Motor angle dihedral What’s the deal with this?

[R_Lefebvre]

IMO, the only reason to angle the motors is a band-aid fix for a flight controller that isn't performing properly. Let's say that you have 7° of tilt as I saw, you aren't going to see much virtual dihedral effect until somewhere around half the angle, or 3.5° of tilt. That's a LOT of angle error to have. The only way you could get that far out of shape is if the flight controller has already failed in keeping it level.

Go do it, measure the before and after and get back to us on that one.

Theory is one thing but reality is another.

Don't be silly. If you don't believe these very basic rules, I hope you're living in a cave because this is one of the most basic laws that our modern society is built on.

 

[Benjamin Kenobi]

Huh? The dihedral angle is a fix for the controller?

 

[Silverburn]

I ran a APM2.5 based quad with and without a 5' dihedral. It made chuff-all noticable difference to the flight characteristics, or to loiter. Battery life was - for all intents and purposes - identical.

I'd like to see someone try it on a V4 though (eg TBS disco)...

 

[jes1111]

Why is it that so many people in this hobby believe the technology is so mystical that the only way to advance our understanding is repeated trial and error? I don't doubt that your quad seemed to fly the same with and without the dihedral and that your battery consumption seemed to be about the same. Does that invalidate the physics involved? Fact is, you tilt a motor and it will reduce the vertical lift available. Period. No argument. No doubt. If you believe otherwise then I've got another one for you - stand in a bucket and pull on the handle: human-powered flight! (apparently it works best if the bucket is red but I have tried it with a blue bucket and the flight characteristics seemed to be identical)

 

[R_Lefebvre]

Huh? The dihedral angle is a fix for the controller?

Well, why else would you do it? The dihedral won't come in to effect unless the controller has already failed to stabilize the craft.

 

[Stacky]

Why is it that so many people in this hobby believe the technology is so mystical that the only way to advance our understanding is repeated trial and error? I don't doubt that your quad seemed to fly the same with and without the dihedral and that your battery consumption seemed to be about the same. Does that invalidate the physics involved? Fact is, you tilt a motor and it will reduce the vertical lift available. Period. No argument. No doubt. If you believe otherwise then I've got another one for you - stand in a bucket and pull on the handle: human-powered flight! (apparently it works best if the bucket is red but I have tried it with a blue bucket and the flight characteristics seemed to be identical)

Im not sure which people you think believe the technology is so mystical to use trial and error over science. My first comment was aimed at the fact that the size of the shim isnt gong to have a big enough impact on thrust to make it worth worrying about. I should have been clearer on that. The people I know who have messed with this havent found it to have any visibly noticeable impact on flight times. I know that it will have an impact, as you point out the science shows that however your assertion that a 5mm shim having an 8% impact which is really close to 1/10th of performance i do question. 5mm shims will reduce a 10 minute flight to 9 minutes or maybe 9 1/2 minutes?. Lets just say that you are right on that, in terms of real world practicality and using a multi with this for a job etc its the difference between your newest battery and your oldest batteries performance, not really a problem. That was my original point which I failed to articulate clearly.

 

[DennyR]

Well, why else would you do it? The dihedral won't come in to effect unless the controller has already failed to stabilize the craft.

It may actually make it worse in some conditions. If for example a strong wind hits from the side then the effect of that would be first seen on the upward tilted side which would then have an upwards component added to the force which the controller has then to overcome. I have always come to the conclusion through many tests, that the more vertically aligned the motors are the better it is with Quads.
The other stuff does not interest me because the more motors you have the less energy efficient it is, coupled with multiple failure points and expense.

The relationship between the center of mass and the aerodynamic center of lift can produce a natural stability if these two points are centered above one another and not too far apart vertically. As the two come together then the craft will become more sensitive to response outputs and require less energy to stay level. Too far apart and the model will sway about during lateral accelerations because the motors have to work harder to maintain level flight due to the moment arm of the forces involved.

The reality of supporting a gimbal underneath means that the ideal situation has to be compromised slightly. When the camera is below it is easier to maintain constant Z axis inertial loads and with the front mounted camera then the yaw inertia is higher but the pitch/roll inertial is less but only in the roll axis.(bad unless very light but can be partially overcome with some interesting asymmetric motor layouts) One element is clear and that is that the weight of the entire gimbal assy. needs to be as light as possible if you want inherent stability incorporated into the basic design.
Asymmetric motor layouts for quads do have some advantages but to work properly require changes to the FC algorithm. Available for WKM. MK and some others.

There are some quite compelling reasons why a tri-copter is the best option. With the same development that quads have had they could have become the way to go. Interesting that DJI never supported that Idea.

 

[Scott Drysdale]

Thanks Denny.... I like the part indicated in orange/underlined text..... makes sense now.
In our industrial business the concept of deadband is introduced to eliminate "chatter" from on/off control systems such as room temperature control allowing the room temperature to swing up and down over a 2 deg F range so that the temp control relays do not burnout from too rapid on/off activity. I suspect though that yaw control using MR FCs don't need that since some yaw deadband already exists in the mechanical design with all motors straight up.

This has been covered here many times before. The radial opposite tilting of the motors is to reduce the yaw dead band so that the yaw input is working immediately. This happens because the torque reaction is supplemented with side thrust as well. Use in moderation because you are introducing cross axis errors. The Dihedral works like a fixed wing aircraft that effectively self levels because the side that tilts down now has more lift and the side that comes up has less. Combined with the downwash spread it can help the overall efficiency and protect the gimbal from some of the wind disturbance. In relatively small amounts it works great with the standard stabilization algorithm that DJI use. It works best on larger Hexas and Oktos. Small Quads are better with the thrust lines vertical because the short moment arms are more vulnerable to cross axis error. Yaw stab. being improved with a little more prop' dia. than the standard Phantom uses. 9x5 e-flight works very well or a cut down 10x5 to 9.25x5.

Cross axis error for example, means that if your model receives a roll disturbance it will respond with roll plus yaw. If you do decide to use yaw offset make sure you turn it the right way or it will cancel out the command.

 

[jhardway]

That a good answer, its a design that many aircraft have to help keep the plane stable on the horizontal plain. If the aircraft slips into a right bank then the wing on the down side is more level to the lift component giving it more wing surface/lift to that side. This in turns will cause the aircraft to inherently favor centering itself. Easy to see it on a Cessna 152, 172, 182 and others.

This has been covered here many times before. The radial opposite tilting of the motors is to reduce the yaw dead band so that the yaw input is working immediately. This happens because the torque reaction is supplemented with side thrust as well. Use in moderation because you are introducing cross axis errors. The Dihedral works like a fixed wing aircraft that effectively self levels because the side that tilts down now has more lift and the side that comes up has less. Combined with the downwash spread it can help the overall efficiency and protect the gimbal from some of the wind disturbance. In relatively small amounts it works great with the standard stabilization algorithm that DJI use. It works best on larger Hexas and Oktos. Small Quads are better with the thrust lines vertical because the short moment arms are more vulnerable to cross axis error. Yaw stab. being improved with a little more prop' dia. than the standard Phantom uses. 9x5 e-flight works very well or a cut down 10x5 to 9.25x5.

Cross axis error for example, means that if your model receives a roll disturbance it will respond with roll plus yaw. If you do decide to use yaw offset make sure you turn it the right way or it will cancel out the command.

 

[DennyR]

Lets not forget one of dihedrals sweeter functions. When you want to make a lateral movement then the high side of the craft is already directing force in the direction you want to go.

The refinement of conventional fixed pitch craft will go on but other options do exist. I cant help but think that what Curtis has done with the Stingray 500 will be the biggest game changer in the business. It offers much more precise control and autorotation. When he demonstrated the position hold it was hands off rock steady in a high wind and it was flying with only manual control. KIS is the name of the game. Variable pitch also opens up the efficiency variables as you can set what ever rpm you want. (A fixed pitch prop. has a very narrow range where angle of attack can operate efficiently without stalling). Stingray is designed for 3D and racing, what will happen when we take it further along the lines of a dedicated camera ship. Smaller and more powerful with interlocked/overlap blade sweep, rigid 3 blade heads etc. Oh yeah, it's coming alright, probably as a tricopter......

 

[jhardway]

Big fan on fixed pitch props and have been waiting for someone to produce on. On the grand scale instead of relying on multiple ESC's to manage prop head speeds you now have one motor doing all the work and what I imagine servos controlling the prop pitch. Now with that control you can probably moderate prop/lift increment with faster and small results.

Thumbs up.

Lets not forget one of dihedrals sweeter functions. When you want to make a lateral movement then the high side of the craft is already directing force in the direction you want to go.

The refinement of conventional fixed pitch craft will go on but other options do exist. I cant help but think that what Curtis has done with the Stingray 500 will be the biggest game changer in the business. It offers much more precise control and autorotation. When he demonstrated the position hold it was hands off rock steady in a high wind and it was flying with only manual control. KIS is the name of the game. Variable pitch also opens up the efficiency variables as you can set what ever rpm you want. (A fixed pitch prop. has a very narrow range where angle of attack can operate efficiently without stalling). Stingray is designed for 3D and racing, what will happen when we take it further along the lines of a dedicated camera ship. Smaller and more powerful with interlocked/overlap blade sweep, rigid 3 blade heads etc. Oh yeah, it's coming alright, probably as a tricopter......

 

[Breezemont]

I cant help but think that what Curtis has done with the Stingray 500 will be the biggest game changer in the business.

I agree. When I first saw the idea of a variable pitch multi being floated around I thought it was an answer to a question that hadn't been asked. Why bother?
Now I get it.
If I had the cash (and I wasn't up to my eyeballs in fixie stuff) I'd buy one right now.