Originally Posted by Aeroplayin

A closer look at the the match-up in the data chart suggests a 515Kv.

8400 RPM / 16.3v = 515.34 RPM/v

It would be interesting to see how that compares with the drill press Kv test.

It would be nice to measure reverse EMF, but for me, it's always interesting how any testing process relates to Kv_Kt or the true relationship between load, torque, and Amps.

For me, there is nothing more important because it reveals how higher and higher loads affect efficiency and performance. Think of it in terms of a gas engine on a dynamometer, which simultaneously measures torque and rotational speed.

For example, this motor spins unloaded to ~515.34 RPM/v taken from a Castle log on Tom's unloaded test, and spins to ~517.26 RPM/v unloaded on Rob's test using equipment that makes it somewhat difficult to match RPM directly with voltage and Amps simultaneously.

It then spins to 5910 RPM on 42.6v with a 28x12 PJN load to 173A. From this, we can then deduce the following:

If we believe the Amps to be 173 and the Kv to be 517, then the RPM must be 5940 because this is where the torque of a 28x12 is 1508.5 oz-in

Kv = 173A / (1508.5 / 1352.4)
Kv = 155.1
(155.1 * 60T) / 18T = 517Kv

So 5940 RPM is where 517Kv and 173A are in mathematical bliss.

This tells us that the efficiency of the system is 89.9% because the unloaded RPM should be (42.6v * 517Kv) or 22,024 RPM at the motor shaft, and 6607 RPM at the prop shaft, so 5940 RPM / 6607 RPM is 89.9%.

This is proven by the calculated torque and Watts-out, which must also be in sync for the formula to work.

173A times 42.6v is 7370 Watt-in
7370 Watts-in times 89.9% is 6625.6 Watts-out
6625.6 Watts-out / 745.7 = 8.885 HP
(8.885 HP *5252 RPM) / 5940 RPM = 7.857 lb-ft of torque where 5252 RPM is the HP_T constant
7.857 lb-ft of torque is 1508.5 oz-in

....which is the value used to create the Kv_Kt sync above.

Now we can also calculate for a 515Kv at 5910 RPM but then the Amps would be 171.
I can also raise the efficiency to 91% but the Kv would be 510 at 42.6v and the Amps would be 170.

So the way to tell for sure is to feed a constant voltage source to the system and record the RPM as accurately as possible as we systematically increase load, similar to what we do on a dyno. This is a better way to get an accurate understanding of Kv and efficiency, which in the end separates the better motors.

So for me and most 3D enthusiasts, the import thing is to be able to see how much load the motor can take and still perform where the Amps still calculate to a value relative to the unloaded Kv without sagging the source voltage too much, keep the temp range real, and not affecting the throttle response too much.

So this motor ran to 517Kv unloaded, then 517Kv loaded with a 178g 28x12 electric beechwood prop, and the video shows a crisp throttle response without too much additional current from the pulsing. It would be nice to also manage the voltage sag a little better because the performance will improve if we can maintain nominal voltage a little longer. The key here could be battery capacity to reach somewhere around a 20C discharge at 173A -- or 9Ah. But sometimes the motor mass and magnetic field just can't make it happen and I'm beginning to think we've reached the limit of these 750-770 gram motors... just a bit under 9 HP.

Hi Tom, Jim, Jack and others
I try to summarize the rewinding (for my personal reference)

Base: Scorpion HKIII 5035-xxx
Windings: 4 turns per arm = 8 conductors per slot
turns made of 2 strands 1.2mm wire
so 16 wires per slot, 8 wires per arm (picture)

Kv 515-517 rpm/V
But could you confirm (Tom) that connectings are YY (or DD or any) ? Am I right ?
Best
Louis
PS note that Scorpion Calc is now v3.70 (march 2019)

That certainly should be a more accurate Kv estimate.

Many are not aware of the difference but Castle's ICE data only collects data on what is going on in the ESC. This thread has a discussion of those considerations:

Castle ICE logging oddity - Neu vs HET motor test - www.rcgroups.com/forums/showthread.php?t=1995693

And Castle Creations themselves discussed the differences between their data and power system watt meter in this discussion:

https://www.rcgroups.com/forums/show...n#post19933161

Some have reacted pretty harshly to the way the CC data is viewed and when it is compared to watt meters but the bottom line is that CC built an ESC data logger into their ESCs and not a power system data logger.

But the RPM and input voltage would vary very little between the readings on an ESC and the data reported by the ESC.

As always, the cost of a power system data logger and the challenge of learning how to implement it's use scares some off. But for those on a tight budget and interested in getting some accurate readings and understanding them, the $3 Watt Meter thread gives us a great lesson on collecting power system data inexpensively:

$3 Wattmeter - https://www.rcgroups.com/forums/showthread.php?949923

If you have an inexpensive multimeter and a short length of copper wire you can be off and running on better understanding your power system.

Jack

The 515Kv was from the Castle log and the 517Kv was from meters and tachs on the same motor, and my calculation were off the same motor under load.

So if Castle is off, it's not by much in this case. And I've always used meters and tachs to confirm log data too and there has been times where Castle has been off by as much as 9%, using the math to confirm the meter more accurate versus the log.

But I've done over 150 load tests on over 50 different motors so the amount of time the log has been off suggests it's not too bad. But I always suggest to use the math to confirm.

In my line of work, you can't defend anything without the math confirming it.

Fordan, yes YY termination

Rob flew the 515Kv this weekend at the event in Tennessee and he was nice enough to let a few very talented pilots fly his plane.

The consensus was that the power was very solid, so I now have a few questions going forward. My goal is to assess the remaining 5035 Kv options so that I can mathematically match them with the propellers that make the most sense for the type of 3D and IMAC performance we want to see with the StinGR on various plane sizes

We will also continue to assess the 541Kv motor. One of the reasons why I asked Tom to send it, and not try to rewind it first, was because there are a few prop options that make a lot of sense with it. We'll know more in a few weeks since some of the props I will be testing are on back order.

All in all, the project is going well, and the winds that Tom did for us are useful. In the meantime, I have a few novice questions, so pardon my ignorance when it comes to motor rewinding.

[1] Can the 541 and 515 be easily duplicated? I ask this question because it seemed that the unloaded RPM/v may have been a surprise. Scorpion suggests that a 4T + 4T YY is a 540Kv so my guess is that at least the 540Kv can become a mark that is easily duplicated. (?)

[2] Since it seems that the only motor manufacturer that I have had a lot of experience with has always been able to hit every Kv within a certain range in 5 Kv increments with 14 pole motors, why does this 10 pole motor seem to have a 26 Kv increments? I may be completely off about this 26 value, but it would help me understand my options if there was some linear relationship -- or a defined non-linear relationship -- to the Kv options.

[3] Is it possible that a list for this type of winding -- YY with the thicker copper gauge for more cooling and efficiency -- could be 541, 515, 489, 463, and 437 for this motor with this type of wind configuration? I can see from Dekker's chart that when Tom thought he had hit the 503Kv correctly, it tested to 515 RPM/v. So perhaps the 539 Kv from the chart is really the 541 RPM/v, which would suggest that the 429 Kv could test to 437 RPM/v and the 451 Kv could test to 463 RPM/v.

I must add here that Tom's Castle log data matched the meter and tach readings so let's not debate the validity of the log readings right now.

This is all speculation of ignorance, of course, but you can see what I'm trying to do.... reach specific performance levels with known propeller performances and known reduction ratios with unknown custom windings. So the missing link to a complete interrelationship is known Kvs.

If the 541 and 515 can be duplicated, great. All I would need to know is if something between 460 and 465 was possible since 451 is too light and 476 is too hot for the third voltage and prop applications I have in mind.

But if not, I'd like to know what the options really are that can be accurately duplicated.

Hello Jim
Here under some informations regarding "rewinding"
Given a stock motor (example Scorpion HK III 5035)
a) Initial (factory) windings
12N10P dLRK
Kv (announced) 500 rpm/V
3+4 turns on two consecutive arms = 3.5 turns per arm = 7 conductors per slot
29 strands 0.3mm
Rph-ph = 0.008 ohm (announced)

b) Rewinding choices
Parameter number of turns
Wire type (strands and diam)
dLRK D
dLRK Y
dLRK DD
dlRK YY
LRK D
LRK Y
LRK DD
LRK YY
All the rewinding modes are related with some relations concerning the "product" turns x Kv
There is some freedom but the "steps of Kv" are "complicated"
A freeware like Turn-Calculator (v8) is a small tool (EXCEL) playing with the Kv vs modes and number of turns

If you consider a "base" of 7 conductors per slot dLRK D, you can change
* the windings mode
* the number 7 (3+4) keeping an INTEGER (n1+n2) ... if n2=0 you get a LRK
It is convenient to use the wording "assymmetric dLRK" if n1 is different of n2

Note that Scorpion generaly is giving the number per slot and not the number of turns per arm
The Kv value (rpm/V) is normaly a value @ zero current, taking into account Rph-ph * Io
The raw Kv (no load) is close and defined as rpm0/V
There is a small spreading of measured Kv (instruments accuracy, battery voltage, magnet strength ...)
Not interesting to give a Kv with decimals
The drill method for Kv measuring suffers also of some details (harmonics, compensating coefficient 0.95 ...)
Louis

Thanks, Louis.

But I'm still not sure about the answers to my questions about predicting the available results in RPM/v for this 5035 motor with a YY 2-strand wire surface >1. Are you saying that if I download the Turn-Calculator v8, the results of the winding choices leading to the RPM/v I listed above will be made accurately available in the spreadsheet?

I understand that this is not that simple, but factory winds seem to be accurate at least most of the time. I can ask Ken for a specific wind for his hand wound Moltrofly motors and he very rarely misses the mark when I test it unloaded. I've had nine motors from him that I can think of and four were custom requests.

I have seen factory Hacker 14 pole motors off as much as much as 17Kv on occasion making one particular motor useless for the 10S with a 22x10 PJN config I had in mind. The correct Kv would have been perfect but because the Kv was off so much, I had no choice but to sell it when Hacker USA told me what the options were and how long it would take.

This whole exercise is not only to get a more efficient wind, but to target the three or four Kv options I will need as closely as possible. All I ask for is that the motor tests close to the unloaded RPM I'm looking for, and in this case, I'm looking for 539-541, 515-517, 488-490, and 463-465 RPM/v.

Originally Posted by Aeroplayin

Rob flew the 515Kv this weekend at the event in Tennessee and he was nice enough to let a few very talented pilots fly his plane.

The consensus was that the power was very solid...

The power was very well received by the four pilots that had the opportunity to test the system out. The overall power is not at the level of a 120cc gas engine but reasonably close...and everyone was amazed how quiet and smooth the system was. I was able to capture video from three of them.
Amazing flying by these three considering it was their one flight on the plane - not always easy considering everyone has different radio and plane setup preferences. I was happy that the electric setup felt pretty natural to these pilots that normally fly 120cc gas.

Electrified AJ 105 Laser StingGR Alex Barron (5 min 29 sec)

Electrified AJ 105 Laser StingGR Mark Czechalien (5 min 19 sec)

Electrified AJ 105 Laser StingGR Justin Owen (6 min 55 sec)

Hello Jim
I have run Turn Calculator 7.3 (attached zipped)
Options :
9t YY (4+5) --> 449 rpm/V
15t DD (7+8) --> 467 rpm/V
8t LRK Y (8+0) --> 481 rpm/V
7t dLRK D (3+4) --> 500 rpm/V (stock)
4t dLRK Y (2+2) --> 505 rpm/V
13t LRK D (13+0) --> 513 rpm/V
13t DD (6+7) --> 538 rpm/V
7t LRK Y (7+0) --> 550 rpm/V
But caution ...
The theoretical computing to pass from dLRK to LRK is not perfectly sure (+10% is not warranted)
There could be some uncertainty on the Kv values (in the real world)
Best
Louis

Thanks again, Louis. I appreciate your help and will review this file tonight.

I picked up some 13 ga (1.7 mm stripped) for another wire option.
Out of all the termination options the YY may be the best at taking the amperage.
I'm thinking a 9T(4+5) may be the ticket as the rewinds seem to come in a bit higher than the Turn Calculators to hit the 460 Kv mark.
Also thinking the Kv difference in the previous motor winds my be due to Scorpion's magnet choice at the time of manufacture.

Nice job this weekend Rob!
Waiting on your decision Jim!

The US distributor said they should have the props by the 28th. I think the best way to do this is to try the 28x13 on the factory 500Kv first. You can then retest on your Castle log so we can at least have a relative comparison, then you can try to get as close to an unloaded RPM/v of 460 as possible.

The 460Kv will be very useful because it's not only the custom power I need for 28 lbs planes in IMAC competition with a 28x13 on 12S, but it keeps the motor speed down while still giving me a very high performance range for hard core 3D on 30 lb planes on 14S with the inexpensive 28x12 PJN while keeping the current down.

Now let's see if the math works out as well as it did matching props to the two resulting Kv's the first time.

Pretty sure a HPdLRK qualify as a DD termination. Trying to calculate Kv options.