It may be very true that a large transformer offers a superior solution for ac motors. Naim Armagedon certainly has a following. I have experimented with regenerated sine waves and know they do produce improvements compared to direct mains connection. The transformers,good as they may be, do not address the problems assosiated with the capacitor second phase generation. Opinions exist that the phase shift for the second phase is not ideally generated with a single cap and the value of the cap needs at least fine tuning to match the motor parameters. I would gladly experiment with isolation transformers, but these experiments are neither cheap, nor reusable. Linn may have chosen to sell Lingo for $1500 but to diy a similar design would cost more like $50 if no account is made for time.
cheers
peter
cheers
peter
Hi,
Maybe so, but the TT manufacturer should have taken care of that in the first place....before shipping the product.
It is reusable and cheap or expensive are relative terms...a dedicated PSU with a digitally generated PSU would be much more restricted in use IMHO....and pollute the mains in the process.
Cheers,
Maybe so, but the TT manufacturer should have taken care of that in the first place....before shipping the product.
It is reusable and cheap or expensive are relative terms...a dedicated PSU with a digitally generated PSU would be much more restricted in use IMHO....and pollute the mains in the process.
Cheers,
Frank, I'll keep your comments in mind while designing the remaining of this contraption. I'm sure we'll have a lot to discuss latter on.
Peter, I said my design was cheap, but I think it will cost a bit over $50.
Regarding phase shift, Rega claims that there is something to be gained by adjusting it to match a specific motor, so I want to try it out and see what happens.
Guilherme.
Peter, I said my design was cheap, but I think it will cost a bit over $50
.Regarding phase shift, Rega claims that there is something to be gained by adjusting it to match a specific motor, so I want to try it out and see what happens.
Guilherme.
In regards to tuning the phase shift of the second winding to a specific motor: I would think a two channel scope and a handful of little caps would be the ticket to checking and tuning that.
I've used a few Hurst motors in work related projects and there has never been any mention of cap tuning in their literature. If the phases weren't right at 90 degrees, there would be a loss of torque, but it's a cosine error and would have to be significantly out for a noticable drop in torque.
As a side note, I've used an isolation transformer to clean up my power going to my turntable and it didn't make much of a difference at all, but I have a VPI with a 20 lb platter.
Sheldon
I've used a few Hurst motors in work related projects and there has never been any mention of cap tuning in their literature. If the phases weren't right at 90 degrees, there would be a loss of torque, but it's a cosine error and would have to be significantly out for a noticable drop in torque.
As a side note, I've used an isolation transformer to clean up my power going to my turntable and it didn't make much of a difference at all, but I have a VPI with a 20 lb platter.
Sheldon
I fail to see your point here. Are you suggesting I do this project and then tell people to tune phase by trial and error with a bunch of capacitors and a scope (that they may not have) when putting that feature in it means writing a few more lines of code ?
I'm a bit skeptic myself, but as I said, wiling to give it a try. Anyway, Rega talks about improvements in motor vibration not torque.
Guilherme.
Heavy stuff
Hi Frank,
I'm with you about heavy platter: my hefty 40 kg platter project is pending implementation. Just one question bothers me: is the rather weak DC motor torque sufficient to support the constant speed rotation (not talking about start)? Should the motor torque characteristics match the platter rotation inertia (like cartrige compliance/ arm effective mass conformity)?
Regards,
Michael
Hi Frank,
I'm with you about heavy platter: my hefty 40 kg platter project is pending implementation. Just one question bothers me: is the rather weak DC motor torque sufficient to support the constant speed rotation (not talking about start)? Should the motor torque characteristics match the platter rotation inertia (like cartrige compliance/ arm effective mass conformity)?
Regards,
Michael
Attachments
Hi Michael,
In order to overcome the high inertia of the heavy platter you need a powerful motor.
I use one with a 30 mili-Newton/m torque.
Once the platter is up to rotational speed the high mass and its gyroscopic effect make the motor pale into insignificance.
You'll need to put the motor on a solitary base and fix it in a high mass box because these tend to vibrate heavily.
I use an 8 kg metal bloc for that.
What kind of belt are you considering to use?
Cheers,
In order to overcome the high inertia of the heavy platter you need a powerful motor.
I use one with a 30 mili-Newton/m torque.
Once the platter is up to rotational speed the high mass and its gyroscopic effect make the motor pale into insignificance.
You'll need to put the motor on a solitary base and fix it in a high mass box because these tend to vibrate heavily.
I use an 8 kg metal bloc for that.
What kind of belt are you considering to use?
Cheers,
Hi,
Not that I'm aware of, SY.
What has been done before though is use the maximum torque of the motor unit and then lower the voltage in order to reduce it.
I also recall the British Dais TT having an ingenious system where the pulley shaft got lifted upwards untill rotational speed was achieved. After that the pulley was idling at much reduced torque.
I can't recall the exact detail but Tom Fletcher at Nothingham Analogue in the U.K. should.
Using several motor units adds noise from spinning shafts.
If using just one motor and a timer this could then be used to drop the feed to a reduced supply once correct speed is achieved.
Alternitavely just a rheastat of sufficient wattage could do the same job but manually.
Theoretically this offers reduced vibrational noises and could potentially increase the motor units' life.
As you may know, most A.C. motors run from a dropped down supply of 100 to 80 VAC so that shouldn't be too hard to implement.
There's no need to resort to that unless you'd use a belt material with low friction such as silk or VCR tape....I do use my paws as brake though.
Which is a great way of warming up your hands during cold winter nights.
Cheers,
Not that I'm aware of, SY.
What has been done before though is use the maximum torque of the motor unit and then lower the voltage in order to reduce it.
I also recall the British Dais TT having an ingenious system where the pulley shaft got lifted upwards untill rotational speed was achieved. After that the pulley was idling at much reduced torque.
I can't recall the exact detail but Tom Fletcher at Nothingham Analogue in the U.K. should.
Using several motor units adds noise from spinning shafts.
If using just one motor and a timer this could then be used to drop the feed to a reduced supply once correct speed is achieved.
Alternitavely just a rheastat of sufficient wattage could do the same job but manually.
Theoretically this offers reduced vibrational noises and could potentially increase the motor units' life.
As you may know, most A.C. motors run from a dropped down supply of 100 to 80 VAC so that shouldn't be too hard to implement.
There's no need to resort to that unless you'd use a belt material with low friction such as silk or VCR tape....I do use my paws as brake though.
Which is a great way of warming up your hands during cold winter nights.
Cheers,
GUILHERME said:
no, with your code, you should know the phase exactly (or within clock jitter and harmonic errors). What I'm saying is that people who don't want to build a frequency synthesizer can play with gettting the two coil phases exactly right (or not right, depending on what floats you boat) with a two channel scope and some extra caps to tune the phase shift cap. The scope allows you to see what the phase relationship really is.
Sheldon
Maxon 110189
Hi Frank,
I'm planning to use Teres motor/controller, actually Maxon 110189, with max. torque 56 mNm (short time permissible overload) and max. continuous torque of 16.6 mNm. I simply have no choise but using mag tape 1/4" belt, just to keep the side load (belt tension) within permissible 5.5 N without slippage. Belt tension is to be regulated by string/counterweight force applied to the motor housing sliding on small balls. Interesting, Chris Brady starts to experiment with mag tape recently.
Thanks,
Michael
Hi Frank,
I'm planning to use Teres motor/controller, actually Maxon 110189, with max. torque 56 mNm (short time permissible overload) and max. continuous torque of 16.6 mNm. I simply have no choise but using mag tape 1/4" belt, just to keep the side load (belt tension) within permissible 5.5 N without slippage. Belt tension is to be regulated by string/counterweight force applied to the motor housing sliding on small balls. Interesting, Chris Brady starts to experiment with mag tape recently.
Thanks,
Michael
Hi Michael,
Ah, side thrust...the Maxons' Achilles heel.
I hope you won't have to use all the engineering tricks mentioned above...it seems rather elaborate.
Maybe you could use a capstan to preload the motors' pulley with a counterforce equal to belt tension?
Just a thought though...I never had to try it out so far.
Cheers,
Ah, side thrust...the Maxons' Achilles heel.
I hope you won't have to use all the engineering tricks mentioned above...it seems rather elaborate.
Maybe you could use a capstan to preload the motors' pulley with a counterforce equal to belt tension?
Just a thought though...I never had to try it out so far.
Cheers,
Interesting thread. There is a kit TT drive available now, see my ad in Audio Asylum's classifieds section.
To answer some of the things raised in the thread, there is an interaction between the precision of the clock, the number of steps in the waveform construction, the phase shift introduced by the filter(s) and the precision of the phase difference (quadrature).
Probably the optimal design approach is that taken by Analog Devices in their DDS chips, which I am using as the basis of some new designs (won't be available as kits).
The compromise on which I settled for the kit builder was to use CMOS (low speed and low current make for minimal noise) and only eight steps in the waveform, therefore requiring only five voltage levels. Eight is divisible by four so quadrature is established on the digital side. The output filtering is easier than with a square wave and is accomplished by a pair of four pole switched capacitor filters.
Once you have a quadrature pair any phase relationship can be created by simple vector arithmetic.
Weird geeky thing - I went looking for E24 resistor ratios which established phase shifts of round numbers of degrees eg 10, 15, 20, 22.5, 30, 40, 45 etc. I was astonished to find that such ratios can be found which give accuracies of better than 0.05 degrees in all but one case (80 degrees where the ratio is too close to unity). As an example if you want 40 degrees you want a ratio of sin 40 which is 0.6429. 3.6 / 5.6 just happens to equal 0.6429.
It looks like coincidence but the chances of it happening were pretty small (the average spread between ratios should be about 0.5 degrees)
To answer some of the things raised in the thread, there is an interaction between the precision of the clock, the number of steps in the waveform construction, the phase shift introduced by the filter(s) and the precision of the phase difference (quadrature).
Probably the optimal design approach is that taken by Analog Devices in their DDS chips, which I am using as the basis of some new designs (won't be available as kits).
The compromise on which I settled for the kit builder was to use CMOS (low speed and low current make for minimal noise) and only eight steps in the waveform, therefore requiring only five voltage levels. Eight is divisible by four so quadrature is established on the digital side. The output filtering is easier than with a square wave and is accomplished by a pair of four pole switched capacitor filters.
Once you have a quadrature pair any phase relationship can be created by simple vector arithmetic.
Weird geeky thing - I went looking for E24 resistor ratios which established phase shifts of round numbers of degrees eg 10, 15, 20, 22.5, 30, 40, 45 etc. I was astonished to find that such ratios can be found which give accuracies of better than 0.05 degrees in all but one case (80 degrees where the ratio is too close to unity). As an example if you want 40 degrees you want a ratio of sin 40 which is 0.6429. 3.6 / 5.6 just happens to equal 0.6429.
It looks like coincidence but the chances of it happening were pretty small (the average spread between ratios should be about 0.5 degrees)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.