Different Yaw Speed for clockwise and counter clockwise

Post Reply
rukie
Posts: 8
Joined: Mon Jun 09, 2014 3:16 am

Different Yaw Speed for clockwise and counter clockwise

Post by rukie »

Hi Guys,

I currently finished a build with the following components.

DJI f450 clone
Witespy EV3 Blackboard
Sunnsysky 2216 900kv
Rctimer 30amps esc

I noticed as said in the subject that my yaw speed differs from the two directions. My clockwise seems to be sluggish even feels that its struggling to turn clockwise. This board was transferred from a bigger quad with different components and the same behaviour can be observed. I do think that I have everything ok mechanically on the build. I think I have to tweak something in the settings or code. Appreciate if you can point me to the right direction. Thanks,

jim lee
Posts: 16
Joined: Tue Apr 22, 2014 10:09 pm

Re: Different Yaw Speed for clockwise and counter clockwise

Post by jim lee »

I see the same thing on my Quad. (3lbs of aircraft) Doesn't like turning right so much, turns to the left much better.

-jim lee

waltr
Posts: 733
Joined: Wed Jan 22, 2014 3:21 pm
Location: Near Philadelphia, Pennsyvania, USA

Re: Different Yaw Speed for clockwise and counter clockwise

Post by waltr »

Re-check that all motors are true and square to the frame.

rukie
Posts: 8
Joined: Mon Jun 09, 2014 3:16 am

Re: Different Yaw Speed for clockwise and counter clockwise

Post by rukie »

All motors are ok. I noticed this both on two frames I used same symptoms.

Anyway I increased my YAW rate and seems to have a helped a little bit. Hope this help others who have the same problems.

happul3
Posts: 44
Joined: Mon Apr 21, 2014 1:54 am

Re: Different Yaw Speed for clockwise and counter clockwise

Post by happul3 »

rukie wrote:Hi Guys,

I currently finished a build with the following components.

DJI f450 clone
Witespy EV3 Blackboard
Sunnsysky 2216 900kv
Rctimer 30amps esc

I noticed as said in the subject that my yaw speed differs from the two directions. My clockwise seems to be sluggish even feels that its struggling to turn clockwise. This board was transferred from a bigger quad with different components and the same behaviour can be observed. I do think that I have everything ok mechanically on the build. I think I have to tweak something in the settings or code. Appreciate if you can point me to the right direction. Thanks,


I had the same problem and after some experimentationand thinking got to the bottom of it. In addition to different responsiveness to two yaw directions, I also noticed that two motors opposing each other (left-front and right-rear) are consistently hotter than the two other motors. Using bluetooth, I've logged input for all four motors (motor[0] - motor[3]) during some test hovers and discovered that two hotter motors were getting about 100-150 units higher input on overage. That explained both temperature difference (well, duuh) and difference in responsiveness - when YAW stick called for rotation such that two harder-working motors were to increase rpms, they were actually beginning to be limited by hardcoded max. No such problem when stick called for opposite rotation.

Ok, so this explained the effects I observed, but why the uneven load on motors in the first place? After a little head scratching, I said Aaahhh!, took a ruler and started measuring distance from flat ground level to prop ends as the props were rotated by hand. Sure enough, I found that two motors (note that it must be motors, not the props), rear ones in my case, were mounted very slightly askew, such that when they rotated in normal (opposite to each other) directions, they blew air slightly off-vertical and both created unplanned-for torque for the quad. That torque was compensated by normal operation of quad: FC detected yaw and increased input two two motors /decreased input to two other motors to compensate. Please note that there were only one of hot motors was mounted askew so motor temperature is not indicative of *where* the problem is.

I do not think that your problem has necessarily the same nature, but it is easy enough to check. Hope this helps - personally, I did not realize there are such unpleasant consequences very minor skew in motor or frame.

Wipo
Posts: 22
Joined: Mon Jan 20, 2014 9:33 am

Re: Different Yaw Speed for clockwise and counter clockwise

Post by Wipo »

When you mount your motors straight on to a frame the yaw is caused by the drag of the props (rotation speed). Thats a small force that almost never provide a fast and good yaw authority. You have to make sure that the distance between the motors (pair) are exactly the same.

E.g.: if distance between the ccw motors is 20cm and between the cw motors 19cm then your yaw will be out of balance. That 1 cm difference will have a huge impact on the small force caused by drag that is used to keep heading (yaw). This can even cause your quad to be unstable on pitch and roll or difficult to tune (finding correct pid). Because your motor pairs are operating in a different powerband and different powerbands need different pid settings to be stable.

To get a fast yaw response, good yaw authority and stability you can mount your motors in a angle of 3 to 5 degrees. like in this project: http://wipo-quadrocopter.blogspot.be/20 ... ority.html

happul3
Posts: 44
Joined: Mon Apr 21, 2014 1:54 am

Re: Different Yaw Speed for clockwise and counter clockwise

Post by happul3 »

Wipo wrote:When you mount your motors straight on to a frame the yaw is caused by the drag of the props (rotation speed). Thats a small force that almost never provide a fast and good yaw authority. You have to make sure that the distance between the motors (pair) are exactly the same.

E.g.: if distance between the ccw motors is 20cm and between the cw motors 19cm then your yaw will be out of balance. That 1 cm difference will have a huge impact on the small force caused by drag that is used to keep heading (yaw). This can even cause your quad to be unstable on pitch and roll or difficult to tune (finding correct pid). Because your motor pairs are operating in a different powerband and different powerbands need different pid settings to be stable.

To get a fast yaw response, good yaw authority and stability you can mount your motors in a angle of 3 to 5 degrees. like in this project: http://wipo-quadrocopter.blogspot.be/20 ... ority.html


Interesting. I have not thought that motor tilt can ever be advantageous in quadcopter, but I see your point. However, it needs to be made clear that simply tilting motors in random direction is not a good idea. Basically, if one tilts motors such that tilts result in airflows enhancing yaw induced by perfectly level prop (two force vectors add), that should make response to yaw stick more snappy at expense of slightly wasted power when quadcopter is not turning. Titling motors in opposite direction will also waste power and make quadcopter *less* responsive to yaw stick input (and less stable with respect to holding heading). Tilting some motors "good" way and some "bad" way will produce intermediate cases. If one is really unlucky, I suppose it is even possible to accidentally compensate yaw induced by prop drag by yaw induced by tilted airstream(s), thus more or less completely loosing yaw authority and stability.

Post Reply