Motor for turntable

As said before the motor current waveform is dominant. The sine wave is very like baby food. It can not add vibration if pure, like babies it isn't really required. Valhalla in later types was 0.05% THD ( current use 10 % THD ? ) . The only error is phase shift of the 0.22 uF. Voltage and PSU ripple matter most.

This means, the motor must be connected like in this circuit for the loudspeaker:
Amplifier topologies for current-drive | Current-Drive - The Natural Way of Loudspeaker Operation
I.e. the motor coil is part of the NFB network.
I haven't find such a configuration for a motor control unit until now.
The schematic in post #16 under
Linn Lingo schematic
belongs to the old valhalla.
Are there a schematic for the new valhalla ?

I don't understand this term: current use 10 % THD
did you mean the old (first) Valhalla version ?
 
Last edited:
b0bujV5.png



This is the current drawn by the LP12 motor. I was testing a simple 45 rpm idea also. A small resistance is used to show this. The phase shift 0.22 nF shows the same. It is not a sign of poor quality. I have often speculated that using a subtraction of this wave as in negative feedback might work. Digital pre-distortion. An analogue subtractor coulld be used to keep it low cost.
 
If you look carefully at the wave shape it is the same as the pole slots inside the motor. It must be if you think about it. When the very low impedance of a Valhalla is clamped over the motor this quality is hidden. You might say " If I can not see it, it is gone ". The coil resistance is large. The coil will see this distortion. If it were a locomotive you have 10 metres of chain joining the train as this resistance. The Valhalla is the loco. The train does what it likes. The flywheel plus oil drag at least tries to pull the chain tighter.

The vibration of the motor is due to the triangle wave imposed by the magnetic circuit. Slightly due to coil phasing also. If you use a square, triangle or sine driving wave vibration will be similar as the motor has a very powerful filtering quality. Try running the motor at > 100 Hz. It is unlikely to move. If you guess the data for a similar 7.5 degree stepper motor a calculator will tell you why.

It is likely that a sine wave is the slightly better wave as it makes phase shift easier if using Wien style oscillators ( SVF etc ) . Mostly the current determins the vibration.

110 Vrms = bass boom and " muddy detail ".
90 Vrms = good bass power with good detail
80 Vrms = ok bass with more detail ( try 85 )
67 Vrms = excellent detail, less bass.


This is how I hear the LP12 with original motor + Valhalla. The Valhalla is supurb if it's poor DC PSU is improved. This tightens the bass.
 
Mostly the current determins the vibration.

110 Vrms = bass boom and " muddy detail ".
90 Vrms = good bass power with good detail
80 Vrms = ok bass with more detail ( try 85 )
67 Vrms = excellent detail, less bass.

This is how I hear the LP12 with original motor + Valhalla. The Valhalla is superb if its poor DC PSU is improved. This tightens the bass.

Interesting, Nigel. I have no doubt that a better PSU could improve the Valhalla. :)

My own experiments have been with a 'Number9' AC motor controller (not a Valhalla) driving a 24v Premotec (as used by Rega). But I would've thought the same results would've held true for the 110v LP12 Premotec?

We (my 'listening panel' and I - which included the designer of the Number9) found that as the voltage being fed to the 24v motor was increased ... the music got 'better'. It sounded more dynamic and the (room) decay heard with choral singing (Vivaldi's "Gloria") was more extended.

IOW, 24v < 25v < 27v < 30v, in terms of SQ. We stopped at 30v as we thought that it might be dangerous for the motor's health to up it any further. (This turned out to be the case as the motor stopped working after a few months; bringing the voltage down to 27v has prevented this from happening again. :) )

Of course, increasing the voltage fed to the motor ... increases motor vibration. :( So this is not a good thing to do ... unless the motor's PS has a way of minimising motor vibration. (The Number9 does this by being able to vary the phase difference between the 2 pairs of motor wires; the optimum is not the 90 degrees that a cap produces!)

Andy
 
Here's one for you gents to try, swap out the bearing oil for something thicker, thick enough to require adjusting the drive frequency to attain correct running speed. See what you think it sounds like once you've adding enough drag to swamp stylus drag variance ( not much) and motor clogging.

I ran a meldano controller with rega 24v motor fitted with linn pulley, driving linn bearing subplatter and platter in my modded Kuzma Stabi s. This setup sounded and measured better than any other ac setup I'd tried, ie all the linn controllers and hercules including original lingo.

Measured better than my sme 10 !!!!
 
There was a very bad problem at Rega when someone in Belgium used the wrong oil for the motors of the Airpax 110 V type. Legend has it motorcar oil. A black substance is seen in the bearings which possibly is a phosphorous compound. The motors stopped! Washing out the bearings seemed fine for six months. All returned with black gunge. The correct oils are SAE 20 to 30 with no additives. Sewing machine oils are OK. This shows that the LP12 Airpax 110V motor isn't very powerful. I never found a better motor all things considered. The Garrard 301/401 are about 10 times more powerful and use a different motor type.


If you remember my locomotive-chain-train analogy some drag would pull the chain tight. Ms Verdier thought this. His latter turntables used Rocol lathe slide bed oil in the centre bearing. The Garrard 501 uses Terresso 77 ( 68 ) turbine jacking oil. Revox, type 43. Verdier liked the Garrard eddy current speed brake as it reacted like thick oil. Albeit the graph is opposite. When in equilibrium both work in a similar way. Oil looks like tadpoles at the microscope level. When stable all the tadpoles swim the same way. At rest they are random. Old style oils for us are better. Motor oils have cold start chemicals which we do not want. Verdier thought if bearing drag could be greater than stylus drag then sound improves.

The main upgrade to the Valhalla DC PSU is change 47uF 250 ( 200 ) VDC to 220uF 250 VDC. Larger makes no difference. Ripple goes from - 41 dB to - 54 dB. An isolation transformer ( floating ) gives -60 dB if in addition. THe graphs show the hidden ripple ( 50 + 50 Hz beat ) when 67.5 Hz ( 45 rpm ) used. Beyond that the Valhalla single ended PSU for the CD4060 ripple counter and flip flops is off center to the 340 VDC class A power amp ( like mini JLH ). - 60 dB is the best you will get. As the motor will filter subsonics - 60 dB is excellent. To be frank the belt is more trouble when this good.

- 41 dB is just over 1 : 100 - 54 dB is 1 : 500 and - 60 1 : 1000.

crGvKrX.jpg
 
yjjci8I.jpg


This is the distortion of a Garrard 501 motor ( as current ). It's mostly 2nd harmonic at a bit under 1%. As 2nd harmomic is in phase with the principle 52 Hz it causes very little vibration. Far less than a LP12 motor whilst being about 10 times more powerful. The 3rd harmonic is very different ( - 24 dB LP12. - 68 dB 501 ). Here it is below 0.05%. 4th at -60 db/ 0.1% is a fine result. In this case THD is unhelpful. The wave looks like a low distrotion sine wave which is slightly ballooned out. Fingers on the motor feel almost nothing.

The 501 is 52 Hz. The beat problem like the LP12 ( 50 Hz/ 60 Hz , and 52 Hz ). 52 Hz allows a 6 mm motor shaft to directly run the platter. German reviews of the 501 show it has very low flutter and drift. More like a studio tape machine. This is the reason. The 501 motor shaft is to a very high standard and is made in Switzerland.

The 501 motor has no " cogging ". It lags the field and has an elastic quality to the torque. Strangely, very cheap fan heaters use this motor type as did Dansette record players. The 501 is 4 pole at circa 1482 rpm. 6 pole types exist for fans, they would give 2/3 the runing speed. The 501 motor was from a medical application. It donated a rotor and field. The rotor has multiple laminations in a step fashion as is typical. That is vastly better than the LP12 cog jumps. If you rotate the LP12 motor using fingers you feel this. Consideing this a LP12 motor is very good. That motor is obviously an expert design.

The 501 motor uses static balancing. Dynamic balancing by a company trusted by F1 ruined the motor! The 501 motor runs in the exact magnetic centre of it's magnetic field. The LP12 motor is good on that.
 
qlhCu5f.jpg


Here are the Red and Blue phases of the original LP12 as voltage waveforms. A dropper is used to protect the oscilloscope. Red is via 47uF and Blue is via 0.22 uF. It looks at first that the Blue phase is totally dreadful. In fact it is not. It's the truth exposed by the higher impedance of the 0.22uF and the phase shift. This scope is able to work on a live mains connection. Most can not. If using a standard scope a floating Isolation transformer must be used. The Valhalla has a 110VAC option which is useful for this. This option works equally well. Retain the transformer for turntable use. 80VA is enough.

I had a crystal to allow 45 RPM use. Quite a lot of work is required to use one as the Valhalla has a steep filter curve. I used a relay to switch the crystals and the second section to switch the filter gain at gain setting pot. How Valhalla is designed made it possible by piggy-backing the parts. Use a higher voltage for 45 RPM.