KDE vs T-Motor -- Picking a Motor Based on Spec Sheets

econfly

Member
I am in the early stages of building a copter to lift my MōVI M5. With solid knowledge of the weight of the M5 and cameras I plan to fly, as well as a general idea of all-up weight, I have started to look at motors. This is a description of that process and what I have concluded so far.

This examination is limited to KDE and T-Motor. I have had good luck with T-Motor in past builds. KDE stands out as a high quality option, and, as we will see, their specs are impressive.

In what may seem as a backwards way to do things, I've decided on using DJI's 1552 folding props in what will almost surely be an X8 build. I reviewed these props about a year ago and still find them appealing. They are efficient for a folding prop, durable, and very cost effective. The convenience of a folding prop is appealing, and these folders remain one of the best options out there. As a plus, rough estimations indicate a nice match between these folders and an X8 build for my needs. Finally, KDE offers a nice mounting option for these and other folding props.

Comparing motor specifications, particularly from one vendor to another, is not easy. Data points don't line up for easy comparison. Makers tend to report results for various throttle levels, and that means you can't easily compare efficiency (thrust vs. watts) at desired power levels. At the same time, reported specs are based on a given propeller, and one maker's "15 x 5" prop may not be the same one used by another.

Thankfully, both T-Motor and KDE report values for a series of motors using "15 x 5" props and 6s power. With a little work we can learn a lot about the competing options. The motors I am considering are:
To begin, I want to do a comparison that has nothing to do with motors. To make certain later comparisons are as fair as possible, I compare KDE and T-Motor values for thrust vs RPM. This is all about the prop used in the testing. Whether a given prop is spun by one motor or another, or a hamster on a wheel, the resulting thrust should be directly related to RPM. If KDE and T-Motor are using comparable "15 x 5" props in testing then we should see obvious similarity in reported thrust for any given RPM. Here are results I gathered from the published motor specs:

thrust_vs_rpm_zps1lszmufe.jpg

This is very encouraging. Over the six motors considered we see a nice curve of thrust vs. RPM. T-Motor is reporting very slightly higher values of thrust, but roughly speaking we can see that over all motors there is a fairly consistent relationship between thrust and RPM. This tells us that T-Motor and KDE are using very similar (if not the same) "15 x 5" props in their motor testing. Hence, we can compare specs among these motors without the complication of adjusting for differences in props under test.

The key consideration when picking a motor is thrust per watt within the desired thrust range. For any given level of power (watts) we want as much thrust as possible. Prop design can be a big factor, but we have eliminated that variable above. Here is thrust vs. watts for the six motors:

thrust_vs_watts_zpsofu55o6b.jpg

What to make of this? Over all six motors we see above a very big advantage for KDE. For any given level of power the KDE motors are reported to produce more thrust -- that means more efficiency and longer flight times. This difference is not trivial. I estimate that KDE is reporting about 20% more thrust over the entire thrust curve for their motors vs. T-Motor's reported values. And keep in mind this is not about the prop. In fact, above we found that, if anything, T-Motor is reporting a little bit more thrust for any given RPM. This huge efficiency advantage reported by KDE is based entirely on the motors.

To be very clear: This is not based on my testing. Everything here is coming from published motor specifications on KDE's and T-Motor's web pages linked above. The raw data I collected are below.

Based on these results I am leaning toward the KDE 4215XF-465. I may pick up a T-Motor U7-490 for a head-to-head comparison. Is KDE really this much better than T-Motor at designing and building a motor? Perhaps. If their specs are accurate KDE is producing some extremely impressive results.

Code:
Make    Motor    Watts    RPM    Thrust
T-Motor    U5-400     95      4,200      990
T-Motor    U5-400     175      5,200      1,490
T-Motor    U5-400     258      5,700      1,900
T-Motor    U5-400     322      6,200      2,220
T-Motor    U5-400     382      6,500      2,480
T-Motor    U7-420     135      4,500      1,160
T-Motor    U7-420     260      5,700      1,870
T-Motor    U7-420     340      6,300      2,260
T-Motor    U7-420     462      6,850      2,690
T-Motor    U7-420     548      7,250      3,030
T-Motor    U7-490     206      5,300      1,600
T-Motor    U7-490     400      6,600      2,400
T-Motor    U7-490     568      7,500      3,060
T-Motor    U7-490     739      8,100      3,680
T-Motor    U7-490     884      8,500      4,100
KDE    4014XF-380     29      2,940      430
KDE    4014XF-380     53      3,630      700
KDE    4014XF-380     95      4,560      1,110
KDE    4014XF-380     143      5,280      1,500
KDE    4014XF-380     205      6,000      1,930
KDE    4014XF-380     279      6,660      2,360
KDE    4014XF-380     359      7,320      2,820
KDE    4012XF-400     27      2,880      430
KDE    4012XF-400     57      3,780      770
KDE    4012XF-400     109      4,800      1,270
KDE    4012XF-400     159      5,460      1,600
KDE    4012XF-400     239      6,240      2,090
KDE    4012XF-400     317      6,900      2,520
KDE    4012XF-400     401      7,440      3,020
KDE    4215XF-465     65      3,885      800
KDE    4215XF-465     119      4,860      1,260
KDE    4215XF-465     197      5,915      1,860
KDE    4215XF-465     317      6,900      2,460
KDE    4215XF-465     452      7,850      3,190
KDE    4215XF-465     632      8,820      4,040
KDE    4215XF-465     823      9,470      4,810


UPDATE - TESTING COMPLETE

Here is the comparison of thrust vs. watts for the KDE 4215XF-465 and T-Motor U7-490. This is with the KDE folding prop bracket, DJI 1552 15 x 5 props, and recommended ESCs (KDE 55+ for the 4215, and T-Motor T80A for the U7-490).

In sum, these motors are extremely close in performance using the 15 x 5 folders. KDE's efficiency claims are absolutely nonsense. Yet, their motor does a very nice job, and only yields slightly to the larger U7 at the high end of the thrust curve.

Superficially, the U7 is massive compared to the KDE 4215. It and its ESC are about 65 grams heavier than the KDE system. The U7 is 4mm taller, and 12-13mm wider.

Both motors appear to be very well constructed. I may post a teardown later if I can find time.

I found the KDE motor/ESC combination to be much more responsive in my tests. Dialing in PWM throttle values resulted in nearly instant reaction for the KDE motor while the U7 motor/ESC system had a slower response and tended to ramp up/down to changes in throttle signal.

For my intended use -- 15 or 16 inch props and all-up weight around 12-14 kg in an X8 setup -- both of these motors provide plenty of thrust and should result in a hover right at 50% throttle.

Both motors sell for right around $150. Given the nearly identical thrust performance with 15" props, for my purposes the superior response of the KDE motor along with its more compact size make it the likely winner.

However, I am very unhappy with KDE's specifications for the motor. As can be seen in the graph, T-Motor provides very accurate and practical information on their motor's performance. KDE, on the other hand, does not provide accurate information.

I calibrated my thrust meter prior to testing. My power measurement process is very accurate. Yet even if my process were flawed, shifting performance one way or another, both motors were tested equally. Despite KDE's claimed performance advantage, both motors performed similarly and right along the path provided by T-Motor's specs.
comparison_zpsxcv7shbf.jpg
 
Last edited by a moderator:


hjls3

Member
I was early adopter of KDE and have to say - i am very skeptical of their reported numbers. just throwing that out there. Nice write up.
 

econfly

Member
I will follow up with a head-to-head comparison of the KDE 4215XF-465 and T-Motor U7-490. I will order the recommended ESCs for each as well (KDE 55A+, and T-Motor T80A), and compare performance with the 15-inch folding props.
 

Carapau

Tek care, lambs ont road, MRF Moderator
I did a replication of the KDE response test on the bench. I took a KDE 5215 435kv motor and KDE ESC and compared it to a U7 420kv with T-Motor ESC. Just as in the KDE video, the KDE setup was MUCH more responsive (both using 18x6.1 Tiger props). Not a particularly scientific test for sure but in my mind I will currently take a KDE setup over T-Motor any day.
 

fltundra

Member
I will follow up with a head-to-head comparison of the KDE 4215XF-465 and T-Motor U7-490. I will order the recommended ESCs for each as well (KDE 55A+, and T-Motor T80A), and compare performance with the 15-inch folding props.
I would only use the KDE 75's or the new Tiger 80's with the 4215xf-465's. The 55's aren't big enough to handle those motors. Just ask pumpkin!
 

fltundra

Member
I will currently take a KDE setup over T-Motor any day.
Same here. I have over 70 hrs flight time with my 515's, and absolutely no bearing play. If I were still running tigers, i would already have to plan on bearing replacement soon. Tigers recommends changing at 80 hrs. KDE 300 hrs.
 

econfly

Member
I would only use the KDE 75's or the new Tiger 80's with the 4215xf-465's. The 55's aren't big enough to handle those motors. Just ask pumpkin!

I saw his thread on the power consumption. I would run 15-inch props with these motors (he had heat issues with 18" props). According to KDE's specs the 4215XF-465 motor with a 15 x 5 prop maxes out at about 825 watts (37A) at 6S. KDE recommends their 55A+ ESC for this motor. Given all of that, would you still go with a larger ESC?
 



econfly

Member
My concern is this: If I can't trust KDE to spec a correct ESC for their own motor, particularly when my intended use is at the low end of the operating current range, then I wonder whether I can trust anything they say. Take your point above about bearing life vs. T-Motor. That's a very complex thing to measure and forecast, while ESC temperature measurement and load handling is fairly simple by comparison. If they can't spec a correct ESC, why in the world would I trust them to spec bearing life?
 
Last edited by a moderator:

fltundra

Member
My concern is this: If I can't trust KDE to spec a correct ESC for their own motor, particularly when my intended use is at the low end of the operating current range, then I wonder whether I can trust anything they say. Take your point above about bearing life vs. T-Motor. That's a very complex thing to measure and forecast, while ESC temperature measurement and load handling is fairly simple by comparison. If they can't spec a correct ESC, why in the world would I trust them to spec bearing life?
The 55's would be absolutely fine if your only running them on a 4S. 6S and above I would not chance It. I just went threw this with my 515's a year ago. Originally running them on Castle 25's and max draw with 14x4.8's on each was around 4 amps and they would always run warm, as in 100 to 105. Same setup with Castle 35's, motors are more responsive and esc's run cool to the touch, and I am only drawing 17 amps at hover with gimbal, tx, and strobes. Also, the 515's are rated at 25+amps. The 465's are 62+ amps. That's why I'm saying don't undersize you esc's .If I were to move up to 6s or larger camera and gimbal I would probably go with the 55's with my motors. Just to have the extra margin of safety.
 

econfly

Member
I guess I don't understand how the ESCs could be undersized if (a) KDE says to use them, and (b) KDE says the motors pull, at most, 823 watts / 37A at 6s with a 15 x 5 prop, and (c) the ESC is rated to 2S - 8S, 55A+, and 1,630 watts. I would be pulling, at most, about half the watts and two-thirds of the amps the ESC is spec'd for, and that's at 100% throttle (which will never happen). I'm guessing I'll need maybe 275 watts / 12A at most to hover, and likely no more than double that at peak.

On top of all of this, KDE's videos really emphasize how wonderfully their ESCs and motors work together to give us that rapid response. If I move to a T-Motor or Castle ESC will I get the same performance?

Don't get me wrong -- you may well be right. I'm concerned, however, when I hear that KDE makes the best motors (a very technical and complex engineering and manufacturing task), but also that somehow they can't figure out how to tell us which ESC to use.

I'm really looking forward to getting the KDE and T-Motor samples in my hands to test. Hopefully will get to it this week.
 

Motopreserve

Drone Enthusiast
It sounds like the math proves the point, no? I can't imagine that KDE is doing the math wrong for their products, but even if they were - YOU are not. :)
 

jfro

Aerial Fun
Just curious, what's your targeted AUW for this and what extra are you hoping to have for future use?
 

econfly

Member
Aiming at around 12kg, with room to go to maybe 14 or so. No firm decisions yet.

For a while now I've wanted a compact X8 with lots of lift and relatively high RPM to really handle the wind. That's the motivation for my leaning toward high rpm motors with relatively small props. Having said that, the motors here all spec out 15-inch props at 6S and the first graph in my original post shows how they just fill in various parts of a thrust/rpm curve. If we look at efficiency (thrust per watt) the Tigers are on one curve and the KDEs are on a uniformly better curve (we'll see about that).

So, again, it's just about picking the motor that covers the thrust range I want.

I've decided on a 950mm Vulcan black widow frame. That gets me X8 with 15 or 16 inch props. For my uses the 15s look like a nice fit. I can always move to 16s if needed.

So far, that's all I have. Plenty more decision making (and backtracking) is still possible.
 

Motopreserve

Drone Enthusiast
That's a nice choice rob. I think the Vulcan has been one of my best choices in this hobby. But of course now I want foldable :).

Wondering if I could mill out the plates to allow the standard style to fold.

Looking forward to seeing your build.
 

econfly

Member
I ordered the folding "black widow" X8 frame with the quick latches (all 4 arms are "mantis"). Also a quick release set of retracts. This thing should travel well. Will do a full writeup of the build. But first I need to get this motor issue solved...
 

Motopreserve

Drone Enthusiast
It's a sweet frame. Really need the folding arms. The damn thing has to be transported in the back of my pickup :)

I have the KDE 4012 and have been very happy with them.

I'm sure you've calculated it - but I know the X configuration with Z arms will effect the prop size allowed for the lower motors on those booms.
 

econfly

Member
It's a sweet frame. Really need the folding arms. The damn thing has to be transported in the back of my pickup :)

I have the KDE 4012 and have been very happy with them.

I'm sure you've calculated it - but I know the X configuration with Z arms will effect the prop size allowed for the lower motors on those booms.

Yes, it's a 16-inch limit on the props with the 950mm Z arms.
 

Top