Why it's ok to mount your FC away from the CG
ฝัง
- เผยแพร่เมื่อ 9 ก.ย. 2024
- Uncertain about what to buy? Check out THE ULTIMATE FPV SHOPPING LIST: www.fpvknowita...
SUPPORT ME ON PATREON at: bit.ly/2Ep5Sia
THIS IS MY FULL TIME JOB. Here are other ways that you can support me: www.fpvknowita...
---
Is it ok to mount your FC away from the CG of your quad? Yes. And sometimes no. But mostly yes. Let's get into it.
2:45 "You kids know what a record player is, right?" - I was there, Gandalf... I was there, 3000 years ago. ;)
I put my flight controller in the back of many of my quads when I ran the RunCam Split I'd put that stack in the front and the FC in the back and it flew great!
the gyro is not in the middle of the flight controller anyway...
but with mine (Aikon F4)
Shh! don't point out the obvious! :)
XD
@S Grim Yeah but when do need that? lol
If you use accelerometers for self leveling then theoretically yes, but practically very little. But from a structural design viewpoint, there is a actually a larger chance that you will get harmonic frequency buildup at the center of the frame, since this is typically the spot where the vibrations coming from all four motors intersect.
at 6:48 just a tiny physics comment: the acceleration of a uniform circular movement is towards the center of rotation (normal acceleration), if the object is changing its angular speed then you also will have tangential acceleration. The direction you are pointing in the video is the tangential acceleration, or maybe the centrifugal force direction (which is a non-existing force created by placing your reference frame in a non-inertial reference frame).
Anyway, your point was clearly stated and explained as usual. Sorry for bothering ;)
Centripetal* sorry lol
@@DirtyLew42 the centripetal force is the responsible of normal acceleration, it is oriented towards the center of rotation in uniform circular movement. On the other hand, centrifugal force is oriented in the opposite direction of centripetal force and it has the same magnitude. Centripetal force exists and it is needed to turn, but centrifugal force only appears when you (reference frame) are rotating (non-inertial, as you need to accelerate to make a rotation) and therefore it is an imaginary force created by the selection of this non-inertial reference frame.
Also it can be a language thing and we call those forces in a different way. This kind of force named are a bit misleading for non-native speakers, I am not really used to speak about these forces in English.
Happy flying! :)
DLabyGmr ah I see. Centrifugal force does not exist, at least I the way centrifugal is defined in english. Centripetal is the force that is most commonly misconstrued with it. In English, people confused a centrifuge, which utilizes centripetal force. Thus came the word centrifugal force, which in this case is incorrect. I wonder what the correct translation would be though.
Perhaps from the non-inertial point of view the object is experiencing centripetal force,forcing it inwards, but from the inertial point of view the object is experiencing centrifugal force, since we cannot feel acceleration, only force. As in a centrifuge something must be actively pushing or pulling the object toward the center at all times but the object itself only experiences outward force. I think centripetal/centrifugal just depends on the frame of reference but both are very important concepts.
I have been think of this as i want to mount a split cam board in the front and then there is no place on the stack for the fc except at the back. Good to know that it does not matter, straight from the horse's mouth. 😊
At the rotational speed that quads can achieve, the gyro can be affected by the g forces created. They work on measuring the difference in angular velocity from one side of the internal sensor to the other. While the difference is the same whether at the axis of rotation or away from it, the data will still be more accurate at the centre. Not to mention the vibrations are amplified as you move away from the axis.
A good gyro should be fully decoupled from non rotational forces.
The whole discussion here is about x/y displacement and yaw influence. I would say that some Z-Axis displacement of CG happens in 100% of all our builds, as the battery _will_ move the CG in Z-axis away from its theoretical 'ideal' position. So if that mattered: we would notice pitch/roll effects, wouldn't we? The only effect I observe: if it falls out of the sky it always lands on the top-mounted battery :)
I'm surprised the FC doesn't use the gyro data to compensate for centripetal acceleration in the accelerometer data. It's not a hard calculation if you know the position of the FC relative to the CoM, but I guess it doesn't know (and can't infer?) where the CoM is. I also didn't appreciate how unnecessary the accelerometer is for acro mode. Thanks, I learned something today.
I use a hyperlite glide frame and mount my flight controller in the front. I have had no issues with it at all and flys great!
Hyperlite ftw
Great question..
Great answer 😉
Joshua I really don't know what the fpv community would do without you... Much love brotha. As I said before anytime your in Nova Scotia you have a place to stay with cold beverages readily available.
I have been doing this since DJI digital fpv came out. I put the AU in the center of my Glide and my stack in the rear. Kabab has been spreading out his stack for a long time with his FC in the front usually.
Thanks!!👍🏼 That question was in my mind for quite some time. Definitely learned something today.🥳
The old Immersionrc Vortex 250 Pro had the FC at the back. That worked well on that quad 👍🏻
Pretty interesting! Thank you, Joshua! 😊
Stay safe there with your family! 🖖😊
The acceleration at a constant angular speed of the motor is towards the center of mass... And it's called centripetal.
To have tangential acceleration it needs to have a variation of angular speed, inducing tangential acceleration..
So... If you have your flight controller on the back.
Yaw rate its coupled with autolevel-pitch (centripetal), and yaw rate changes (yaw acceleration) is coupled with autolevel-roll.
Didn't understand a word but still hit on the like button.
There are 2 sensors in the flight controller. Accelerometer and Gyro. Placing the flight controller off center does not effect the reading of the gyro. It does effect the accelerometer. Most miniquads fly in rate mode which uses only the gyro and ignores the accelerometer. But yes, a flight controller mounted off center can have acceleration introduced if mounted off center.
The off-center yaw rotation induces tangential and centripetal linear accelerations that act on the body and have to be integrated with the rotational velocity to form an overall percept of the motion through the environment.
Yeah, I was going to say that. Thanks
As I was watching I was thinking about what you were going to say about vibration, but really, as long as the flight controller is mounted with some form of soft mounting to stop most of the vibration, the quad should fly just fine even if it it’s out to one of the extremities, might make a little bit of difference in angle mode, but as you said, the effect probably wouldn’t be that noticeable.
Very helpful for some of us. I've been curious about this for a long time now. I asked Timecop about it years ago and just got a filthy response riddled with insults and profanity. Thanks for taking the time to actually go into it a bit.
Hi Josh Great video thanks again for the information
Nice video!!! I think the key of good fly is to have the CG in the same position than the "iddle" trust vector.
Normal mounting on the centre of bottom plate with a top battery is not around the centre of mass either for pitch or roll axis
The FPVCycle Glide by Kababfpv is designed to take the Fc in the front, esc in the middle and vtx, reciever in the back
Thank you
I thought we'd sussed all this out many years ago. Thought this must be an old video until I saw the release date for this.
There are still newcomers to the hobby I guess.
There always will be
Youre an idiot.
@@coolhanddruid You're a troll.. and "youre" isn't a word. Learn to use basic English if you dont want to be seen as one yourself. ;):D
@@paulhope3401 You think because "we'd sussed it" then it would never be talked about again? Youre a pretentious pos. And you were probably never involved in the "we" in the first place.
@@coolhanddruid It appears you either can't read.. or have chosen to deliberately misquote me. You are clearly the fucking idiot here.
Eccentric rotation is just rotation + movement. So the gyro shouldn't have a problem but maybe the accelerometer would. Although I'm not sure how much the accelerometer matters to the FC. EDIT: I should watch the whole video before commenting😅
yap: 5:40
7:48 it should be customised in betaflight there you could set position of FC, by the way in dead some gyros is not centered in the FC :) this means it does little bit floating moves :)
I was just wondering about this and saw that Joshua Bardwell answered the questions so I immediately dropped everything and clicked on this video. The milk can clean itself up.
sorry, on the acceleration part for the yaw spin, the are all accelerating 'in' toward the CG and the directions you pointed out was the current direction of motion of that part.
Thanks you have answered so many questions I had on this topic 👍👍
In the example of the ANGLE mode with FC mounted way back and yaw, the FC will try to stay in the center of the rotation and yaw the entire quad around itself. That will cause all that pitch and roll compensation Joshua explained as a consequence.
Thanks Josh, one again I learned something.
In a yaw movement the motors are accelerated towards the point the quad turns around. What you describe is the acceleration in an accelerated reference system.
Joshua doesn't really know it all ;-)
Fantastic explanation. Thank you.
I came in to this thinking your point was the opposite lol... great info thanks sir!
Thank you for the video. I am designing a 10" hexacopter with a spider style frame and was wondering where the optimal position to mount the fc would be
usually there are settings in the firmware to set the exact accelerometer offset from the center of gravity
What is missing is the discuerssion what the gyro sensor actually measures. Is it the rotational speed or "something" which is very close to the rotational speed if the sensor is near to the center of that rotation.
Gyro just measured 3-axis angles. Accelerometer measures about the 3 axes acceletation
I have always wonderd about that
pretty much agreed. remember the old eachine racer250 and its off centered fc? mine flies great in any mode to this day. acro, angle even gps hold. I've always been skeptical about this subject even though the centering does makes sense.
great thanks
Makes sense to me great video JB.
Hmmm…. I agree with you IF the gyros only measured rotational rates. Do they? I don’t know. I’m just learning. If they also measure translational rates then location would matter. …just trying to learn…
Next step in drone frame technology, frame resonance testing data. Graphs, spreadsheets and all that jazz.
Great vid. Better Q - why don't we mount the fc vertically, in fact the whole stack rotated 90 degrees - horizontally. Always the same height for any quad. Slot in board on the baseframe if you really like :)
Physics guy here, just wanted to clear some things up.
"If you have a rigid object any point will rotate at the same rate": dθ will be the same, but the linear torque *force* will not. The closer you are to the pivot, the less torque you experience. Think of what would happen if you were to sit on a rigid lever 10000 miles long. Being on the far end away from the pivot = you would be obliterated. Yet both the pivot and the far end have the same rate of angular acceleration.
"Penny at a center of a record player is not experiencing any acceleration". It *is* experiencing centripetal acceleration. Think of being spun around, you get dizzy because there's a rotational force acting on you.
All correct. But a gyro doesn't measure acceleration. It measures rotational rate. And acro flight only uses the gyro. So for acro flight, distance from CG doesn't matter.
Hey, Joshua. For much larger quads, say with a 700mm wheelbase and a 200mm long bottom plate, would it matter to have the FC mounted forward of center by 2-3" if flying autonomous mapping missions (mostly lawn mower pattern....flying straight, until rolling briefly and yawing 180° at the end of the line to repeat for the next leg over)? Thanks!
Quads (and other things) rotate about the centre of inertia, not mass - the two can be quite different
ok have can. but I am already skeptical with the gyros not in the middle of the FC.. lol I remain on the idea of the center of gravity .. otherwise yes, we can move the battery itself too, but that's not why it will work as well as is the better?, I doubt it ..
You are just looking at theoretical caluclations basedon perfect sensor data. Sensors are not perfect. Assuming that a gyro sensor can deliver better data if it sits in the center of the rotation it is measureing is at least feasible. I looked up the MPU-6000 datasheet. What they mention is a 0.1%/g (error per g-force). This g-force will be bigger off center with the same angular speed. That is looking at the simplified movement of only rotating around the center. But I think even with more complex movement we can assume that the g-force is lower newr the center than further away. The error still should be low enough to igonore it.
centripetal acceleration always faces the center of rotation
Another great JB video (that tag line never gets old). Ironically, this applies in the fixed wing world as well, where there is the possibility to place the FC much farther from the CG and angle mode is used more often. Better to be as close to the CG as possible within the limitations of the design of the model, and a few inches forward or aft is sufficiently close enough. It seems like vinyl is making a comeback so a few more folks might actually own a record player these days. 😜
When I flew wings there was no flight controller, just a Rx. I have 3 planes with a 6 foot wingspan, and I'm looking forward to making one a long range electric. In the 90s there was no level mode, lol
On my Hyperlite Glide the FC is in front behind the camera, away from the ESC, VTX, and receiver's electrical noise. Flies fine even in angle mode, might rotate around a point an inch and a half ahead on the center of the frame, but i certainly can't tell, and if that's closer to the camera's POV then that's better as far as i'm concerned.
I would think gyro works better closer to center of rotation 🤔
Ok explanation makes sense.. angle is the same in all locations.
In my head just seemed weird timing pitching yawning lol
But I fly angle a lot of times
They DO fold and flex. But only when it doesn't matter anymore because it's time to buy a new frame and maybe a new gopro.
Looks like a Marriott property. Sometimes I miss the road but mostly love WFH.
I just switched frames form a iFlight iX5 to a iFlight DC5 frestyle. Now I have a wobble at the end of hard yaw movements but only in horizon mode. Acro/Air mode does not have this problem. The biggest difference I see between the frames is that the DC5 has the flight controller in the front part of the square formed by the motors.
Just curious why you use horizon mode? Only use I've found is a quick Los test with flips,and rolls. I don't use it for that anymore since I learned how to fly acro.
@@RCOverKill Horizon is not my main mode. Due to camera uptilt I'll use it to land or if I'm line of sight flying in the dark. The Leds are cool but its hard to really see the drone angle at all times.
@@rmedal4594 I see, I like to fly up fast about 5 feet away from my chair, do a quick 180 to a stop and set it down. I'm working on a split second landing, but much farther away, lol Nurk does a nice one at the end of the train, flight of the year.
A few years ago I gutted and rebuild my mom's 4ft lift top chest stereo...I couldn't find a "record player". Type in "turntable" and got all kinds of choices. Point being terminology helps hide the age gap
Just on time, thanks for saving my money 🙏
It wont effect it from the gyros perspective but it will effect it from PID tune perspective. If you place it in the rear of the quad on center line then Pitch and Yaw gains would be different but roll would stay normal (in a flat flight model) Now put it at cg on pitch axis but off to the side of the quad and pitch would be unaffected and roll and yaw would be wonky. So yeah it kinda does matter specially when you rely heavily on filtering like you tend to do as the filters are I believe written based on a cg mounted FC model and would have nothing to compensate for off cg placements. Would be a good point to address the guys who write the code for all this to answer.
No that's incorrect that's the whole reason I made this video is to clear up claims like this. PIDs are not affected by the FC position on the quad.
@@JoshuaBardwellYou are correct PIDS do not matter to the FC by position. However they do matter in what it is controlling and would have to be changed as FC position changes for optimal control. IF the quad still rotates around the FC as center then correct no changes. If you are talking no change if the center of rotation is still at the center between the motors as normal then I guess we have to agree to disagree.
You also state pids dont matter with the betaflight filtering. I have yet to see a quad using only betaflight filtering and stock pids that did not fly like total ass. Even your stuff until you started to fly with the Drew crew wasnt the bestest most smooth with all the black boxing magics.
PIDS do matter. No matter what you fly without them controlled flight would not be possible. Its what ALL filtering is based off of manipulation of pids. If you start with a wet noodle the best you will ever get is a wet noodle that is over working itself. Betaflight crew did an awesome job making flying easier with various gear but in the end its still all based on pids.
Id make a bet if you shut off all filtering and tuned a quad at CG with just pids then moved it away from perfect CG there would be a noticeable difference the further away you get. Betaflight did such good work on filters it masks a TON of bad tendencies which can be very misleading as to what is really happening.
Joshua please can you tell me why my Matek F722SE flight controller knows it's compass orientation even before I add and configure gps/mag module ??
It powers up and assumes whatever direction it's facing is north. Then after powerup it uses gyro integration (dead-reckoning) to estimate direction. So it knows it yawed 100 dps left for 2 seconds, therefore its new direction must be such. Accumulated error will make this inaccurate over time.
I’m not sure it’s so simple. The more a point is far from centre of gravity, the more it will be affected by the centrifugal force. And I’m affraid the gyro can be confuse between orientation and weird forces.
The gyro doesn't know anything about orientation. It only reports rotational speed (degrees per second). You're mixing up the gyro and the accelerometer.
This has been discussed for years actually. It’s still optimal to have the FC near or at the CG and it has to work less.
If you have evidence to support that claim present it. I've made an argument why that's not true.
Several old RCGroup forums mentioned it - that was back in the early days of weaker FCs. There as a Painless360 video discussed awhile back about FC position too. In my personal testing, FC position didn’t make a difference with modern ones.
This might become an issue with an airplane. If you mount the FC way up in the nose with the fpv stuff. Or for some reason on the wing tips there could be problems.
What about RESCUE MODE, would it get affected?
I always mount my fc just behind thr cam. But insont fly stability at all so it dont matter
Im surprised there's no setting along the lines of Gyro Offset from CG with 1 to 3 boxes to fill in...
Maybe not so much in BF but could definitely be of benefit to iNav users...
What is in the bottle?
Good morning. You seem to have a little pep in your step this morning.
Wouldn’t placing the flight controller closer to the front increase the pitch/roll/yaw resolution?
I can't think of any reason why that would happen.
Shark byte review , come on
yes
No rescue mode whit another place... The good question is how the velocity affect the gyro ?
"Aerodyne Nimbus" had it's flight control board mounted on it's edge...how does that work?
There are board alignment settings in BF.
@@flaparoundfpv8632 interesting, I didn't know there was settings for other than degrees on a horizontal plane
Is that your new print ed GoPro couch?
Sauce? Don't leave home without it!
You should rename your channel Mr Wizard lol...you so brought back my childhood memories of that show in this Vlog lol!
There would be more centrifugal forces exerted on the FC at different stages of flight though, would this have any effect on the giro?
No, because centrifugal force is an acceleration and the gyro measures only angular rate (degrees per second) not acceleration.
I fly the auskwads montser Frame and have my fc at the front of the frame and the escs at the back and fine I do t have any trouble with the the where the fc is mounted
I missed the (oldschool) Motor Mixer Part...
👍 nice 👍
Woo-hoo learned something today 🤓
good stuff :)
what about FC position if you are using a GPS module? Does that matter?
This video feels like it may be related to something having short cables between a camera and VTX...
Every other video people have said No it doesn't matter and I'm not going into detail again.
Private pilots: Adverse yaw will kill you if you don't handle it properly!
Quad pilots: Yeah well you know, that's just your opinion man.
newbeedrone lanyard is the best lanyard
He knows...it’s a holiday inn!
Waaaait up a moment - quadcopters don't fold? That must be what I've been doing wrong...
How about the vibrations?? Will i experience more than mounting at the center of the quad?
I would think cg might be the least vibrations, but there's not enough to make a difference now days. When I started they were just starting using gummis to mount the FC, b4 they used silicone pads for the motors. It's come so far in the 2.5 years I been flying, I can't even imagine flying without air mode, well I can because I've done it, but I've never tried doing freestyle with air mode off.
The Gyro needs to be as close to center of gravity or mass as possible. Angular motion will be more severe further away from the center of mass, MEMS style Gyros micro-structure uses a 2d structure measuring capacitance in movement within the structure. The further away it is the stronger the centripetal force is or Coriolis effect. So in a yaw movement if the Gyro is being Pushed away from center of mass the angular motion will be stronger than actually being experienced by the mass itself at the center.
If the FC is trying to compensate for the angular motion or re-level the mass it will have to overcompensate on a whatever axis the gyro is centered on. Being that a Quad is designed to be centrally balanced this can cause issues for FCs and overworking the Gyro/accelerometer
Hotel jb edition ...is back ?
rotational rate (speed of rotation) vs directional velocity (speed in a direction) is an extremely simple concept that many people just can not understand. Just look at the flat earthers...
Imagine not flying acro :)
What’s a record player? Jkjk love the video JB!
Joshua ... is that sweet and sour or worcestershire sauce you took with you there? Right next to your portable mouse? ... :D:D:D
It's hot sauce. I went to Tijuana Flats but their hot sauce bar is shut down because of the pandemic. So I had to buy my own bottle.
@@JoshuaBardwell A man who brings his own hot sauce with him on his trips is a man I can trust.
I'm getting a serious Vsauce vibe from this video.... 🤔😁
When JB was hoping people still know what a recordplayer is.
I felt old and felt the generation gap...
As if he is old
@@svosprey I'm guessing mid to late 30's. Not old by a long stretch (also, consider that ours is the generation that never really grew up), but a 20-something year old or a teen would consider him somewhat old, yeah...
If you just learn how to fly well in acro there is no need for the accelerometer. It's like training wheels. 😆 😄
That's the ticket! It can still come in handy in certain instances tho.