 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Turntable speed control
Page 1 of 2   |
|||||
| Author | Message | ||||
Mixtel90 Guru  Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
This is just theory at the moment, but... Any of you into hi-fi will know that switching between 33-1/3rpm and 45rpm isn't always quite as easy as you might think. Early systems are mechanical, using different diameters of pulley on the motor shaft. The posher systems use a synchronous motor but change its supply frequency. The 24-pole motors usually used rotate at 250RPM on a 50Hz supply (300RPM at 60Hz). This is geared down by a belt drive to give a turntable speed of 33-1/3rpm. To get 45rpm you need to increase the supply frequency from 50Hz to 67.5Hz (or from 60Hz to 81Hz). Of course, we are talking about sine wave supplies here. The mains is a more or less reasonable sine wave, the motor "cogs" a bit and the give in the belt, coupled with the inertia of the turntable, even things out. The idea then is to produce a sine wave at about 90V that can be switched between 50Hz and 67.5Hz. Additionally, a capacitor is used on one winding of the motor and ideally this should have a slightly different value for each speed to minimise motor vibration. My idea is to use a PicoMite with a R2R ladder, at least 8 bits and preferably 10bits or more, to generate the sine wave from a lookup table. The resulting output would feed a class D bridged power amp module from ebay - probably about 30W - to feed a step-up transformer to get the voltage required for the motor. I'd be relying on the crystal oscillator on the PicoMite for the master frequency. I've not worked out whether I can get the correct division values yet, or whether the lookup table can be processed fast enough for higher bit rates. Things similar to this are already done - usually with boards of expensive logic rather than microcontrollers. They also have the usual 1500% audiophool price markup. This solution costs a PicoMite, a £6 amplifier board and about £20 for the transformer plus a few odds and ends. It shouldn't, in theory, perform any worse than £500 motor speed controllers. Any comments? Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4222 |
The audio outputs with low pass filter provides perfect sine waves at near 12 bit resolution. The frequency can be set with sufficient accuracy. But it may be possible to drive a power h bridge directly with the 44kHz output an do the low pass filtering at 90v level. Edited 2023-01-08 00:29 by Volhout PicomiteVGA PETSCII ROBOTS |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
Aren't the frequencies in PLAY TONE and PLAY SOUND integers? In which case it couldn't generate 67.5Hz so the 45RPM would be out on a 50Hz system (44.6667 or 45.3333). It's irrelevant anyway as it'll be more accurate than a motor that's synchronised to the mains. :) Do you think there'd be much 44kHz left after it's been squirted through a reverse-connected toroidal mains transformer? I doubt if they have a great frequency response. I'd be a little wary of trying to filter at 90V because of possible transformer resonances. The H bridge is an attractive option though. I might get a traansformer on order and test this. I'm liking this idea more. One pin of the PicoMite could be pulled high or low to use it on 50Hz or 60Hz motors. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
TassyJim Guru Joined: 07/08/2011 Location: AustraliaPosts: 6097 |
I needed 91.5 Hz for a CTCSS encoder. The pico came to the rescue. Floats do work for TONE. Jim VK7JH MMedit  MMBasic Help |
||||
phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Did this 40 years ago. Simple square wave inverter, 2 transistors driving the secondary of a very small 240V to 12-0-12V transformer. The capacitor value isn't critical. It still worked at 78 RPM. The new Pico PWM -Duty option would be ideal for producing the inverted and un-inverted waves for the push-pull drive , if only I could get it to work. There doesn't appear to be any inversion. "Duty cycles are specified as a percentage and you can use a negative value to invert the output (-100.0 <= duty <=100.0)." I expected this to do it. Setpin 1, PWM Setpin 2, PWM PWM 0, 67, 50, -50 |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4222 |
I posted a single register change (poke) to make pwm symetrical. So you can have 45% duty cycle at one, and 45% at the other pin, and 5% gaps at bpth leading and trailing edges. Works better when driving a transformer PicomiteVGA PETSCII ROBOTS |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Re the new PWM. It's the PicoVGA that I used above, Just tested the LCD one and it does work. @Volhout Yes, for bigger transformers that is necessary but for the tiny one needed for a turntable motor it didn't matter. Edit Very nice! > setpin 1,pwm > setpin 2,pwm > Poke word &h40050000,3 > f = 45 / (100/3) * 50 * 2 > ? f 135 > pwm 0, f, 45, -55 > Differential between pins 1 & 2  https://www.thebackshed.com/forum/ViewTopic.php?FID=16&TID=15127#191042 CH 0- Poke word &h40050000,3 'CH 0 CH 1- Poke word &h40050014,3 'CH 1 CH 2- Poke word &h40050028,3 'CH 2 CH 3- Poke word &h4005003C,3 'CH 3 CH 4- Poke word &h40050050,3 'CH 4 CH 5- Poke word &h40050064,3 'CH 5 CH 6- Poke word &h40050078,3 'CH 6 CH 7- Poke word &h4005008C,3 'CH 7 RP2040 Datasheet pp 549 - 555 . Edited 2023-01-08 13:53 by phil99 |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
Right, I have some bits on order. I decided to stick with the power amp and filter before it, Volhout. The reasons being that the output protection is better and there are no heatsinks. The idea is that this unit fits into a recess in the bottom of a 1.5" thick plinth so there's not a lot of height available. The total height of the board now is only about 22mm, the space available is 30mm. The platter and arm are mounted on a second 1" thick plinth, mounted above this on some sort of isolating mounts that I'm not sure about yet. Could be springs or sorbothane in some form. That isolates the platter and arm from the motor, which is mounted on the lower plinth and belt-coupled. I've never built a turntable before - this is fun. :) Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Any reason to not use a DC motor with a pulse generator (encoder) as a tacho? Not that a turntable should experience load variances but a closed velocity loop would maintain a constant velocity, regardless. Craig |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
Good DC turntable motors *cost!* This will work on just about all medium range, and some high end turntables without any modification. It's a minimum of £400 for a DC motor and the minimal controller (from Origin Live). The same model goes up to over £1000 when options are added. Most others start at over that. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Who cares about a "turntable motor"? Let me guess, an audiophile is gonna tell me that he gets "greater spatial definition"  Any old DC motor with feedback can maintain any velocity within its spec. Craig |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
One of the main problems with turntables is vibration from the motor. People go to great lengths to minimise it, including mounting the motor completely separately and driving the platter using a long belt. That's not practical if the turntable is a mid-range one and already built. Most of these turntables use very similar motors, consuming about 1.8W. The very cheap "suitcase" turntables, based on a Chinese copy of an old BSR combined platter and arm, almost all use the same motor. It's a little brushed DC motor with a chip mounted on it. It's speed stabilized by sensing the back emf. Unfortunately they aren't too good after a bit of wear. Reducing the belt tension to the minimum helps, as does dropping the motor voltage. In both cases it's easy to go too far - the turntable will run fine once up to speed but it needs a manual push to get it started as there is insufficient torque. The system I'm playing with puts fiull voltage on the motor at startup and reduces it to a low level once it's running. Stepper motors look ideal at first but most of them have too much vibration to be useful. (a 7.5 degree stepper can be made to run at 250 rpm from 50Hv AC with only a capacitor). Brushless motors are better but tend to need messy electronics. There is a theory that DC motors can't run at the correct speed anyway. :) They depend on constant feedback in the loop to stat approximately correct and will always, to some extent, oscillate in speed. That's just something for the purists. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Hi-fi is the only hobby where willful ignorance is seen as a virtue. With hobbies like RC aircraft, amateur radio, photography, and astronomy, people learn to build and fix things, and strive to learn as much as they can. Only among audiophiles do we see practitioners embrace magical thinking while rejecting of science and logic. Without extremely precise velocity control, sub-micron machining would not be possible. Craig |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
Oh yes, it's very possible - but only at silly prices. I've been trying to find a motor for this turntable and it's not as easy as you may think. You can get them, if you are willing to pay lots of dosh. The motor manufacturers won't supply you though - how nay thousand would you like, sir? Even simple stepper motors, if reasonable quality, have to be imported from the US or Japan. Of course, the motor manufacturers won't supply the electronics - you need to source that from somewhere else. You'll be very lucky if you can get some sort of anti-vibration mountings out of them. It works out *much* cheaper to buy a second-hand "official" turntable motor (complete with the correct pulley). Yes, there's a lot of smoke and mirrors in the hi-fi world, but there is some very good engineering too. Anyone can get a noise out of a record. That's the easy bit. Getting that noise to sound believable is something else. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
stanleyella Guru Joined: 25/06/2022 Location: United KingdomPosts: 2120 |
run at 78rpm via a AC light dimmer :) |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4222 |
Try to find a coreless motor. They are not very powerfull, but do not have a lot of mass in the moving parts, thus cannot produce a lot of mechanical vibration. Alternatve, take a drive motor from a 5 1/4 innch disk drive. In essence a 3 phase stepper motor. Edited 2023-01-09 08:08 by Volhout PicomiteVGA PETSCII ROBOTS |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
It's quite possible to run at 78rpm. :) As long as the motor can take it and the belt drive is ok. No light dimmer needed. Just a different stylus. I've got the software very usable now: > 33/off/45 switch input. > Pull a pin low to select a 60Hz rather than 50Hz system. > Output is at full voltage for several seconds at startup to get up to speed, then it reduces until Off is selected. > The switch has a LED on the handle, which is lit while running. To do: > Add an output to control the Shutdown pin on the power amp. No sense in leaving it all running unless needed. > Add temperature monitoring to the amplifier chip and transformer. Probably more useful during development than on the final thing. > I might try to monitor the amplifier current. That would let me know when the platter is up to speed and also detect overload conditions. > Might add an additional control mode so that it will work with a push button input with speed indication by LEDs. > Probably other things. :) @Volhout: I did consider a spindle motor. I'd have had to get a pulley made somewhere though. The platter I'm using (and not willing to change) is for a flat belt so it's not just a simple v-notch. The motor I've decided to use is a cast-off from another turntable which is having a later model fitted. Edited 2023-01-09 08:24 by Mixtel90 Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| PeterB Guru Joined: 05/02/2015 Location: AustraliaPosts: 651 |
G'Day Mick etc This may be of some use. Mr. Fourier claims that adjusting the pulse width can be used to reduce the harmonic distortion and can reduce the third harmonic to zero. If the waveform is, +V for 120 degrees, zero for 60 degrees, -V for 120 degrees, zero for 60 degrees, there will be no third but a bit more fifth. The amplitude will be 88%. That will reduce the load on any LPF. At least it may be worth a try / play. Peter |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6779 |
These little motors are pretty crude. A permanent magnet armature surrounded by two coils in a steel cage. Bent over "teeth" on those cages are the stator pole pieces. You can feel the 24 steps as you turn them by hand. However, mount them resiliently and couple them to a relatively big, heavy flywheel with a drive belt and you are firmly in "good enough" territory. If you fiddle with the phase shift between the windings you can reduce the vibration, but even different samples of the same motor will often need different values. 90 degrees is a compromise. Some will be better at 92 degrees, others at 85. Also note that there are some of these motors which look identical that don't self-start no matter what you do with the phase shift - not suitable for turntables usually. Once again, the value of the phase shift capacitor seems to be pretty none-critical in practice, providing that the windings stay within their voltage ratings. Generally anything from 0.15uF to 0.22uF can be considered normal. This can change the vibration characteristics a bit too. Note - there is always some vibration from these no matter how you derive the 90 degree phase shift. You can never get rid of it entirely and while the additional complexity of producing the phase shift at low level then using a second power amp (and transformer in my case) usually helps it comes at a cost. That's because the voltages on the coils don't have to be the same - it's the phase shift that matters. You can always equalise the voltages a bit with a series resistor in the other coil circuit anyway. All this falls to bits unless you drive from a reasonably clean sine wave. Anything less results in something more like stepper motor behavior (not surprising as that's more or less what these are). This is where this sort of speed control wins - the sine wave is much better than you'll get from the AC mains, especially at peak load times, the frequency is more accurate and has much better short term stability, which is what you hear when listening. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
0.25%??? How about 0.05% with a cheap motor and mediocre resolution feedback. Craig Edit: I guess they mean wow/flutter Edited 2023-01-09 18:54 by Tinine |
||||
| Page 1 of 2 |
|||||
 Print this page |
