advertisement


Simple DC motor control

davidsrsb said:
The hard part is measuring the rotation. Measure once a revolution and you can only have a time constant of about 5 seconds. Ok for long term stabilty but not for drag due to modulation, warps etc.
If yo could find a way of putting a strobe marking around the platter, then the loop can be much faster. The challenge is that this pattern must be very uniform
Click to expand...

Alternatively, measure rotation at the motor - which is much faster (1200 rpm is typical for DC motors), so you can have tight loop. I've built circuits like this; easiest method is to use a motor with built in sensor.
 
What's the point of tight loop control if you're driving it with a belt. Better to control drift and leave it at that.
 
sq225917 said:
What's the point of tight loop control if you're driving it with a belt. Better to control drift and leave it at that.
Click to expand...

Dynamic wow. The Oracle turntable used an open loop DC motor, and was notorious for dynamic wow - the motor speed changes with torque, slows on crescendos and recovers in quiet passages.

The simple technique described in the OT mitigates this by increasing applied voltage in proportion to torque; this doesn't help with drift.

Local feedback at the motor doesn't help with belt creep and related sources of drift, but fixes motor drift AND dynamic wow.

Very slow feedback loops (like Linn use) fix drift, but not dynamic wow.

Best of all, but complex, is two loops; tight loop at motor for dynamic effects, slow loop derived from platter to eliminate long term drift and inaccuracy.
 
Ill stick with stiffly powered, regenerated a.c. motor and enough bearing drag to render programme based effects moot.
 
PigletsDad said:
Dynamic wow. The Oracle turntable used an open loop DC motor, and was notorious for dynamic wow - the motor speed changes with torque, slows on crescendos and recovers in quiet passages.
Click to expand...
So long as the dynamic wow is due to the motor and not dynamic belt slip.
I do agree that measuring motor speed with a reasonable bandwidth is easier.
Many dc motors have a tacho output for use as fan speed sensors
 
Chaps,

I'm trying to make an OL Advanced controller at least work reasonably on my PT Too...

It's the Gen 6 circuit and I've already got the balanced traffo, and swapped the diodes for Phillips avalance items, 10KuF supply cap and a 0.1uF on the inlet, also added as large a heatsink as poss to the reg. Thing is it's still not really doing the job.

I think Simon, Martin C or maybe PD had another mod that involved adding a pair of matched transistors and 2 resistors as a bias to the op amp, can someone please walk me through this :)

Yes I know this is turd polishing, but I'm not really in the mood for building another controller (if the OL can even be called that) and if there was a quick way to make this one tolerable then that would be great.

I'm also not sure about going down the AC route for the PT, as there is very little bearing friction - another subject

Dan
 
Hi

Any boards or advices where to check power supply for Maxon 226774? I was thinking maybe try this on LP12. How about speed controller? Is that in power supply or different board?

Regards

Jake
 
Will do Martin, Simon has pretty much already covered this though

You'll laugh, when you see the circuit.

In the end I expect I'll have to buy a new motor/ pulley and build one of your controllers
 
Dan, i've done some experimenting with DC motors etc on an LP12. I originally used the diy motor controller which I got from someone here and used it with several different pulley arrangements and speeds.

Essentially the controller could not compensate for stylus drag and so ultimately was no use. Also subjectively it was dramatically inferior to the Linn Radikal. I also learned that they inject ridiculous amounts of mechanical noise in to the top plate. However I recently installed a new DC set up which works on a similar (ish) principle to the Radikal from Edmond (of Hercules fame) - this was an early prototype unit and uses 4 markers per revolution to track speed as opposed to Linn's single one.

Again the mounting system for the motor was a weak area being directly connected. I changed the motor and also mounted it in a foam pot to reduce resonance. It is not as sophisticated as the Radikal in terms of getting up to speed and stabilising, however once there it now measures very similarly to the Radikal and subjectively sounds very similar. At 45rpm it is almost identical and hard to tell apart.

I have limited electronic expertise (non existent) so my work has mainly been focused on how the motor is mounted physically and this has had considerable influence on the outcomes. I have not gone back to re try the diy dc controller with this new mounting system so must get round to trying it. I'm not familiar with how the PT Too arrangement works but would certainly suggest investigating ways to isolate the motor.
 
Thanks TPA. I need to get in touch with Edmond. The PT motor mounting is essentially the same as an LP12, but at 7 o'clock rather than 10.
 
Dan K said:
Thanks TPA. I need to get in touch with Edmond. The PT motor mounting is essentially the same as an LP12, but at 7 o'clock rather than 10.
Click to expand...

Yep, just google'd some images. In that case yes, I'd have though the same principles would apply. Staying with the larger pulley, lower rpm set up is again probably beneficial from my experience.
 
Dan that's quite a bit different to my pcb, they were slots for two transistors to form a ring of two ccs around that zener voltage reference. Sadly my pics are long gone now.
 
Mine is the MK6b. So what can I do to that transistor and resistor to get the thing stable/ remotely listenable.
 
I think this should go to a new thread?

Dan, what actually is the symptom?

I think the voltage regulator for a dc turntable motor is almost irrelevant, anything will do the job as long as it can supply the current. The physics are that a dc motor must slow under load. The speed of the motor is a consequence of its windings, magnets, geometry and the applied voltage. The motor spins at a rate that generates a voltage that balances the supply. With a perfect motor, as you apply load the current rises and any regulator will continue to supply the same voltage balanced against the voltage generated by the spinning. But because real motors have resistance as the current rises this generates a voltage, which must be added to that generated by the spin, to match the supply. So the motor slows. It doesn't matter how excellent the supply is, this must happen.

There are two ways around this.

1. Use a big motor turning slowly, it has a low internal impedance and the slowing under load drops to insignificant at the loads seen in a turntable. I think this is the Radikal approach. The speed measurement on the Radikal is so that the user doesn't have to adjust the speed for temperature, pressure, weight of record, whatever. It's not about moment to moment compensation.
2. Use a PSU with a negative output impedance, as discussed in this thread. The effect of this is that as the current rises under load the voltage delivered also rises, hence constant speed by magic.

'2' is difficult, especially to make general. Probably best with a small motor turning quite fast. You actually want a high internal impedance to make the compensation more sensitive.

No idea what the OL supply does, I suspect from other posts in this thread that it is a simple regulator. And hence rather pointless if expensive.

Paul
 
Yes it is Si, makes no sense to me either.

Paul, the symptom is an annoying pitch bending on piano and harmonica, the more busy the track the worse it is. The deck has a new belt and is otherwise properly setup. It's a new deck to me and my first DC motored one.
 
You must log in or register to reply here.


advertisement


Back
Top