Hopefully time to put this to bed once and for all. I have the ruined clip to show you all and I also have the flight log showing the drone losing control when there was absolutely no reason for it to do so. This is very much releated to the braking issue, as the drone is over correcting movements, causing it to oscillate.

I shot a clip for a wedding on a beautiful calm day, with no wind. A standard slow and controlled flight over a vineyard, nothing unusual.

What's important about this flight is that it is diagonally flown, where the stick sits between the forward and left direction, which causes the drone to oscillate. I am seeing it time and time again that the Mavic 3 cannot fly in a controlled manner when a diagonal input is used.

The exact moment the flight happens is between 9m 5s - 9m 29s and the video below is this flight, where the drone clearly cannot remain stable, even though my stick inputs remain in the same direction.

https://www.phantomhelp.com/logviewer/117X1C3AZUOEZ1N1LRBT/

The oscilations are even visible on the map.

Disable OA, it's definitely the cause of this. It will interfere if you  want perfectly smooth footage when you are that close to the ground  (and even higher when it picks false positives).

Brake mode works  to do a general flight, or if you are exploring, doing photos or using  the drone for things like inspections. But if you want smooth video  footage and a perfect flight path you need to temporally disable OA and  make sure you are feeding the drone with perfect signal. This is less  than 200m from the drone, direct unobstructed line of sight, the  controller pointed towards the drone, FCC mode and the smartphone in  airplane mode. In any other condition, you are introducing variables  that will lead to various problems and the results will vary.

OA  off, drone near, controller pointed, that's how drone videography is  done. On photography on the other hand you can bring the drone to  whatever distance you want because having bad signal or OA giving false  positives from now and then means nothing.

PS: Furthermore, you  can also play with the controller settings on the different modes and  change the sensibility and smoothness, so it filters the movements you do  on the sticks to perform a smoother trajectory.  

Huginn Kenningar Posted at 8-1 03:12
Disable OA, it will interfere if you want perfectly smooth footage when you are that close to the ground (and even higher when it picks false positives).

Brake mode works to do a general flight, or if you are exploring, doing photos or using the drone for things like inspections. But if you want smooth video footage and a perfect flight path you need to temporaly disable OA and make sure you are feeding the drone with perfect signal. This is less than 200m from the drone, direct unobstructed line of sight, the controller pointed towards the drone, FCC mode and the smartphone in airplane mode. In any other condition, you are introducing variables that will lead to various problems and the results will vary.

So I really should have said in the original post, but I never fly with OA turned on. This is not an OA issue because it was never turned on.

When I talk about brake settings, I’m talking about the tuning of the drone, not the brake OA setting.

CloudVisual Posted at 8-1 03:15
So I really should have said in the original post, but I never fly with OA turned on. This is not an OA issue because it was never turned on.

When I talk about brake settings, I’m talking about the tuning of the drone, not the brake OA setting.

Then we'd have to wait for that forum member that analyzes the logs, maybe he can point out the problem because you are raight, input seem to remain constat but the drone is doing it's own thing.

Hi,

Had a look at your log.
See my chart of your data between 9m5 - 9m29s.

Bottom sensor had a hard time to get the correct height info, so as 1 + 1 = 2 ;
this had apparantly effect of the low speed flying stability.

cheers
JJB

JJB* Posted at 8-1 03:45
Hi,

Had a look at your log.
[Image]

Thank you for this, good to see it graphed out this way.

I did spot on the logs that there were several missed readouts of the ground VPS distance. I just can’t see how that would tell the drone to aggressively overcorrect when there was excellent GPS.  

CloudVisual Posted at 8-1 03:52
Thank you for this, good to see it graphed out this way.

I did spot on the logs that there were several missed readouts of the ground VPS distance. I just can’t see how that would tell the drone to aggressively overcorrect when there was excellent GPS.

Hi,

Good question, afaik with good GPS only GPS is used for stability.
But at low heights (and low speed flying) somehow the bottom sensor is used as well....This add-on is not decribed in the manual.

This is almost always noticed flying low over streaming water; hover and height fluctuations, sometimes drone becomes wet....  ;-(

cheers
JJB

So nothing to see here ….except maybe a problem with bottom sensors something that can probably be seen replicated with all dji drones.

hallmark007 Posted at 8-1 08:32
So nothing to see here ….except maybe a problem with bottom sensors something that can probably be seen replicated with all dji drones.

I wasn't holding my breath for your input on this. Even though the logs show that it wobbled and the footage was unusable at the end - the exact thing you asked me to prove.

CloudVisual Posted at 8-1 08:41
I wasn't holding my breath for your input on this. Even though the logs show that it wobbled and the footage was unusable at the end - the exact thing you asked me to prove.

I think what you tried to do was fly in such a place that had nothing to do with the Mavic 3 stopping abruptly and apart from showing nothing to that affect , what you showed was clearly explained by JJB showing exactly what caused your problems and it was nothing to do with what was discussed on the other thread. You need to be more clear about what your actual problem is I can post a video flying low over long grass and it won’t show this problem at all . Just let me know if you need it posted.

hallmark007 Posted at 8-1 09:08
I think what you tried to do was fly in such a place that had nothing to do with the Mavic 3 stopping abruptly and apart from showing nothing to that affect , what you showed was clearly explained by JJB showing exactly what caused your problems and it was nothing to do with what was discussed on the other thread. You need to be more clear about what your actual problem is I can post a video flying low over long grass and it won’t show this problem at all . Just let me know if you need it posted.

I know you find it difficult to differentiate between issues. So here is the thread about the drone stopping and ruining clips, in case you need reminding:

https://forum.dji.com/thread-270200-1-1.html

Super delighted you can fly a nice straight line. I can do that too. What I can’t do, is help when the drone decides to do its own thing, which I’ve proven with the evidence above.

Please do post your footage of a nice flight over grass. By the way you talk, I’m expecting absolute perfection.

CloudVisual Posted at 8-1 10:08
I know you find it difficult to differentiate between issues. So here is the thread about the drone stopping and ruining clips, in case you need reminding:

https://forum.dji.com/thread-270200-1-1.html

Well its not absolute perfection, but I thought I’d try to slow it down 120 fps just that if there were any imperfections caused by sudden flight movements they would show up easily at that speed . But you wouldn’t have to go far to see some quite extraordinary videos shot on mavic 3.

Video filmed on a pretty windy day slowed to 120 fps 5 ft above top of the grass , OA off .

hallmark007 Posted at 8-1 10:32
Well its not absolute perfection, but I thought I’d try to slow it down 120 fps just that if there were any imperfections caused by sudden flight movements they would show up easily at that speed . But you wouldn’t have to go far to see some quite extraordinary videos shot on mavic 3.

Video filmed on a pretty windy day slowed to 120 fps 5 ft above top of the grass , OA off .

I find it funny that you actually posted a video to prove you can fly in a straight line.

hallmark007 Posted at 8-1 10:32
Well its not absolute perfection, but I thought I’d try to slow it down 120 fps just that if there were any imperfections caused by sudden flight movements they would show up easily at that speed . But you wouldn’t have to go far to see some quite extraordinary videos shot on mavic 3.

Video filmed on a pretty windy day slowed to 120 fps 5 ft above top of the grass , OA off .

Now post one while flying diagonally. That is what give the OP problems. Straight flying works fine for him.

Mobilehomer Posted at 8-1 11:09
Now post one while flying diagonally. That is what give the OP problems. Straight flying works fine for him.

Sorry your saying flying straight there is no problem with how the mavic 3 brakes. Because thats the crux of the problem people are complaining. So now the problem is flying diagonally? The video looked like it was flying diagonally across the field.


Below is curved right to left lowering the craft flying straight and backwards while ascending.

hallmark007 Posted at 8-1 11:31
Sorry your saying flying straight there is no problem with how the mavic 3 brakes. Because thats the crux of the problem people are complaining. So now the problem is flying diagonally? The video looked like it was flying diagonally across the field.

Read the OP's post again. Flying straight he had zero rpoblems. When he flew diagonally, the M3 would oscillate. Look at his posted picture.

Mobilehomer Posted at 8-1 11:09
Now post one while flying diagonally. That is what give the OP problems. Straight flying works fine for him.

So really is the problem now not with the braking on the Mavic 3 because that is what the previous thread was about . I think the title was when would we be able to stop the mavic 3 “softly” it seems now there is a different problem.

Mobilehomer Posted at 8-1 11:34
Read the OP's post again. Flying straight he had zero rpoblems. When he flew diagonally, the M3 would oscillate. Look at his posted picture.

Again you are completely mixed up here. The op was posting a video showing the problems caused by the mavic 3 stopping abruptly. I think it needs to be explained how braking abruptly is causing this “NEW” problem.

Mobilehomer Posted at 8-1 11:34
Read the OP's post again. Flying straight he had zero rpoblems. When he flew diagonally, the M3 would oscillate. Look at his posted picture.

You know Im sure that the easiest way to film anything diagonally is to take two points on a diagonal and just fly left to right or right to left in a straight line.

hallmark007 Posted at 8-1 11:46
You know Im sure that the easiest way to film anything diagonally is to take two points on a diagonal and just fly left to right or right to left in a straight line.

Just go away. You are zero help.

Mobilehomer Posted at 8-1 12:04
Just go away. You are zero help.

Considering you don’t even own a Mavic 3 maybe you should go away because it has diddly squat effect on you and considering your have no clue whats going on it seems like trolling is what you’re here for.

hallmark007 Posted at 8-1 12:24
Considering you don’t even own a Mavic 3 maybe you should go away because it has diddly squat effect on you and considering your have no clue whats going on it seems like trolling is what you’re here for.

To the contrary, he seems to know exactly what’s going on thanks to the fact he’s reviewed the footage, logs and evidence and was able to verify there’s an issue.

Your input, so far, is to show us your videos where you managed to fly in a straight line or a bit of a curve, as some kind of way to prove that there isn’t an issue. Also you keep dragging up the braking issue, which is a separate, but slightly related problem.

I think you’re the troll here bud

Have you tried diagonal flight across flatter ground? Say a soccer pitch? Looking at the graph JJB posted, I wonder if the rows of grapevines were affecting the bottom sensors? Just a guess, but worth a try.

CloudVisual Posted at 8-1 12:47
To the contrary, he seems to know exactly what’s going on thanks to the fact he’s reviewed the footage, logs and evidence and was able to verify there’s an issue.

Your input, so far, is to show us your videos where you managed to fly in a straight line or a bit of a curve, as some kind of way to prove that there isn’t an issue. Also you keep dragging up the braking issue, which is a separate, but slightly related problem.

Ok so now its a different problem and slightly related to the abrupt braking .

This thread was directly born out of the previous thread Titled “When will it be possible to softly stop the Mavic 3?” You are now in post 30 saying it is only “slightly” related to the braking problem. But in your opening post and I will Quote “ This is very much releated to the braking issue” so it seems it is you thats very much confused here.

You posted video and no one is disputing whats in them but after going into jjb’s frap there was no proof as to why your drone behaved as it did and certainly nothing to suggest braking had anything to do with it, but you’re insisting now its only slightly to do with braking something I might add you never mention until post 30 . And this post really has nothing in common with others posts regarding braking. If it does then be my guest and show us.

Mobilehomer Posted at 8-1 13:05
Have you tried diagonal flight across flatter ground? Say a soccer pitch? Looking at the graph JJB posted, I wonder if the rows of grapevines were affecting the bottom sensors? Just a guess, but worth a try.

I’m glad you asked this.

So I’m heading back home on Friday and my plan is (in a very British fashion) to fly the drone in a pattern which will resemble a Union Jack over the fields near me. My hope is that I can recreate this over flat ground and prove that only diagonal flight causes these wobbles.

CloudVisual Posted at 8-1 13:26
I’m glad you asked this.

So I’m heading back home on Friday and my plan is (in a very British fashion) to fly the drone in a pattern which will resemble a Union Jack over the fields near me. My hope is that I can recreate this over flat ground and prove that only diagonal flight causes these wobbles.

Honestly, I hope it works perfectly for you. This would indicate the vineyard is to blame rather than a hardware or firmware issue. But either way, I think you will discover the reason/cause. Good flying.

May be worth experimenting with flying the same flight path with your controller set to mode 1, so that you can split the 'forward elevator' and 'left aileron' stick commands between the two gimbals. This way we can eliminate the possibility of a faulty/poor quality gimbal in your controller when flying diagonally in mode 2. I had problem with one of the gimbals on my mavic pro controller a long time ago, which cause the drone to behave in a similar way.

CloudVisual Posted at 8-1 12:47
To the contrary, he seems to know exactly what’s going on thanks to the fact he’s reviewed the footage, logs and evidence and was able to verify there’s an issue.

Your input, so far, is to show us your videos where you managed to fly in a straight line or a bit of a curve, as some kind of way to prove that there isn’t an issue. Also you keep dragging up the braking issue, which is a separate, but slightly related problem.

If Uncle Claddagh doesn't behave himself he will get banned again.

Monkey007 Posted at 8-1 13:36
May be worth experimenting with flying the same flight path with your controller set to mode 1, so that you can split the 'forward elevator' and 'left aileron' stick commands between the two gimbals. This way we can eliminate the possibility of a faulty/poor quality gimbal in your controller when flying diagonally in mode 2. I had problem with one of the gimbals on my mavic pro controller a long time ago, which cause the drone to behave in a similar way.

As much as I like the idea of this, mode 2 is so ingrained into my flying that I wouldn’t ever be able to swap and make a stable comparison.

I seem to be coming late to this party but here is my stupid questions:
1. You say you have the OA disabled, Did you just turn it off in the app? Reason for this is as Mobilehomer pointed out that it could be the vineyard because although you turn off the OA sensors, you can never really turn of the bottom sensors. The bottom sensors, if flying very low and not holding the stick down to land but forward will try to avoid anything below it and the drone movements will be erratic.
2. Have you done an RC Stick calibration to see if maybe your sticks are out of sync?
3. Have you tried putting some tape over the bottom sensors and trying this flight route again?

Suren Posted at 8-1 15:51
I seem to be coming late to this party but here is my stupid questions:
1. You say you have the OA disabled, Did you just turn it off in the app? Reason for this is as Mobilehomer pointed out that it could be the vineyard because although you turn off the OA sensors, you can never really turn of the bottom sensors. The bottom sensors, if flying very low and not holding the stick down to land but forward will try to avoid anything below it and the drone movements will be erratic.
2. Have you done an RC Stick calibration to see if maybe your sticks are out of sync?

Good suggestions.  Any downside to taping the bottom sensors?  I have not done that, but wondering how the drone reacts when those are blocked.

TonyPHX Posted at 8-1 16:12
Good suggestions.  Any downside to taping the bottom sensors?  I have not done that, but wondering how the drone reacts when those are blocked.

Lets just say you dont want to hold the stick full down with that taped up But I flew a little with a tape over the bottom sensors in some fields because the drone when flying forward was raising and lowering itself constantly even in sports mode, with the tape the drone flew steady

Suren Posted at 8-1 15:51
I seem to be coming late to this party but here is my stupid questions:
1. You say you have the OA disabled, Did you just turn it off in the app? Reason for this is as Mobilehomer pointed out that it could be the vineyard because although you turn off the OA sensors, you can never really turn of the bottom sensors. The bottom sensors, if flying very low and not holding the stick down to land but forward will try to avoid anything below it and the drone movements will be erratic.
2. Have you done an RC Stick calibration to see if maybe your sticks are out of sync?

The OA is turned off at the app level only, I've not covered up the sensors. I'm aware you can tape up sensors, but I'm worried that doing this may turn up more issues. I once blocked an Inspire 1 bottom sensor with a 360 camera and the onscreen warnings and messages just didn't fill me with confidence.

When I fly low, I almost never encounter the drone adjusting height, unless it's due to me flying at the minimum altitude before a landing would be initiated, but this is a very rare shot for me to do. You'll see the vineyards were much higher than this and the height readout was around 16-19ft.

I haven't recalibrated the sticks, but I will. Good suggestion

I won't tape up the bottom sensor. If it can't be turned off, then it'll only further confuse the drone and I'm sure the downward positioning does come in handy in some areas of bad GPS.

Interesting thread.  I'm curious about disabling the bottom sensors fully for flying low over objects.  I imagine landings would potentially be a bit more like a controlled crash.

sky_makai Posted at 8-2 12:14
Interesting thread.  I'm curious about disabling the bottom sensors fully for flying low over objects.  I imagine landings would potentially be a bit more like a controlled crash.

If you tape the bottom sensors for low flying then you should hand catch for landing.

Suren Posted at 8-1 16:28
Lets just say you dont want to hold the stick full down with that taped up  But I flew a little with a tape over the bottom sensors in some fields because the drone when flying forward was raising and lowering itself constantly even in sports mode, with the tape the drone flew steady

I only do the full stick down thing with my M2P.  : )

I flew one of my drones with various sensors taped up as an experiment once (might have been my M2P) and yes, it'll fly ok. But it gave warnings all over the place and it was a challenge to land with the bottom ones taped up. It just drops right into the ground...

For as much money as it cost, I probably will just make do with my M3 as-is. Mine does the same thing, slowing down and halting, etc., as I approach the ground even at high altitudes like 50 or more feet. I can make it better by turning off OA in the Fly app, but it won't let me get close to the ground. I guess for a $2000 aircraft, I'm ok with it not flying into the ground even if it spoils a shot.. .

Hopefully we'll one day get the ability to calibrate the sensors in the DJI Assistant, like we had in the previous aircraft. A proper calibration is probably all that's needed....

luciens Posted at 8-3 04:03
I flew one of my drones with various sensors taped up as an experiment once (might have been my M2P) and yes, it'll fly ok. But it gave warnings all over the place and it was a challenge to land with the bottom ones taped up. It just drops right into the ground...  

For as much money as it cost, I probably will just make do with my M3 as-is. Mine does the same thing, slowing down and halting, etc., as I approach the ground even at high altitudes like 50 or more feet. I can make it better by turning off OA in the Fly app, but it won't let me get close to the ground. I guess for a $2000 aircraft, I'm ok with it not flying into the ground even if it spoils a shot.. .

I agree with you luciens.  I think that although the M3 did not get overwhelming love, it still has a lot of potential if they get a few of these nagging issues handled.