Mavic Mini uncommanded descent tests

[droneloverair]

So all in all. The fix would be stronger/stiffer blades to combat flexing.

I wonder how that carbon fiber blade sar linked would perform.

 

[FoxhallGH]

So all in all. The fix would be stronger/stiffer blades to combat flexing.

Or - stow it in the Fly More case as suggested by DJI ...

 

[Ice_2k]

Or - stow it in the Fly More case as suggested by DJI ...

View attachment 100694

I am always paying attention to how I pack it in the case and with all that I just looked at my logs the other day and I can see my rear left motor spinning at almost 3000 RPM more than the right one. Also, I’m guessing part of the problem is that it might happen that the prop moves slightly as you put it in the casr and you can’t really see if it’s under the belly of the drone.

 

[FoxhallGH]

I am always paying attention to how I pack it in the case and with all that I just looked at my logs the other day and I can see my rear left motor spinning at almost 3000 RPM more than the right one. Also, I’m guessing part of the problem is that it might happen that the prop moves slightly as you put it in the case and you can’t really see if it’s under the belly of the drone.

That's my guess as to what's happening - yes. I don't know if it helps to tip the drone forward as you put it in? - Or a maybe a small rubber-band put loosely around the props to keep them east-west would help??

 

[Zen Fly]

That's my guess as to what's happening - yes. I don't know if it helps to tip the drone forward as you put it in? - Or a maybe a small rubber-band put loosely around the props to keep them east-west would help??

Or a little plastic clip

 

[bigswifty]

I was actually in a shop today which had the DJI consumer range on display. It was the first time I'd seen a Mini in the flesh and after all this I just had to check the props, and I was shocked how flimsy they felt, both for bending and twisting. A Tello was sitting next to it, and the Tello props felt stiffer. I walked away rather amazed that a Mini could fly at all with those props.

 

[KiloMeater]

CsvView which can be found here
CsvView Downloads

You might also want to check out
Mavic Flight Log Retrieval and Analysis Guide

Is there some reference which explains which attribute is what in CsvView?
Some are easy to understand like the motors rpmbut I was trying last night to detect vertical and horizontal speed and could not locate them

 

[Keule]

Is there some reference which explains which attribute is what in CsvView?

Begin here:

V3 .CSV column descriptions

 

[slup]

Is there some reference which explains which attribute is what in CsvView?
Some are easy to understand like the motors rpmbut I was trying last night to detect vertical and horizontal speed and could not locate them

In general ... no.

You can though get some guidance through the link that @Keule provided & in section 5 here --> Mavic Flight Log Retrieval and Analysis Guide

But it will be confusing as it's hard to figure out the more complex signals shown in the logs & further more what they actually indicate ... To get out similar info that sites like Airdata.com or Phantomhelp.com show is the easy part ... if those signals is enough then those sites lay it out in a much more understandable way actually. But when it comes to more complex forensic tasks involving sensor behaviors it's first a matter of finding the correct signals to look at in CsvView, then understand what they show & lastly knowledge how the normal behavior is regarding those sensors so you can spot the deviation.

I personally have found that follow all the crash & flyaway threads here at the forum gives the fastest understanding ... replicate what's done in those threads (with your own logs for instance) regarding chosen signals in CsvView, read up on the explanations given & ask questions there if something isn't obvious. With this time spent you will slowly gain more understanding & knowledge about "normal DJI drone" behavior & which signals to look at in different situations.

 

[Deep Blue Witch]

Not impressed with the speed of the dji response with this issue. I recently ordered a MM and am considering just returning it until the problem is resolved

 

[RichR]

A few simple checks before you fly and the issue is unlikely to occur. People who fly radio control fixed wing or helis spend plenty of time doing checks before taking off, because it means you're not going to be surprised by something loose or not fitted correctly causing a crash. Drones really shouldn't be any different - and my Mini cost more than any of my helis or fixed wing planes, so I feel I should treat it with perhaps more respect with regards to storage and pre-flight checks.

If you look at the rear props and they look any different to the front ones, then change them. If on takeoff the rear goes up after the front, change the props. Fly gently for the first 5-10 minutes to see if it's behaving or sounding unusual, and if it is land and check it over. Keep it in VLOS and if it looks to be doing something other than what you expect, gently bring it down and back to home.

Enjoy flying it, but be aware that it is a complex piece of electronics and mechanical parts and so can go wrong in some circumstances. Read this forum to get an idea of what kind of things can cause problems, and how best to react should they occur. If yours does develop the uncommanded descent issue (and there's nothing to suggest that everyone will experience it), then understand what you can do to minimise damage and get it back safely, then change the props and get back in the air.

At some point either DJI or some third party will release better props. It's only been out less than a year and seems to have been exceptionally popular. Many people have been flying regularly since last year with no issues at all, so don't assume that you'll experience them.

 

[Max67]

If on takeoff the rear goes up before the front, change the props.

I think you meant to say, "If on takeoff the front goes up before the rear ...".

 

[sar104]

A few simple checks before you fly and the issue is unlikely to occur. People who fly radio control fixed wing or helis spend plenty of time doing checks before taking off, because it means you're not going to be surprised by something loose or not fitted correctly causing a crash. Drones really shouldn't be any different - and my Mini cost more than any of my helis or fixed wing planes, so I feel I should treat it with perhaps more respect with regards to storage and pre-flight checks.

If you look at the rear props and they look any different to the front ones, then change them. If on takeoff the rear goes up before the front, change the props. Fly gently for the first 5-10 minutes to see if it's behaving or sounding unusual, and if it is land and check it over. Keep it in VLOS and if it looks to be doing something other than what you expect, gently bring it down and back to home.

Enjoy flying it, but be aware that it is a complex piece of electronics and mechanical parts and so can go wrong in some circumstances. Read this forum to get an idea of what kind of things can cause problems, and how best to react should they occur. If yours does develop the uncommanded descent issue (and there's nothing to suggest that everyone will experience it), then understand what you can do to minimise damage and get it back safely, then change the props and get back in the air.

At some point either DJI or some third party will release better props. It's only been out less than a year and seems to have been exceptionally popular. Many people have been flying regularly since last year with no issues at all, so don't assume that you'll experience them.

Looking for prop damage has always been an important pre-flight check but, in this case, the problem is that the deformation isn't at all obvious with the props on the motors. And flying gently to begin with often doesn't manifest the problem. Better props is definitely the only real solution here.

 

[droneloverair]

Looking for prop damage has always been an important pre-flight check but, in this case, the problem is that the deformation isn't at all obvious with the props on the motors. And flying gently to begin with often doesn't manifest the problem. Better props is definitely the only real solution here.

Why hasnt any aftermarket seen this as a potential product?

Isnt master airscrew gonna release a stiff props soon?

 

[Knoepel]

Yes, they will release them soon. Master Airscrew STEALTH Props for Mavic Mini?

 

[RichR]

I think you meant to say, "If on takeoff the front goes up before the rear ...".

You're right - corrected

 

[droneloverair]

Yes, they will release them soon. Master Airscrew STEALTH Props for Mavic Mini?

I hope they are following this thread.

Anybody here told them yet about the MM problem?

 

[tino.v]

I was wondering if there are any more uncommanded descents after upgrading to 1.0.0500
and replacing props if indicated by motor speed error ESC?

I searched the forum but couldn't find any referrals.

Regards
Tino

 

[sar104]

I was wondering if there are any more uncommanded descents after upgrading to 1.0.0500
and replacing props if indicated by motor speed error ESC?

I searched the forum but couldn't find any referrals.

Regards
Tino

It's in this thread.

Mavic Mini uncommanded descent tests

@sar104 ——Would you be willing to do the test on a set of the new carbon fiber props for the MM? If so, I’ll have a set sent to you from China. Sure - if I can get some I'll test them. I wasn't aware that any were yet available.

mavicpilots.com

 

[Zbip57]

[...] in this case the problem is clearly caused by deformed props, not the motor axis.

I'm starting to suspect that the motor axis angles actually do have influence on how the props become deformed. It has to do with the way the props mesh when stored with the arms folded.

This diagram shows how the front and rear motors are mounted on different angles. I don't know the actual precise angle, these are just estimates, but the rear motors are noticeably canted outwards, while the front motors are canted rearward.


I've always been cautious to fold the props correctly when storing the Mini in its Fly More case, so I really don't believe I ever jammed them in there improperly. But I nevertheless received the Motor Speed warning on my first flight with the v01.00.05 firmware. I replaced the rear-left blades as instructed and haven't seen the warning again.

However, I did notice this. Previously the left and right prop blades slid neatly together when nested with arms folded. But, the tips of the newly installed fresh blades stand higher than the old ones, and they no longer slide under the other blades. Because of the way the rear motors are canted outwards, the tips of the prop blades stand higher than the opposite motor.



I subsequently also replaced the rear-right blades. They now also point higher than the opposite motor.



With fresh blades installed on both rear motors, you can see the tips of the blades now cross each other. The props can no longer be nested together without forcing one set to bend under the other set.



Storing the Mini for any period of time with the props crossed under each other will eventually cause the blades to permanently deflect. Even when stored "correctly" within the Fly More case, I still suspect forcing the blades to fold together is what's causing them to take on a permanent deflection.

This might even explain why some people have experienced Uncommanded Descents with brand new Mavic Minis straight out of the box. The Mini is shipped with the arms folded. There's no telling how long they've been stored like that before their first flight.

So how come we're seeing the Motor Speed warnings almost exclusively on the rear motors, and practically never on the front motors?

Well, with the arms folded, the fronts motors on my Mini are almost exactly level with each other, maybe even canted slight inwards.



I also replaced the front-right blades. These are the fresh blades (below).



And these (below) are still the original front-left blades which have been on the Mini since December.



When folded together, the front blades mesh easily without putting any bending stress on each other.



What do you guys think? Does any of that make sense?

Edit: Correction, I replaced the left-front leaving the right-front blades as the only ones still original.