 = -0.5sin - 0.866cos
Phase 3 = -0.5sin +0.866cos
I wish I had thought of that. I generate sine an cosine by DDS to drive two phase motors, so three phase would be a handful of op amps and some resistors.If you have an oscillator that produces sine and cosine, appropriate addition of the two will produce 120 degrees and -120 degrees
Phase 1 = sin
Phase 2 = -0.5sin - 0.866cos
Phase 3 = -0.5sin +0.866cos
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
- Thermionic Monk Status
- Posts: 5899
- Joined: Thu May 24, 2007 11:22 am
- Location: People's Republic of South Yorkshire
#18 Re: Hi from Germany!
I have nothing to add to this thread, apart from hello!, but it's fascinating. Please keep up the discussion.
-
- Old Hand
- Posts: 260
- Joined: Fri Mar 25, 2016 10:58 am
#19 Re: Hi from Germany!
That's right; a few op-amps using those relationships and you're done.
Incidentally, for the Airpax-style motor, what you're really looking at is two motors, one stacked on top of the other. Each motor produces sinusoidally modulated torque, but cos^2 + sin^2 = 1, and that's why you need the 90 degree phase shift between the two motor coils. But simply shoving a capacitor in there doesn't work terribly well. You can have 90 degrees or equal voltages, but not both. Ideally, you drive each motor separately from sine and cosine. When you do that, you find (surprise, surprise) that the motor isn't perfectly made. Assuming you have equal voltages to the coils, the rotor needs to have equal height in each motor. So the original Connoisseur motor had a little bracket across the bottom of the motor with a central screw and thrust bearing that allowed rotor height to be adjusted for minimum vibration. That helps a lot. Never seen that trick openly deployed anywhere else. But you still don't get silence because the two motors are not in perfect rotational alignment. So what you then do is use an op-amp to exploit the trigonometrical relationships seen earlier and allow +/-10 degrees of phase tweak on either sine or cosine. When you do that, the difference is quite remarkable. A tweak of the phase pot and the motor suddenly doesn't vibrate any more. And the nice thing about all this is that the trigonometry works at any frequency. So you can adjust oscillator frequency (fine speed control) and everything still works.
But direct digital synthesis is a perfectly valid way of doing things. Just don't go following the DACs with low-pass filters that have any noticeable phase shift at 50Hz or you've thrown away all the digital advantages. Mostly, that's exactly what people do. You really want audio quality DACs. Which is why the Altmann turntable suggests using a portable music player as the source.
You're right; this discussion probably should be split off.
Incidentally, for the Airpax-style motor, what you're really looking at is two motors, one stacked on top of the other. Each motor produces sinusoidally modulated torque, but cos^2 + sin^2 = 1, and that's why you need the 90 degree phase shift between the two motor coils. But simply shoving a capacitor in there doesn't work terribly well. You can have 90 degrees or equal voltages, but not both. Ideally, you drive each motor separately from sine and cosine. When you do that, you find (surprise, surprise) that the motor isn't perfectly made. Assuming you have equal voltages to the coils, the rotor needs to have equal height in each motor. So the original Connoisseur motor had a little bracket across the bottom of the motor with a central screw and thrust bearing that allowed rotor height to be adjusted for minimum vibration. That helps a lot. Never seen that trick openly deployed anywhere else. But you still don't get silence because the two motors are not in perfect rotational alignment. So what you then do is use an op-amp to exploit the trigonometrical relationships seen earlier and allow +/-10 degrees of phase tweak on either sine or cosine. When you do that, the difference is quite remarkable. A tweak of the phase pot and the motor suddenly doesn't vibrate any more. And the nice thing about all this is that the trigonometry works at any frequency. So you can adjust oscillator frequency (fine speed control) and everything still works.
But direct digital synthesis is a perfectly valid way of doing things. Just don't go following the DACs with low-pass filters that have any noticeable phase shift at 50Hz or you've thrown away all the digital advantages. Mostly, that's exactly what people do. You really want audio quality DACs. Which is why the Altmann turntable suggests using a portable music player as the source.
You're right; this discussion probably should be split off.
In the end, it's all physics...
#20 Re: Hi from Germany!
I don't, I use a 2nd order with a -3dB at 1k. I did find getting the distortion down below 0.02% makes a difference (when I found not to use ceramic caps in a filter). Not sure though what difference any phase shift would make as long as the filters are matched between the sin and cos part. I do also allow the phase between the two to be changed from the nominal 90. Though of course anything other than 90 would break creating 3 phase.But direct digital synthesis is a perfectly valid way of doing things. Just don't go following the DACs with low-pass filters that have any noticeable phase shift at 50Hz or you've thrown away all the digital advantages. Mostly, that's exactly what people do. You really want audio quality DACs. Which is why the Altmann turntable suggests using a portable mjusic player as the source.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
- Old Hand
- Posts: 260
- Joined: Fri Mar 25, 2016 10:58 am
#21 Re: Hi from Germany!
Perzackly. You'd be amazed how many people use switched capacitor filters with ceramic capacitors so that the two filters do not match. Obviously, you can take that out in digits, but ceramic capacitors are temperature sensitive, so the error changes with temperature. I forget what the distortion of my oscillator is, somewhere around 0.02%. In an analogue implementation, low distortion is important when you use diode clipping as your amplitude stabilisation mechanism because it implies high Q, and a high Q means minimal phase noise, which is important. No idea why low distortion per se might matter. Lots of people use Wien bridge oscillators, which are low Q and have lots of phase noise.
In the end, it's all physics...
#22 Re: Hi from Germany!
Yep, I get paid to write software, so DDS seemed the way to go, and TBH, it started life as part of the control system for the DD turntable, and there is no way I was going to do that with analogue, the tweekability of digital wins every time.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
- shane
- Social outcast
- Posts: 3456
- Joined: Sun Sep 16, 2007 12:09 pm
- Location: Kept in a cool dry place.
#23 Re: Hi from Germany!
When we designed the TT2, which used the Airpax/Premotec motor, we found that putting a nylon cap on the bottom bearing of the motor containing a spring and a nylon thrust pad pushing up on the bottom of the shaft gave a noticeable improvement in the sound. I always assumed this was because it lifted the rotor off whatever served as a thrust bearing internally and reduced bearing noise, but from what you’re saying it seems likely that it was helping to centre the rotor between the two windings. It would certainly account for why it made so much difference. I can’t remember where the idea came from, but others using that motor had similar arrangements.Morgan Jones wrote: Wed May 01, 2024 2:26 pmSo the original Connoisseur motor had a little bracket across the bottom of the motor with a central screw and thrust bearing that allowed rotor height to be adjusted for minimum vibration.
What will happen comes closer. What has happened goes further away.
-
- Old Hand
- Posts: 260
- Joined: Fri Mar 25, 2016 10:58 am
#24 Re: Hi from Germany!
Yes, Nottingham Analogue found the same. But when I designed a turntable power supply for them, I originally had a tweak for electrical drive and was surprised to find it often needed to be unequal. I then found that if electrical drive was set to be equal, minimum vibration could be achieved by fiddling with height of the rotor, so I concluded that it was all about each coil seeing an equal amount of rotor. This is important because it means you saturate both ends of the rotor at the same drive, but if it's not at the right height, one saturates early and limits your maximum drive. Entirely by chance, I was later given an old Connoisseur motor by a work colleague and when I saw the adjustment on it realised that Sugden was way ahead of me. Beware that the later Connoisseur motor was horrible.When we designed the TT2, which used the Airpax/Premotec motor, we found that putting a nylon cap on the bottom bearing of the motor containing a spring and a nylon thrust pad pushing up on the bottom of the shaft gave a noticeable improvement in the sound.
In the end, it's all physics...