Of course it could be essentially perfect but the summary of the past few replies was just that today's technology allows much better performance than what was traditionally considered acceptable. No offence.
Ah yes, that's probably true for larger motors with a small number of poles and may help with smoothness, but any 3-phase motor driven sinusoidally ought to be fairly smooth due the overlap of phases - there are some 5-phase motors (steppers usually?) that take things a step further.
In the case of the Papst outer rotor (Aussenlaufer) hysteresis motors, when driven by a 3 phase supply with the correct phase relationship,
I don't think there are any AC motors with less vibration. I use Pyramid's SG4 design and even with larger motors it's almost impossible
to detect whether the motor is running or not. The downside is the inefficiency, my setup uses 3 x 80VA transformers and a 3 channel 200W
amp (Linn motor for comparison)
I don't think there are any AC motors with less vibration. I use Pyramid's SG4 design and even with larger motors it's almost impossible
to detect whether the motor is running or not. The downside is the inefficiency, my setup uses 3 x 80VA transformers and a 3 channel 200W
amp (Linn motor for comparison)
I see there are two speeds and power ratings on that motor, presumably one for star connection and one for delta? Also, noting the relative size, the motor for the Linn has a tiny rotor spinning at 250rpm, whereas the outer rotor motor (by comparison) has a humungous flywheel spinning at twice speed (x4 inertia) or four times speed. You'd have to try very hard to make something that big not run smoothly.
Nice post,
@ralphfcooke , I'm sure that these Papst aussenlaufers hysteresis motors for reel to reel capstan are the best possible thing to drive TT, but they are not easy to find, and when one is for sale it is about 400 USD. I spent a year looking for one w/o success (did not want to pay that much). They are really wanted for repairing old Revox-es and Teacs.
@Herbmeister In few weeks (after holidays) I will be assembling this https://www.diyaudio.com/community/...eded-with-motor-and-drive.412984/post-7764422
The hearth is made in UK, by kind member @richb who shared whole project, it is rpi pico based 3-phase sine generator and 3x power amps need to be done by yourself (when my simplistic amp is tested I can send you details to build). For motor, as best alternative IMO, after looking a lot, I will go testing with drumhead and capstan motors from VCR's. My favorite so far is drumhead motor (also aussenlaufer) from huge sony pro HD digital tape recorder. However, I will show to all the tests and I hope that motors from scrapped Phillips VCR (you can get for no cost) will also be good,
But that will be later September - October...
@ralphfcooke , I'm sure that these Papst aussenlaufers hysteresis motors for reel to reel capstan are the best possible thing to drive TT, but they are not easy to find, and when one is for sale it is about 400 USD. I spent a year looking for one w/o success (did not want to pay that much). They are really wanted for repairing old Revox-es and Teacs.
@Herbmeister In few weeks (after holidays) I will be assembling this https://www.diyaudio.com/community/...eded-with-motor-and-drive.412984/post-7764422
The hearth is made in UK, by kind member @richb who shared whole project, it is rpi pico based 3-phase sine generator and 3x power amps need to be done by yourself (when my simplistic amp is tested I can send you details to build). For motor, as best alternative IMO, after looking a lot, I will go testing with drumhead and capstan motors from VCR's. My favorite so far is drumhead motor (also aussenlaufer) from huge sony pro HD digital tape recorder. However, I will show to all the tests and I hope that motors from scrapped Phillips VCR (you can get for no cost) will also be good,
But that will be later September - October...
Yes, the stators of all sinusoidal-flux motors are the same but 3-phase is not a must because 2-phase is equivalent, related by way of Clarke transform.
Last edited:
The reason 3-phase is so common is because it uses the wiring more efficiently - 4 wires (3 phases + neutral) carry 3 times the power of 2 wires (single phase + 'neutral'), or 3/2 times the power of 3 wires (2-phase + neutral). Assuming each phase is the same voltage and current, where P = V*I
single phase power P over 2 wires. 0.5 P per wire
two-phase. power 2P over 3 wires 0.667 P per wire
three-phase power 3P over 4 wires. 0.75 P per wire. 12.5% better than 2-phase, 50% better than single phase.
So most industrial AC motors are 3-phase, to reduce the amount of copper in the power distribution network.
2-phase is easy to derive from any other scheme using a center-tapped transformer, so wanting 2-phase devices doesn't impact the choice of power distribution much, hence power distribution is commonly 3-phase. For domestic single-phase (or two-phase) power its usual to just cycle through the 3 phases from house to house down the road to keep reasonably good balance on the phases through statistical averaging.
This means if some clever individual contrives to switch off power to every third house in a long street, the power network will likely trip out... Induction motors in particular don't work well with poor phase-balance, so its wise for a precision motor to run from a motor driver rather than direct from 3-phase mains - not that that's likely for a TT motor(!) - maybe someone somewhere has a 2 ton tungsten platter requiring an industrial-scale power source (!). A 2-ton tungsten platter would be about 1.5m thick BTW...
single phase power P over 2 wires. 0.5 P per wire
two-phase. power 2P over 3 wires 0.667 P per wire
three-phase power 3P over 4 wires. 0.75 P per wire. 12.5% better than 2-phase, 50% better than single phase.
So most industrial AC motors are 3-phase, to reduce the amount of copper in the power distribution network.
2-phase is easy to derive from any other scheme using a center-tapped transformer, so wanting 2-phase devices doesn't impact the choice of power distribution much, hence power distribution is commonly 3-phase. For domestic single-phase (or two-phase) power its usual to just cycle through the 3 phases from house to house down the road to keep reasonably good balance on the phases through statistical averaging.
This means if some clever individual contrives to switch off power to every third house in a long street, the power network will likely trip out... Induction motors in particular don't work well with poor phase-balance, so its wise for a precision motor to run from a motor driver rather than direct from 3-phase mains - not that that's likely for a TT motor(!) - maybe someone somewhere has a 2 ton tungsten platter requiring an industrial-scale power source (!). A 2-ton tungsten platter would be about 1.5m thick BTW...
Hi all, thanks again for all the replies and discussion, it's fascinating to learn about everyone's setups etc. In case it's any use for people coming upon this thread with similar questions in the future, my current progress is as follows:
An attempt at running a 'GBM5208-75T' aliexpress BLDC gimbal motor using a simplefoc shield and arduino. I tried both open loop and closed loop with an as5048 magnetic encoder, both suffered from cogging (not smooth) rotation which is probably due to the cheap chinese motor being used in the wrong application.
I then received a BLWS231S-24V-2000 ANAHEIM AUTOMATION motor via Radwell components in the uk (an age after ordering it!), this could be driven very smoothly using the simplefoc shield but with limited torque due to the low power rating of the shield. So I have built an SG4 controller and the matching MA3D amplifier, which was always my backup plan. The setup works but I need to get back to university so I can use an oscilloscope to balance the outputs, as at the moment the motor has much more vibration than in my arduino test.
At the time of ordering the ma3d/sg4 components I hadn't come across richb's arduino controller but going forwards i will certainly have a closer look as the sg4 ended up adding up to a pricey bunch of parts! But I want to thank everyone for making these projects open source and for helping with questions as I know how much work it is. I'll post again once I settle on a motor setup
I really like the look of the papst motors, is it just a case of keeping an eye on the forum marketplace for people selling theirs? I couldn't seem to find any on ebay/etc.
An attempt at running a 'GBM5208-75T' aliexpress BLDC gimbal motor using a simplefoc shield and arduino. I tried both open loop and closed loop with an as5048 magnetic encoder, both suffered from cogging (not smooth) rotation which is probably due to the cheap chinese motor being used in the wrong application.
I then received a BLWS231S-24V-2000 ANAHEIM AUTOMATION motor via Radwell components in the uk (an age after ordering it!), this could be driven very smoothly using the simplefoc shield but with limited torque due to the low power rating of the shield. So I have built an SG4 controller and the matching MA3D amplifier, which was always my backup plan. The setup works but I need to get back to university so I can use an oscilloscope to balance the outputs, as at the moment the motor has much more vibration than in my arduino test.
At the time of ordering the ma3d/sg4 components I hadn't come across richb's arduino controller but going forwards i will certainly have a closer look as the sg4 ended up adding up to a pricey bunch of parts! But I want to thank everyone for making these projects open source and for helping with questions as I know how much work it is. I'll post again once I settle on a motor setup
I really like the look of the papst motors, is it just a case of keeping an eye on the forum marketplace for people selling theirs? I couldn't seem to find any on ebay/etc.
Most 3-phase power systems are 3-wire with a delta-star transformer (with secondary grounded neutral) at the consumer end only to support single-phase loads. And, 2-phase was traditionally 4-wire and unbalanced (both phases did not add to zero) and was therefore ditched by the power distribution industry about 100 years ago.
However, record player motors are really low power and do not really suffer from the consequences of using a 2-phase system.
A 3-phase motor has the added ability to reject the triplen harmonics in addition to the even order ones (e.g. by disconnecting the star neutral), further improving torque ripple, acoustic noise etc. However, a record player needs maybe a <50W motor, which is not very far from 0W, from a power distribution point of view.
I think two special transformers in Scott-T confirguration would be needed to obtain compatibility with standard 3-phase, which complicates things.
https://en.wikipedia.org/wiki/Scott-T_transformer
Yes, I agree. Motors would run much better from a drive than from the mains. This also allows the user to easily select any number of record speeds / rpms, without much problem.
A small induction motor run using an open-loop V/f controller (MCU based PWM) should work reasonably well for the torque requirements. For example, a 230V/50Hz 4-pole induction machine (say 1450rpm) could be run at 5.3V/1.15Hz to get 33 rpm at the rated torque.
However, for a 4-pole synchronous motor, the rotor speed would be accurately 33.333rpm at 1.111Hz, even without speed feedback, which is not true for the induction machine, due to the slip phenomenon.
Hi,
It is very difficult. Papst made 100's of models and not all are suitable. Best Papst motors (and other brands) are produced in mid 20th century and are today very much looked for. I did not see one single of those on market in 1 year of search. Attached is brochure of good stuff that is not made at all any more, just rotor is so expensive and nobody needs it today.
Further, most of these motors are made for 100V and more and 2 phase. At that time PMV frequency drivers did not exist, and building sine generator + amps was expensive, so they normally connected to mains via resistor and capacitor. This makes it mandatory to use transformers or mains frequency inverter to drive them. See one good original scheme for 2 speeds (black lines above go to mains):
You can still find some (not Papst) ; look at ebay "reel to reel capstan motor" , and check the prices, nothing good under 3 - 400€
Papst is also today making good motors, I got 2 at home https://www.diyaudio.com/community/...eded-with-motor-and-drive.412984/post-7688090
But these are now optimized for different purposes and generally not available for diy. If I select some interesting motors at Mouser it says not available in my country (even we are EU). Besides they cost equivalent mass in gold.
Last industrial scale need for precision low power, low speed, low rumble motors was VCR, so I put my eggs in this basket. Lot of scrapped VCR's around 🙂
I was lucky enough to buy a few Papst Aussenlaufer motors a while ago, before they became so desirable. While some of them are the synchronous, others are 'near synchronous ', with a small slip angle. If you are lucky enough to come across one look for a part number that starts 901, these are synchronous. The 902 versions are close to synchronous, while the 903 variant is designed to be used at variable speed, for example as as spool motor on a r-to-r tape machine.
ALL of the Aussenlaufer motors I have are 3 phase delta wound, and work best when fed by a 3 phase supply, but can, as many are, be powered by a single phase supply in conjunction with a phase capacitor. This latter method is less efficient, and provides a 90* phase shift rather than the optimum 120*.
ALL of the Aussenlaufer motors I have are 3 phase delta wound, and work best when fed by a 3 phase supply, but can, as many are, be powered by a single phase supply in conjunction with a phase capacitor. This latter method is less efficient, and provides a 90* phase shift rather than the optimum 120*.
https://www.google.com/url?sa=t&sou...AQFnoECB4QAQ&usg=AOvVaw1ay4FCTxJZcupmA68AzNjw
There is also a possibility of using a tiny Arduino or RPi PLC to control a variable frequency drive, with a sensor to provide feedback, to control the speed.
YMMV, look around on the net.
There is also a possibility of using a tiny Arduino or RPi PLC to control a variable frequency drive, with a sensor to provide feedback, to control the speed.
YMMV, look around on the net.
Ah yes, a non-stretchy belt. But how to do the join? I've wondered about recording tape, from 1/8" cassette to 1/2" video cassette, but although editing blocks and appropriate (non-oozy) tape were once available for 1/4" tape, I haven't seen anything recently. The other possibility seems to be a cotton thread and ignore the knot. Any thoughts?
Superglue another piece of belt across the join?
You can make a long-splice in thread to avoid a knot/bump.
You can make a long-splice in thread to avoid a knot/bump.
With inverter drives, you could couple the platter to the shaft of the motor directly. With AC motors at synchronous / sub-synchronous speeds, most controllers do not use flux weakening and therefore provide full torque at any speed including speeds very close to standstill.
There are equations for speed variation in flywheels in Mechanical Engineering, design the thing to stay stable, and a keep slow sampling interval, with a slow corrective action, on the control circuit.
A flywheel on the heavy side will be less prone to speed variation due to high moment of inertia.
The speed variation can be kept within tolerance expected in comparison to the medium to premium tables made by famous makers, even those had tolerances for speed variation.
This must gave been done earlier, it is simply a matter of patient searching, and getting results using newer controllers, it can even be done in analog.
A flywheel on the heavy side will be less prone to speed variation due to high moment of inertia.
The speed variation can be kept within tolerance expected in comparison to the medium to premium tables made by famous makers, even those had tolerances for speed variation.
This must gave been done earlier, it is simply a matter of patient searching, and getting results using newer controllers, it can even be done in analog.