Dec 31, 2012, 08:01 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by Fourdan Hello Modisc I am not a great specialist in motor drivers. See Wikipedia VFD http://en.wikipedia.org/wiki/Variable-frequency_drive Maybe another guy could answer My idea is that when the rpm are steady .. torque produced by the motor is equal to the loading torque. When there is some acceleration or deceleration, the problem is more complicated. When starting from stop if you have position sensors or not (sensorless) the algorithms are different. Louis
I am aware that motor consumes current depending on the loading torque. But I also notice that when throttle is small, it is very easy to hold the rotor still; while in higher throttle, it can be very difficult to stop the rotor. However, for servo drive, even when the motor is running at low speed, the torque needed to stop the rotor can be very big; in other words, constant torque.

This may explain why high speed inrunner + gear to reduce the output speed is favoured. For example, the maximum torque of a inrunner (15000 rpm)+ gear (5:1) at 3000rpm, may be better than a motor directly running at 3000rpm, at 50% throttle.

I am also not sure, just curious. Thanks for the input.
Dec 31, 2012, 09:17 AM
Aka: Tom Jenkins
United States, FL, West Palm Beach
Joined Aug 2008
4,698 Posts
Can you tell me where to get one of the capacitors in this video? I need one for 110v ac, want to build a generator for a tiny lipo charger
 Overunity free energy generator (2 min 37 sec)
Dec 31, 2012, 03:59 PM
Registered User
Chicago 'burbs
Joined May 2006
1,737 Posts
Quote:
 Originally Posted by modisc I am aware that motor consumes current depending on the loading torque. But I also notice that when throttle is small, it is very easy to hold the rotor still; while in higher throttle, it can be very difficult to stop the rotor. However, for servo drive, even when the motor is running at low speed, the torque needed to stop the rotor can be very big; in other words, constant torque. This may explain why high speed inrunner + gear to reduce the output speed is favoured. For example, the maximum torque of a inrunner (15000 rpm)+ gear (5:1) at 3000rpm, may be better than a motor directly running at 3000rpm, at 50% throttle. I am also not sure, just curious. Thanks for the input.
Maybe I don't completely understand your point (it happens!), but at low throttle, you just need to think that you are applying a low voltage to the motor. Under those circumatances, even stalled only a small current can flow (I=V/R, when the motor is stalled). Small current =small torque..

Now if you increase the voltage (higher throttle), more current can flow, so the torque also can increase.

It isn't actually the load that pulls current and produces the torque, it is the rpm the motor is turning at. That rpm creates a back emf that fights the impressed voltage from the power supply.

So I guess you know all that, but that's my explanation for the low torque at low throttle.

Now for a servo motor, you have to ask what the controller is trying to do. My guess is that it is trying to move to a position, or to a velocity (rpm). If the motor can't move or move fast enough, the controller increases the applied voltage (or equivalently increases the PWM pulse width to the motor) until the motor starts accelerating to the target velocity.

Our RC motors aren't doing that, the controller is just producing a fixed PWM width (or equivalent voltage level), and not trying to hit a rpm. The rpm they get is just set by when the motor torque = load torque. At that point the motor stops accelerating.

The only time that isn't true is when you have a set rpm target in the motor--but then you are letting the controller find the throttle level to reach that rpm.
Dec 31, 2012, 04:25 PM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by Alan Hahn Maybe I don't completely understand your point (it happens!), but at low throttle, you just need to think that you are applying a low voltage to the motor. Under those circumatances, even stalled only a small current can flow (I=V/R, when the motor is stalled). Small current =small torque.. Now if you increase the voltage (higher throttle), more current can flow, so the torque also can increase. It isn't actually the load that pulls current and produces the torque, it is the rpm the motor is turning at. That rpm creates a back emf that fights the impressed voltage from the power supply. So I guess you know all that, but that's my explanation for the low torque at low throttle. Now for a servo motor, you have to ask what the controller is trying to do. My guess is that it is trying to move to a position, or to a velocity (rpm). If the motor can't move or move fast enough, the controller increases the applied voltage (or equivalently increases the PWM pulse width to the motor) until the motor starts accelerating to the target velocity. Our RC motors aren't doing that, the controller is just producing a fixed PWM width (or equivalent voltage level), and not trying to hit a rpm. The rpm they get is just set by when the motor torque = load torque. At that point the motor stops accelerating. The only time that isn't true is when you have a set rpm target in the motor--but then you are letting the controller find the throttle level to reach that rpm.
I thought about what you have said about the RC controllers, but what you said about how the servo controller maintain constant torque output, is very inspiring indeed. It is clearer thinking like you said. Thanks Alan.

Do you have any idea about sinosodial magnetic field vs rectangular magnetic field of the rotor? I have been playing with a dozen of servo motor till now, and i found that most of the rotors (maybe for smaller ones) are made with a single piece multiple-pole ring magnets (4 pole, 6 pole, or 8 pole, or etc), which has a sinosodial magnetic, while rc inrunners mostly have rotors made by several segments of rod magnets which has a rectangular magnetic field. For servo motors, there are also rotors assembled by several segments of rod magnets.
Is there any obvious difference between the two types of magnetic fields, when using a RC controller to drive them?

 Jan 01, 2013, 02:53 AM Life begins at transition Australia, VIC, Sale Joined May 2007 3,593 Posts If you have a sinusoidal magnet, the flux distribution will be sinusoidal, and so will the bEMF (roughly). In this case, it doesn't matter what the winding is. With a "square" rotor flux distrubution, caused by lumped magnets as normally seen in RC motors, you may or may not get sinusoidal bEMF. With concentrated, full pitch windings like we all use, the bEMF will be roughly trapezoidal. If you use a distributed winding scheme though, you can get a bEMF that's sinusoidal, even will a hobby rotor. It makes a few % difference in efficiency, but that's it. If a sinusoidal machine (magnet or winding distribution) is driven by a sinusoidal signal, you can get almost perfect torque (FOC for example). If you use a trapezoidal machine and a sinusoidal signal, there are some regions where the motor is capable of producing 100% torque, but the signal commands as low as 86% (off the top of my head). The result is a torque ripple and noise. If that doesn't make sense, let me know and I'll see if I can draw some bEMF and current waves. (sorry for the OT!)
Jan 01, 2013, 05:56 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by Odysis If you have a sinusoidal magnet, the flux distribution will be sinusoidal, and so will the bEMF (roughly). In this case, it doesn't matter what the winding is. With a "square" rotor flux distrubution, caused by lumped magnets as normally seen in RC motors, you may or may not get sinusoidal bEMF. With concentrated, full pitch windings like we all use, the bEMF will be roughly trapezoidal. If you use a distributed winding scheme though, you can get a bEMF that's sinusoidal, even will a hobby rotor. It makes a few % difference in efficiency, but that's it. If a sinusoidal machine (magnet or winding distribution) is driven by a sinusoidal signal, you can get almost perfect torque (FOC for example). If you use a trapezoidal machine and a sinusoidal signal, there are some regions where the motor is capable of producing 100% torque, but the signal commands as low as 86% (off the top of my head). The result is a torque ripple and noise. If that doesn't make sense, let me know and I'll see if I can draw some bEMF and current waves. (sorry for the OT!)
How about sinosodial machine driven by trapezoidal signal. To be more specific, a sinosodial rotor, with concentrated winding, driven by RC ESC (trapezoidal signal), how is this combination?
Jan 01, 2013, 06:11 AM
like a rock!
Finland
Joined Oct 2007
2,530 Posts
Hi,

Quote:
 Originally Posted by Odysis ... ... With a "square" rotor flux distrubution, caused by lumped magnets as normally seen in RC motors, you may or may not get sinusoidal bEMF. With concentrated, full pitch windings like we all use, the bEMF will be roughly trapezoidal. If you use a distributed winding scheme though, you can get a bEMF that's sinusoidal, even will a hobby rotor. ,,,
would this explain why so many ESCs have difficulties with single-strand rewound motors?
If so, it might be interesting to hook up stock winding and rewound to an oscilloscope, drive with a power drill and look at the waveforms.
 Jan 01, 2013, 06:26 AM Life begins at transition Australia, VIC, Sale Joined May 2007 3,593 Posts A sine machine driven by a trap signal will have the same torque ripple / noise issue (and for the same reason). GK, I think the issue is more inductance-related rather than sine vs trap. I haven't seen a stock vs rewound bEMF trace. I can't think of why they'd be different, as most of the time we replace one CFP winding scheme with another CFP wind... I have been wrong in the past, and will be wrong in the future! Anyone got a scope?
 Jan 01, 2013, 06:53 AM like a rock! Finland Joined Oct 2007 2,530 Posts could be, I think they usually favor higher Kv (should mean less turns => lower inductance, right?) Just asking as re-wound motors are the rage right now, gotta have one so you have something to say in all those "why-does-my-motor-cut-out-in-flight" and "OMG my ESC is on fire" discussions BTW, speaking of ESC fires, I had a 4035 in my heli. One thread of the windings ripped out, and according to the burn marks inside the bell I think I got very lucky. Scorpion, never again ...
Jan 01, 2013, 10:19 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by gravityKills could be, I think they usually favor higher Kv (should mean less turns => lower inductance, right?) Just asking as re-wound motors are the rage right now, gotta have one so you have something to say in all those "why-does-my-motor-cut-out-in-flight" and "OMG my ESC is on fire" discussions BTW, speaking of ESC fires, I had a 4035 in my heli. One thread of the windings ripped out, and according to the burn marks inside the bell I think I got very lucky. Scorpion, never again ...
a normal scorpion's factory winding really is not good. But the hardware of a scorpion is good enough, light and strong design, plus good magnets. Scorpion is a very good re-winding candidate.

I just got a motor, 5030 stator 12N10P, made by a Chinese factory. Good enough for me. One thing I learnt is that however good i say the motor is, it does not count since i have no hard data or credit. I agreed to a technician of the factory that i would help contact someone has testing hardwares and credits here to help do a test of their motor. Here is some pic after i rewind it though.
rewind is using STSD, single strand 1.8mm wire(O.D around 1.9mm), 13T delta, Y, Kv around 250.
a similar scorpion would be S5030 12N14P, winded with 16T 15 strands of 0.33mm wire, weighted 606gram. mine is 660 gram, heavier due to slightly sicker rotor flux ring than scorpion, but also 61% more copper.

Thanks Odysis, for the explanation. It seems a sinosodial rotor will only be superior when driven by a sinosodial ESC. I have a few rewinded 1515 (O.D. and length in inch) sized (stator size) 12N8P inrunner with sinosodial 8P rotor sleeping here.

# Images

 Jan 05, 2013, 05:30 AM Aka: Tom Jenkins United States, FL, West Palm Beach Joined Aug 2008 4,698 Posts Ok! Its the new year, hope you all are happy! I have another project of a 4035 800 12n 8p to rewind. I need about 28 thousand solid rpm on 12s for a ducted fan. ABCABCABCABC I need input on how big of wire can be used and how many turns. I don't understand Modisc's plus two rule of thumb and don't know if I've done the XLS calculator correctly. PS: any suggestions as to how much timing and what PWM on a hv 160 for the 4035 550Kv wind? HNY Tom Last edited by ApexAero; Jan 05, 2013 at 05:42 AM.
Jan 05, 2013, 05:49 AM
Registered User
Antony (France)
Joined Sep 2003
3,380 Posts
Quote:
 Originally Posted by ApexAero Ok! Its the new year, hope you all are happy! I have another project of a 4035 800 12n 8p to rewind. I need about 28 thousand solid rpm on 12s for a ducted fan. ABCABCABCABC I need input on how big of wire can be used and how many turns. I don't understand Modisc's plus two rule of thumb and don't know if I've done the XLS calculator correctly. HNY Tom
Hi Tom
So your Kv could be around 630 rpm/V
If you have N1 turns for 800 rpm/V
you need to wind N1 x 800 / 630
For wire size the best is to try yourself on few teeth some AWG to fill the slots
Cu less area = old x 630/800
Cu diam = old SQRT (630/800)
Simple reasoning you could learn yourself for your future knowledge.

For a priori computing it is difficult as we don't know the magnets strength and airgap you have in the kit.
Louis
 Jan 05, 2013, 06:28 AM Registered User Delft, Netherlands Joined Jul 2001 239 Posts I did a 4035 8p wind in yy with 9turns. That got me 800rmp/V (18deg YGE setting) So if you want something in the 630 range: With 10turns delta you will get ~620rmp/V. With 11turns yy you will get ~650rpm/V.
Jan 05, 2013, 07:15 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by ApexAero Ok! Its the new year, hope you all are happy! I have another project of a 4035 800 12n 8p to rewind. I need about 28 thousand solid rpm on 12s for a ducted fan. ABCABCABCABC I need input on how big of wire can be used and how many turns. I don't understand Modisc's plus two rule of thumb and don't know if I've done the XLS calculator correctly. PS: any suggestions as to how much timing and what PWM on a hv 160 for the 4035 550Kv wind? HNY Tom
The scorpion's factory winding for their 12N8P 40XX motors are a bit different from a standard winding. For ABCABCABCABC, note that wire_A is jumping from tooth_1 to tooth_4, right? Normally, a standard winding will just have the "jumping wire" on the down side surface of the stator. However, for the scorpion's factory winding, the "jumping wire" will be hidden on tooth_2 and tooth_3. You see what i mean? So i found from my experience with 4 of my scorpion 40XX 12N8P motors, if it was a "8T" stated by the factory, then it should be viewed as "10T", when you do the Kv calculation as Louis posted above. Or you should follow Dekker's suggestion, since i believe he is using standard winding technique, no such hidden wires. For my 4035 12N8P. 6T delta give me 1028Kv, while 10T delta give me around 630Kv.

if Y termination give you Kv1000:
YY: 2000
D: 1732
DD: 3464

exact Kv depend on ESC, timing, some winding details~~.