New motors and esc's causing yaw issues... Help!

[jwk72]

I'm busy rebuilding a customers Discovery Pro and they decided to try new motors and esc's because the TBS 900kv's are out of stock. We settled on Lumier 900kv motors and the Castle 35amp quad pack. Those are the only parts replaced. We have build a ton of Discos with the exact same gains and they fly great. Now for some reason the Naza v2 has a significant yaw wobble. I've tried yaw gains from 60-150 with no luck. End points are set properly (castle esc's don't/can't calibrate). Any ideas???

 

[kloner]

take all your props off and fire it up, feel each arm for any vibrations, is there any?

 

[jwk72]

No vibrations, even downloaded a seismometer for my iPhone (saw someone else did that) and everything seems nice and smooth

 

[kloner]

If your confident enough in your skills, put one prop on a motor and remeasure, do one prop at a time to keep too many finger slicers from spinning while your holding it....

 

[jwk72]

Just curious, if a motor (props are well balanced) is out of balance and causing a vibration wouldn't it effect more than just yaw? Roll arms pitch work great

 

[kloner]

it will affect whatever axis it is fighting.... if your saying i've been doing 50mph for mile stretches and all i see is a yaw wobble is a whole different animal than i am hovering this new build and the yaw is wobbly so i guess it depends on when your seeing this, but yea, if a rig flies like crap and is un tuneable with gains, 99% of the time it is motor/prop related.

yaw problems would normaly be part of a motor alignment issue. Are you on Aerial mob arms or stock dji stuff?

 

[jwk72]

DJI arms but that is likely to change once the 400kv motors are back in stock.

 

[kloner]

is there a pwm setting on the castle esc's? if so you might find out what setting is working for others with those motors.... usually 12 i believe is middle timing and usually works on anything

 

[jwk72]

I'll have to hook it up and see. According to castle they are set up and ready to use out of box other that setting endpoints. They tell me these are nothing more that the Talon 35's that are pre programmed for MR use and missing bec's on 3 of the 4. I installed the Talons on a quad a while back and they were perfect. At first I was sure it was a gain issue I was having but not so sure anymore. From everything I e seen online there are very few people using gains outside of the range I've tried. I'll have to take a closer look at vibrations I guess.

 

[kloner]

if you feel any as you ramp the throttle up and down that is not normal.....

 

[jwk72]

Just did a few tests and nothing seems unusual. If I really try had to feel something, arm #2 may possibly have a a slight amount more that others but it would really be stretching it... So much so I'm not sure I'm not grasping at something that isn't there. Here are a few pics from the seismometer on my phone... Apparently the angle of the phone. May effect the axis reading
View attachment 20198
View attachment 20199
View attachment 20200
View attachment 20201

 

[kloner]

are you catching the part about revving the throttle up and down and feeling the full range of throttle for vibes in each one..... that's pretty big deal to feel anything there. if your showing a static picture it leads me to believe that's all your looking at, a certain rpm...

 

[kloner]

something else just came to mind, make tripple sure the motor bolts are not protruding into or touching the windings.

do you hear anything when it's doing it's wag? can you hear the motors revving or whistling or???

 

[jwk72]

Went back and double checked for vibration at all rpm's and nothing jumped out at me. Even compared it to another Disco with the TBS stuff and no noticeable difference.

No, there are no noises and the screws seem to be well clear of windings. I've got a call into Castle to see if they have any ideas

 

[cootertwo]

I had too long of motor screws once, and it drove me nuts, trying to figure out what was going on, until one motor threw out some sparks! I was blaming the ESC's. Got lucky, and didn't hurt anything. Shorter screws, and all was fine. An easy check, is with an ohm meter, I used the continuity setting that beeps if there's a short. Check between each screw and each motor wire, one at a time.

 

[jwk72]

I'll try a meter because I'm out of ideas but I doubt that's it. These are essentially the SMS motors as the TBS 900's on DJI arms with supplied screws... The screws are barely longer than the motor mount us thick. I've checked visually but I'll pull out the meter

 

[cootertwo]

Some motors have more clearance than others. I've only had the problem 1 time, out of 12 builds. But when chasing problems, you never know.:nevreness:

 

[kloner]

fly this in manual and either have somebody look or glance down at the throttle stick and tell me what the hover throttle position is.

 

[jwk72]

I will do that but first... After a 30 min call to Castle today (in which they say it is impossible to have anything to do with the esc's) getting nowhere I grabbed my multi meter as was recommended. For starters, they way the motors are laced out it is virtually impossible for the screws to make contact with the windings. You can actually see into the bottom of the motor and there is a good 1/8" of clearance everywhere. What I did find that is confusing is that motor #1 gives me a continuity tone with the probes one way but not swapped. I think I had red on the windings and and black on the motor leads. But when I swapped sides, no tone. Also, the Castle Quad Pack esc's only have 1 BEC and this is the one... Does this make sense?

 

[kloner]

so yea, thats a short. Can you get another motor? when there shorting, it's crazy. random things hapen, usually a notable whine but there untuneable from gains, fly like crap to even worse depending how many are shorting

you can pull the clip off the bottom of the motor and slip off the barrel, inspect the windings, maybe a piece of wire or a washer or a bolt is up in there.

on the motor in question, switch to Ohm on the multimeter and measure all the combos of the three motor wires,,,, see if the numbers are consistent