In case anyone is confused over how the drive works Brushless DC (BLDC) motor with Arduino ? Part 2. Circuit and Software has a nice diagram (and in part one even an animation). Hopefully this will put to bed the idea that the motor is 'stepping' between positions!
Nigel, there is nothing to cure ! Yes drag is good and on my SL-1200's with captive oil bath's I do use a very viscous oil to give that drag.
But drag and this thread ? Mmmmmmmmmmm ................
But drag and this thread ? Mmmmmmmmmmm ................
To quote
"All these problems with jerkiness (Wikipedia says: Jerkiness, sometimes called strobing – somewhat ironic given that this will be a stroboscope when we’re done) of the spindle are caused by driving it from the binary output (either HIGH or LOW) of the MCU. Three phase motors like these are ideally driven with alternating currents and voltages creating a sinusoidal waveform, 120° apart on each of the three ends of the windings. This requires a lot of effort in a digital world – a DAC chip (Digital-to-Analog Convertor) and plenty of computations. But there is an easier way to “emulate” the effect of alternating current in the binary world of microcontrollers: pulse width modulation or PWM."
The thing to see is the coil most likely forms a first order filter if an ideal stepper running at optimum speed. DD are unlikely to do that. Regardless of what the drive gives out mostly the motor will take what it needs. There should be a smalll advantage using sine waves as the leading edge isn't going from zero to maximum in uS.
I had the same problems when trying to improve a Linn LP12. The phase shift ( bi-phase using 0.22 uF if 50 Hz ) is a small thing. Voltage is the main thing. To be frank the motor was not very good. The Lingo has a 90/66V switch. The class A and the exact phase shift do nothing special. The 66 V changed the sound but would make the JVC L3-E seem even better in terms of verve.
Until I do my own tests I will be wildly guessing. Hopefully the SP10's work as they are BBC versions. There is the DD75 also. Goodness knows how that compares. I won't have more than lets say 5 hours to give to this so hope for some luck. If I get somewhere with this I must see if I can get the 201 to work, I have two I can play with. One is pre war 78 only. One is about 1952. The plan was an acrylic platter to keep the PU away from the coils. I have an SME 0ne and SPU somewhere. More likely a Shure M44-7. The Shure is a very good PU if the pre amp suits. Leak Varislope is one if output Z is respected from second EF 86, the scruffy cable usually works as it is 63 pF per metre and is only about 40 cm. Don't change cable unless the same.
The oil bath gamble is that I could actually reduce vibration whilst causing it to work harder. More likely I shift harmonics. An eddy current brake is another way. As Verdier pointed out the eddy brakes damps as would oil even though the graphs look different. As he said the speed is constant so it is the graph at that point we should look at and not if modulated by lets say +/- 1 rpm. Verdier said drag is non notchy. The gamble is keeping the motor sweet. When a belt drive has two motors it has both torque and drive train advantages. The two motors being less than identical helps. Same in a way as the idea is to reduce the abilty of the turntable to slow under load albeit minute. Truely great turntable don't, nor do they over compensate. Denon looked at stylus forces. They are 1000's of time greater than seems likely. The LP is fancy sandpaper.
This thread has been a displacement activity for an electrical engineering problem I started to solve last night. Now to do the PCB. DD TT was the thing I needed to clear my head. The circuit was fine at 207 and 260 V, not great at 200 V ( 10 % below limits although working fine ). Think I just got there and the circuit has no voltage regulators now ( less loss, easier, well inside upper limits ). That can't be bad. The cure was fully saturating a PNP transistor to work as a near zero loss switch. A 12 pence cure. The funny thing is the circuit is exactly like a DD TT although vibration is not important. That is pure coincidence. Money at the end of it so not just for fun. One disadvantage of engineering this way is it makes it look cheaply designed. Usually it takes me a month to make it look cheap. If I get rushed I do it the usual way. It often costs double and it slightly less reliable 7 years down the road. One silly example. I made a capacitance multiplier running at 500 V. To say it was fantastic would be to put it mildly. Once in a while it went wrong. I tried increasingly complex protection circuits. Being FET nothing was fast enough. So I said this. What is 500 V at 11 amps. 47 R. What is 47 R at 80 ma ? 3.8 V, that seems an ideal trade off. To know the truth of it I used a 0.6 watt resistor. It and the FET never failed. The FET has a drain source diode and I fitted a 12 V zener to gate source. That would be ideal as the gate looses about 5 V. The big deal is the zener will pass through the origin if reverse biased without letting go and fast enough. The problem is it looks like a 13 year old did it to another engineer. Someone will say " I can do better". suddenly problems arrive. Instead of saying " I caused that " the engineer says " what idiot designed this ". I never assume the question I ask has not been asked long before me messing it up.
"All these problems with jerkiness (Wikipedia says: Jerkiness, sometimes called strobing – somewhat ironic given that this will be a stroboscope when we’re done) of the spindle are caused by driving it from the binary output (either HIGH or LOW) of the MCU. Three phase motors like these are ideally driven with alternating currents and voltages creating a sinusoidal waveform, 120° apart on each of the three ends of the windings. This requires a lot of effort in a digital world – a DAC chip (Digital-to-Analog Convertor) and plenty of computations. But there is an easier way to “emulate” the effect of alternating current in the binary world of microcontrollers: pulse width modulation or PWM."
The thing to see is the coil most likely forms a first order filter if an ideal stepper running at optimum speed. DD are unlikely to do that. Regardless of what the drive gives out mostly the motor will take what it needs. There should be a smalll advantage using sine waves as the leading edge isn't going from zero to maximum in uS.
I had the same problems when trying to improve a Linn LP12. The phase shift ( bi-phase using 0.22 uF if 50 Hz ) is a small thing. Voltage is the main thing. To be frank the motor was not very good. The Lingo has a 90/66V switch. The class A and the exact phase shift do nothing special. The 66 V changed the sound but would make the JVC L3-E seem even better in terms of verve.
Until I do my own tests I will be wildly guessing. Hopefully the SP10's work as they are BBC versions. There is the DD75 also. Goodness knows how that compares. I won't have more than lets say 5 hours to give to this so hope for some luck. If I get somewhere with this I must see if I can get the 201 to work, I have two I can play with. One is pre war 78 only. One is about 1952. The plan was an acrylic platter to keep the PU away from the coils. I have an SME 0ne and SPU somewhere. More likely a Shure M44-7. The Shure is a very good PU if the pre amp suits. Leak Varislope is one if output Z is respected from second EF 86, the scruffy cable usually works as it is 63 pF per metre and is only about 40 cm. Don't change cable unless the same.
The oil bath gamble is that I could actually reduce vibration whilst causing it to work harder. More likely I shift harmonics. An eddy current brake is another way. As Verdier pointed out the eddy brakes damps as would oil even though the graphs look different. As he said the speed is constant so it is the graph at that point we should look at and not if modulated by lets say +/- 1 rpm. Verdier said drag is non notchy. The gamble is keeping the motor sweet. When a belt drive has two motors it has both torque and drive train advantages. The two motors being less than identical helps. Same in a way as the idea is to reduce the abilty of the turntable to slow under load albeit minute. Truely great turntable don't, nor do they over compensate. Denon looked at stylus forces. They are 1000's of time greater than seems likely. The LP is fancy sandpaper.
This thread has been a displacement activity for an electrical engineering problem I started to solve last night. Now to do the PCB. DD TT was the thing I needed to clear my head. The circuit was fine at 207 and 260 V, not great at 200 V ( 10 % below limits although working fine ). Think I just got there and the circuit has no voltage regulators now ( less loss, easier, well inside upper limits ). That can't be bad. The cure was fully saturating a PNP transistor to work as a near zero loss switch. A 12 pence cure. The funny thing is the circuit is exactly like a DD TT although vibration is not important. That is pure coincidence. Money at the end of it so not just for fun. One disadvantage of engineering this way is it makes it look cheaply designed. Usually it takes me a month to make it look cheap. If I get rushed I do it the usual way. It often costs double and it slightly less reliable 7 years down the road. One silly example. I made a capacitance multiplier running at 500 V. To say it was fantastic would be to put it mildly. Once in a while it went wrong. I tried increasingly complex protection circuits. Being FET nothing was fast enough. So I said this. What is 500 V at 11 amps. 47 R. What is 47 R at 80 ma ? 3.8 V, that seems an ideal trade off. To know the truth of it I used a 0.6 watt resistor. It and the FET never failed. The FET has a drain source diode and I fitted a 12 V zener to gate source. That would be ideal as the gate looses about 5 V. The big deal is the zener will pass through the origin if reverse biased without letting go and fast enough. The problem is it looks like a 13 year old did it to another engineer. Someone will say " I can do better". suddenly problems arrive. Instead of saying " I caused that " the engineer says " what idiot designed this ". I never assume the question I ask has not been asked long before me messing it up.
I built a quite large house on never accetping that. For music one needs that. Usually anything can be made to work better and the cures cost pennies or far less. Time is money and a fair bit of autism need. Can you imagine Harrison making clock No 4. It looks so wrong although beautiful and the No 1 so right. I did a motor project with Greenwich University. I was asked to go to patent. I didn't bother as I felt the idea was more than obvious. I might do a unipivot arm with them this year. The effects were to reduce vibations in motor down to two sinusoidal components with no chaotic components. The data was vastly better than hope ( factor of 100 at a crude guess ).
Clockwork would be the ideal turntable motor. Alas, who would build it ? A string drive as tall as a light house and even so too small for ideal bobbin size.
Nigel, you are not listening. The output of the envelope detector on the vast majority of turntables is a sine wave. That's how they are driven.
......and he refused to sell his excellent tracking force balance when he found that it was being used by those in the drug pushing world. Fortunately I still have one which is very accurate and useful.
I got it on steroids . You should suspect that from my word go analysis that insisted it must be a variation of synchronous. I was very surprised to see it is a continuous low frequency type that I had no sucess with when cheaper motors. As said I will look at it and see if my picture is better. I do get paid a bit of money for seeing things differently. People don't want same old same old when doing something new and better. I watched about the Big Bang being impossible yesterday. At least with this it's just looking at it differently rather than saying wrong or right. I think you are too hooked up on pictures I am not looking at and with luck won't have to. For example do we define an engine by it's mapping or how energy flows. I think the latter the more productive although mapping is useful. I am not trying to repair an idea that isn't broken. I am asking how does the energy flow. Can I manipulate it at low cost to fit my vission of what is best. What is sad is we can not program our cars to do what we like best. It would be very easy to do if built in from word go. I would tune mine for comfort and no performance. 0 to 60 15 seconds , 70MPG with kick down if wanted. Fly by wire in aircraft was when somone said " what if". It's bonkers and it works. The thing is the aircraft were being held back by the pilots. If so why not make a plane they could not fly and get a bonus. The 1950's Flying Wing would have worked if fitted with year 2000 computing. It's designer was allowed to see the Stealth Bomber when it was top secret to show how his design matured. The FW was already hard to spot on radar.
Some say NASA never went to the Moon as no way they could they simulate a vacuum landing. It turns out a vacuum makes it very easy as it is constant and virtually no aerodynamic add ons . Mars is hard because what is almost a vacuum is nothing like as good as a vacuum. Mars weather is able to make simulation nearly impossible. People tie themselves in knots when the basic reasons tell the story. We also have a situation here thought to be Moon like and actially is Mars like. The vaccum if you like is the DD flywheel which although good is not perfect. The servos fight the residual between groove friction and flywheel energy storage. Doubtless the design is highly intelligent. This I want to see.
BTW. As my picture looks to drag how would anyone understanding the subtleties of the electronics help that ? I don't give a stuff how it gets what it does. I just want to see the scope working and take what I think to be propper readings. Not sure it's worth saying anything if I find an engineering Elephant in the room as looks to me no one is interested. The thing is I can get data both with and without drag including playing a record. It is likely it will be worse, sometimes we get lucky and we see and get a result. Not seeing is not knowing.
I have to say apart from being told what is already known have we gone a step forward?
I also got my PCB done. It is 300 x 70 so not a small one. Somehow wasting time here wasn't . I needed a distraction.
Nige, I am afraid you are now coming across as needing to up your meds. We don't give a flying what you are paid for to 'think different'. I also watched that program and do not need a synopsis from someone who did not understand it.
Your knowledge of control theory seems very very very limited. As such can I respectfully ask that you stop posting here as you are not adding to the discussion at all but doing a very good job of coming across as someone who wants to show how clever he is and is failing miserably at it. Unless you can contribute your ramblings are not appreciated.
Your knowledge of control theory seems very very very limited. As such can I respectfully ask that you stop posting here as you are not adding to the discussion at all but doing a very good job of coming across as someone who wants to show how clever he is and is failing miserably at it. Unless you can contribute your ramblings are not appreciated.
Would anyone care to say the rights and wrongs of this PDF. I like idler drive as it problems are easy to solve which is not to say cheap. I am not happy with 20 kg platters although the Verdier solution works. Belt drive to my ears is the worse solution. It seems always to have the sound of a budget priced turntable when verve comes into it. The Lenco is better and betters the SP10 sometimes. Treble dynamics. The interesting thing is both the SP 10 and GL75 have similarities of sound. Both sound like live broadcasts for want of a better way to say it. Linn sounds like a second generation tape off of a Revox.
http://www.brinkmann-usa.com/inhalt/en/technical/a_short_study_on_turntable_drives.pdf
http://www.brinkmann-usa.com/inhalt/en/technical/a_short_study_on_turntable_drives.pdf
How can anyone know how much a person understands? When looking at people I work with I have often underestimated how much they understood.
I have someone who works for me who does the imposssible when I need it. He can not read nor write if above age 10 ( 6 in reality, 10 on a good day ). And yet he can make things work that shouldn't. He can always see the picture. That is rare. He never makes mistakes that seem to bog people down on this forum. He was a gunsmith and just knows what is best. He is a very quick learner and yet is 64 now. When he was intervied he said he owned a BSA Gold Star. I asked if he maintianed it ( He is in the film the Leather Boys on the same ). That was interview over and job was his. He forgot to say he was on 9 Concords he helped build ! His mates did his paperwork as he would have been unable to get the job if not. Real world engineering is not always about ordinary people.
I have someone who works for me who does the imposssible when I need it. He can not read nor write if above age 10 ( 6 in reality, 10 on a good day ). And yet he can make things work that shouldn't. He can always see the picture. That is rare. He never makes mistakes that seem to bog people down on this forum. He was a gunsmith and just knows what is best. He is a very quick learner and yet is 64 now. When he was intervied he said he owned a BSA Gold Star. I asked if he maintianed it ( He is in the film the Leather Boys on the same ). That was interview over and job was his. He forgot to say he was on 9 Concords he helped build ! His mates did his paperwork as he would have been unable to get the job if not. Real world engineering is not always about ordinary people.
Hiten: metal is your friend here. Whether the affordable DD donors have enough of the right sort of metal I have not measured or calculated.
That's what I am saying. If a close magnet can not affect the cartridge as shown in the video; I am sure a coil magnetic field through a small distance, metal platter, rubber mat and vinyl should not be that much of a concern.
Regards.
Regards.
A bigger problem is the magnet in the cartridge. I was looking at affordable platter conversions last night. Some of the Thorens 2 part platters looked as if they would fit the bill, but magnetic steel so MC cartridges would be a no no. My hunt continues.
The more I look the more I can't see issues with DD turntables other than decades of attack by flat earthers.
The more I look the more I can't see issues with DD turntables other than decades of attack by flat earthers.
Although never spoken of and I can hear no problems an acrylic patter to raise the height and add a little mass. That might be OK for many reason. For the Garrard 201 this was the plan. Although claimed by some to be non magnetic rumour says some TD 124's had iron platters. I have seen them, but not tested with a magnet. Same idea again.
5.5 Hz might not get through if DD anyway.
I searched without luck for the waveform screan grabs of a Technics free running or under load. A finger load would be interesting. One would hope a repetitive waveform somewhere. I can understand why these tests are not shown as it means cutting the PCB and insering a very small resitance. More ideal would to take the track at the chip and hope the resitance to the coil withing the scope range. It could be fed into an op amp to make a few uV measurable. It can be filtered if like class D it upsets the analyser. To stress, the voltage waveform is of zero interest. It is the series waveform from chip to coil. It is three phase so a quick test for how well coils match. If pulse position sampling or whatever I have no great interest as I am not trying to change that. All that interests me is the good and the bad of loading the motor. Overshoot and distortion under load. Even if a crude test we can infer the subtle reaction to a stylus. This tests the Verdier conjecture. That is if the drag is 10 times greater than the stylus then the stylus is no longer the force to be corrected. If this comes with no wow, flutter, or rumble penalty if it a massive free lunch. I suspect if the reputation of the 1210 is right it could feed the 5000. The measurments would be to be certain bias of exspectations do not state it better when it measures worse. When valve amps I might allow that, not when a turntable.
To be clear. Vibtaion in the platter is directly related to the distortion in the current waveform if the design has no avoidable bearing problems etc. The exact reason it is that way is not for now of interest. Just to see if worse under load. Odd harmonics being worse. Second harmonic almost is of no imprortance. Books I have read on motors which have no interest in hi fi state about harmonics and vibration. They start looking at 3 rd harmonics as a prime problem.
Looking at the motor the obvious difference DD to synchorous is that the latter has two simple bobbin coils and some magnetic material forming triangular windows. The DD has shaped coils in what one might call a horse collar look. This shape mostly fits the space . However it might do more.
Everyone says if disrgarding the black box electonics that the motor likes a 5.5 Hz sine wave of some type by whatever PWM system etc. My question is how close is it to sine. That is not black box output and how it got there. It is in motor demands and wants. On the Garrard 501 I got the motor current demands beflow 1 % THD. Data becomes unreliable below that. Using a core feedback system it LOOKS to be 0.3 % and mostly second. The output of the amplifier is close to 0% . That means nothing. If I have one regreat with the Garrard motor. I should rewind it for 24 Vrms.
On that point. To change the black box demands is silly as so much trouble has gone into the design . Also it disrespects the engineers who must have spent 1000's of hours designing it. This echo's the point about second guessing at a distance another engineer abilities. What we can do is ask if the DD can be asked to work harder and give us more. Anyone who sees fault in that should use CD as a prime source. LP is imperfect, one can play with how good or bad. For example RIAA 3180 uS isn't cast in stone. It was a recomendation. 318uS should be fixed and 75 uS seldom is 75 uS. IEC 7980 uS or whatever should be rejected. 2 uS seems to work.
The structure of the turntable can be changed. This almost is as a loudspeaker and needs nothing more said.
The first thing I will do is get a working SP10 to test. I have been told I can borrow one. The two came from Lockwood Audio who swapped them for one Garrard 301. Since then they have been in a box doing nothing for perhaps 10 years. I thought the l Lockwood cabs I took to Austria for a friend were the biggest pile of junk until I heard them. I was wrong. Being wrong is the greater part of being right. When I said I have, I mean I can borrow. It is borrow without asking or keep if I want. I always ask.
Why I thought these waveforms might exist is for years I was a service engineer. The Sony video recorder manuals always had waveform type. It was of no importance technically. It was a way of saying good or bad.
I don't understand technical aspects of direct drive turntable electronics and effects of motor being right below the platter. Nor the moving platter affecting the cartridge magnet+coil. My experiment was pretty crude but extreme. A headphone magnet attached to the cartridge should show some irregularities, but it didn't. Turntables which are not direct drive has mechanical moving components which if not made with tight tolerance looses the advantage I guess. Direct Drive has single bush and spindle rotating. Other turntables have two bushes in motor, shaft with pulley, belt or idler which drives rim and platter holding spindle bush. All needs to be precision made. Not that I don't like belt or idler drive. This is just sharing of thoughts. I once visited a frail old electronics repair man. He had a rickety Garrard changer and Phillips valve radio. Very likeable sound. May be the limited band the 78 rpm was playing its trick. or it was quiet and fresh in the morning. I don't know. Music is so much related with the mood. But I have also heard high end various drive turntables. Mid-fi direct drive turntable can be good value for money. Tonearms and plinth can be upgraded.
Regards.
Regards.
The JVL3-E shocked me how good as it looks very ordinary. Like a USB turntable of now. Genuinely a Linn LP12 Ekos DL110 sounds slightly less good in areas Linn think they are the best in. Now in my own terms I want to know why. I have to do it in ways I understand or can get simple data from.
On the subject of magnetic fields. Some think a small current into a PU coil might be good. 47 k gives 9.5 mV NE5532 and 14.5 mV MC33078 as an indication of a DC coupled input at 47K . If 15/50 uA matters it is hard to say. The BH curve will change a bit and maybe better ? Call it DC bias as in cheap cassette recorders that didn't have AC bias. At less than a uA I doubt it matters.
On the subject of magnetic fields. Some think a small current into a PU coil might be good. 47 k gives 9.5 mV NE5532 and 14.5 mV MC33078 as an indication of a DC coupled input at 47K . If 15/50 uA matters it is hard to say. The BH curve will change a bit and maybe better ? Call it DC bias as in cheap cassette recorders that didn't have AC bias. At less than a uA I doubt it matters.
http://www.audiomods.co.uk/papers/pardee_recordfriction.PDF
Some measurements on bearing drag vs stylus drag. roughly equivalent for a mid 70s belt drive. but vinyl coefficient of friction varies a lot! So amount of additional inertia or bearing drag needed is small to tip the balance. Also a reference to a 1970 hifi news article on the same. Questions answered 30 years ago, box thought out of.
Some measurements on bearing drag vs stylus drag. roughly equivalent for a mid 70s belt drive. but vinyl coefficient of friction varies a lot! So amount of additional inertia or bearing drag needed is small to tip the balance. Also a reference to a 1970 hifi news article on the same. Questions answered 30 years ago, box thought out of.
Thanks for the link billshurv.
With the hazard of thread going in to contentious stylus drag debate, would some kind soul give some figures of torque required to over come the drag friction calculated in the above link. Since this is coefficient which I guess will change for 45 rpm and suppose we have odd cartridge which requires around 3 gm tracking weight. Just to set the tough conditions. (Please be gentle I am technically challenged).
Thanks and regards.
With the hazard of thread going in to contentious stylus drag debate, would some kind soul give some figures of torque required to over come the drag friction calculated in the above link. Since this is coefficient which I guess will change for 45 rpm and suppose we have odd cartridge which requires around 3 gm tracking weight. Just to set the tough conditions. (Please be gentle I am technically challenged).
Thanks and regards.
- Status
- Not open for further replies.