Dec 29, 2012, 03:53 AM Aka: Tom Jenkins United States, FL, West Palm Beach Joined Aug 2008 4,732 Posts Well if I average the two it fits perfect to Modisc's predictions of 570.
 Dec 29, 2012, 04:20 AM Registered User Antony (France) Joined Sep 2003 3,382 Posts Hi I agree 100% with Modisc I have tested more than 300 BLDC motors, varying V. Kv stays in a very small fork (generally < 1%) Louis
Dec 29, 2012, 04:53 AM
Registered User
Belgium, Flemish Region, Oosterzele
Joined Dec 2001
5,646 Posts
Quote:
 Originally Posted by modisc I forgot to talk about the tenique of how to achieve the best copper fill, after you have choosen the wire size. I am gonna label the 12 teeth of the stator as from T1 to T12. I will be using single-tooth-single-direction winding technique, so each tooth is winded with one wire of appropriate length, which you will be able to estimate since you know how many turns will be eventrully on that tooth. 1, Finish the 1st layer on T1. 2, Finish the 1st layer on T3 3, Finish the 1st layer on T2. Then do the 2nd layer on T2. 4, Finish the 1st layer on T5 5, Finish the 1st later on T4, Then do the 2nd layer on T4 6, Do the 2nd layer on T3, if possible. Or maybe there is no more room for a 2nd later on T3, which is commonly the case. Then keep going like this... ... 1，the 2010 year limited edition of HK4225 in the market，12N8P，single strand of 1.3mm，10T（which is actually 12T），delta，610Kv， Copper fill/copper cross section area in one slot：15.72mm^2（100.00%） （http://www.hobbyhut.com/Scorpion_HK_...10.htm） 2，Powercroco's HK4225 ，12N10P，single strand 1.32mm,6+7T，YY，541Kv Copper fill：17.78mm^2（113.10%） （http://www.scorpion.powercroco.de/12N10P4225Kit.html） 3，Powercroco's HK4225，XXXX，single strand 1.5mm,5+6T，X，XXX Copper fill：19.42mm^2（123.54%） （http://www.scorpion.powercroco.de/12N10P4225Kit.html） 4，My HK4225,12N8P，single strand 1.45mm，6+7T，delta，XXXKv（585？） Copper fill：21.45mm^2（136,45%） By the way, i personally donot prefer LRK on this sized motor. One main reason is that LRK will surely contain multiple layers of winding (at least 3 layers), which does not look good to me. Another reason is that LRK will have less copper fill comparing with dLRK, by my winding technique and from my experience. I am sure that there are other opinions about this issue, but to note that I am always talking about approaching the best possible copper fill. So it is possible for LRK to have better copper fill than dLRK when using 0.6mm wire on this motor; but the copper fill achieved by using 0.6mm wire will not be even close to what i can achieve using 1.4mm wire. Using the above 4225 for example, using LRK will only get 12 turns or even less in each slot, and the coils will not look pretty; while dLRK can have 13T.
well, this is 4020 Scorpion stator I did myself , 13*1,32 mm LRK (DD):

I've had it rewound to 14 turns 1.4 mm, LRK DD (left),and 2*LRK (right) ( for use with 2 separate controllers), by a more experienced "winder" :

I calculate those new winds to about 21.5 mm2 copper fill...

https://www.rcgroups.com/forums/show...1486613&page=4

Impressive wind for a first try though... Oh, and here is what I lost on those first tries :

 Dec 29, 2012, 05:26 AM Aka: Tom Jenkins United States, FL, West Palm Beach Joined Aug 2008 4,732 Posts So it must be the tach, its the only other variable.
Dec 29, 2012, 06:34 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by WimH well, this is 4020 Scorpion stator I did myself , 13*1,32 mm LRK (DD): I've had it rewound to 14 turns 1.4 mm, LRK DD (left),and 2*LRK (right) ( for use with 2 separate controllers), by a more experienced "winder" : I calculate those new winds to about 21.5 mm2 copper fill... See this thread : https://www.rcgroups.com/forums/show...1486613&page=4 Impressive wind for a first try though... Oh, and here is what I lost on those first tries :
Good work on LRK, it seems you are another LRK fan ^^.

I personally do not prefer LRK cause from my personal experience, LRK can not have more copper fill than a dLRK for large wires. BUt this is just personal experience, and there are others having different opinion.

by looking at the stator with coils, for your double LRK and standard LRK , and comparing with my 4225 stator with "my 1.4mm wire" (as shown in the attachment pic), it seems your wire is "a bit smaller" than 1.4mm.

# Images

Dec 29, 2012, 06:47 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by WimH I've had it rewound to 14 turns 1.4 mm, LRK DD (left),and 2*LRK (right) ( for use with 2 separate controllers), by a more experienced "winder" :
I just used Powerpoint to do a rough comparison for clear demonstration:
For 42XX vs 40XX, 40XX's stator diameter is 95% of 42XX, so i reduced the size of your image a little bit.
All three pic is a square, just big enough to contain the stator's cross section view.

For my 4225, 5 turns of 1.4mm wire (outer diameter around 1.5mm) is basically the same as the stator tooth length. For your 40XX, 6 turns of "1.4mm" is slightly bigger than the stator tooth length. To note that the "red line" in the pics are exactly the same length, as i used "copy and paste". Very clear.

# Images

Dec 29, 2012, 07:07 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by Fourdan Hi I agree 100% with Modisc I have tested more than 300 BLDC motors, varying V. Kv stays in a very small fork (generally < 1%) Louis
And this is well explained by the Kv formula provided in the leading thread in this forum ^^

Louis, have you noticed significant variation in Kv when using different ESC to test the same motor?

Another issue that I am not sure now, is that whether ESC has a low-speed limit? I Used a Hobbywing ESC trying to drive a servo motor with original coil and connection, Kv of the motor being 15rpm/v. The motor cannot start at 6S voltage.
Dec 29, 2012, 07:16 AM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by ApexAero Well if I average the two it fits perfect to Modisc's predictions of 570.
I nearly forgot when i predicted that ?

Since Kv = actual rpm / (Input voltage - resistance of motor X current), and we know the resistance of the motor is basically a constant, and the no-load running current is directly related to input voltage. So I suggest it would probably be more accurate if you use as low voltage as possible (as supported by your ESC) to do a no load run and roughly calculate Kv by directly using (actual rpm tested by a tacho, divided by input voltage tested by a multimeter while motor is running).

I believe some elogger can measure and show Kv as well as other values, but i do not have such a luxury. I used computer/microphone based software to do the kv test which seems to work well, but i start to feel lazy now.
Dec 29, 2012, 07:36 AM
Registered User
Antony (France)
Joined Sep 2003
3,382 Posts
Quote:
 Originally Posted by modisc Louis, have you noticed significant variation in Kv when using different ESC to test the same motor? Another issue that I am not sure now, is that whether ESC has a low-speed limit? I Used a Hobbywing ESC trying to drive a servo motor with original coil and connection, Kv of the motor being 15rpm/v. The motor cannot start at 6S voltage.
Hi
Yes I observed some variations of Kv depending of the ESC
** Brand
** Timing
Also it happens that some ESC are not starting well if voltage is too low.
After starting-running some ESC could go down in voltage or PWM to some
low speed limit.
SLS ESC are very good to go low speed (FOC algorithms with a lot of DSP)
Louis
Dec 29, 2012, 10:33 AM
Aka: Tom Jenkins
United States, FL, West Palm Beach
Joined Aug 2008
4,732 Posts
What do you think of these?

# Images

Dec 29, 2012, 03:17 PM
Aka: Tom Jenkins
United States, FL, West Palm Beach
Joined Aug 2008
4,732 Posts
Well I go off my tail, got the 160 hv connected. Looks like the Kv is around 550 + - resulting from two tests on 3s and 5s

# Images

Last edited by ApexAero; Dec 29, 2012 at 03:31 PM.
Dec 30, 2012, 10:55 AM
Registered User
Delft, Netherlands
Joined Jul 2001
240 Posts
Quote:
 Originally Posted by modisc I just used Powerpoint to do a rough comparison for clear demonstration: For 42XX vs 40XX, 40XX's stator diameter is 95% of 42XX, so i reduced the size of your image a little bit. All three pic is a square, just big enough to contain the stator's cross section view. For my 4225, 5 turns of 1.4mm wire (outer diameter around 1.5mm) is basically the same as the stator tooth length. For your 40XX, 6 turns of "1.4mm" is slightly bigger than the stator tooth length. To note that the "red line" in the pics are exactly the same length, as i used "copy and paste". Very clear.
I am the winder of WimH's motors and I can assure you the wire is 1.4mm: in fact it measures 1.48mm with insulation...
Reducing the size of images is not the way to go: the 42xx and 40xx stators both have about the same slot surface, so in both cases, 14wires of 1.4mm will fit.
Actually I found winding the "pure" LRK easier, the double LRK was a tighter fit because you have to do some wire transitions that prohibits good nesting of the wires.
Just to be clear, here is a 4225 12N10P I did, this is 13turns, with the same 1.4mm wire:

On this stator there is 5+5turns in the first layer like you also posted for your motor.
The only reason the 14th wire wont fit is because of the crossover between each second tooth due to the dLRK winding scheme.
Last edited by Dekker; Dec 30, 2012 at 11:10 AM.
Dec 30, 2012, 07:26 PM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by Dekker I am the winder of WimH's motors and I can assure you the wire is 1.4mm: in fact it measures 1.48mm with insulation... Reducing the size of images is not the way to go: the 42xx and 40xx stators both have about the same slot surface, so in both cases, 14wires of 1.4mm will fit. Actually I found winding the "pure" LRK easier, the double LRK was a tighter fit because you have to do some wire transitions that prohibits good nesting of the wires. Just to be clear, here is a 4225 12N10P I did, this is 13turns, with the same 1.4mm wire: On this stator there is 5+5turns in the first layer like you also posted for your motor. The only reason the 14th wire wont fit is because of the crossover between each second tooth due to the dLRK winding scheme.
Great work.

If we have both 1.4mm wire, and assuming you are using 0.2mm DM insulation paper while I am using 0.125mm kapton for this 4225, I can only think of that my wire is not as good as yours and have thicker insulation layer. :P

By the way, my 4225 is not a perfect 5+5 for first layer. every other tooth has 5T perfect first layer, but the other 6 teeth in between, has 4T first layer + 1 outer turn that is slightly above the 1st layer, since 2 turns of my "1.4" can not be placed parallel along the inner side of the slot.

# Images

Dec 30, 2012, 07:35 PM
Inspiration
United States, MI, Ann Arbor
Joined Aug 2011
895 Posts
Quote:
 Originally Posted by Fourdan Hi Yes I observed some variations of Kv depending of the ESC ** Brand ** Timing Also it happens that some ESC are not starting well if voltage is too low. After starting-running some ESC could go down in voltage or PWM to some low speed limit. SLS ESC are very good to go low speed (FOC algorithms with a lot of DSP) Louis
Thank you Louis. Though this is a bit off the topic, but do you have any idea how servo motor driver can make the 3-phase motor running at constant torque even at low speed, while when using RC ESC, when reducing the speed by turning down the throttle, the torque is also reduced. Is this due to the difference in design of the 2 types of brushless driver?

Also, is this why high speed motor + gear can be a good combination, cause this gearing system kind of "reduce' the effect of reduced torque when turning down the throttle. Is this why high speed motor + gear may have better acceleration than using low speed motor for direct drive?

Thanks
 Dec 31, 2012, 03:43 AM Registered User Antony (France) Joined Sep 2003 3,382 Posts 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