DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Motor failure imminent?

Mynah

Well-Known Member
Joined
May 1, 2021
Messages
137
Reactions
140
Location
South Africa
My Mini 'refurb' is nearing the 200km mark and started displaying weird behavior. It started with unexpected sudden yawing during slow acceleration in cine mode, it will recover instantly and be happy for the rest of the flight.

I also started noticing right front props touching the body. This happens occasionally during windy flights in sport mode but now it happens during normal flight. So I changed all props and re-enforced the front arms with G10, something I've meant to do long ago. Problem persisted. Then, finally, I received a single brief motor error, first time ever. I prevented Fly App uploading the flight data and had a look at the .dat file and noticed a slow running right back motor.
The error occurred at 93 secs into the flight:
Capture.jpg
Errordisplay.jpg

I then did a hover test indoors and yet again right back motor seems to run slower than the rest:
hovertest.jpg

Am I right to suspect that the right front motor is working harder to compensate and thus prop to body contact occurs even though I re-enforced the arms?

Those in the know please advise if my presumption is correct in believing the right back motor needs replacement.

The de-formed prop phenomenon is not at play here since I have not stored the drone folded and I've replaced all props. All prop screws are properly tightened and correctly aligned. I have checked all motors for obvious mechanical blockage/fouling.

I'm tying to attach .dat file in question but get a message that the file is too big for the server to process. Will try again in separate post.
 
Last edited:
There's very little to go wrong with the motors and they should last many thousands of hours.
I'd look elsewhere for the cause of the issue.
 
  • Like
Reactions: BigAl07
Thanks Meta4, I have re-installed the firmware on the Mini, could or might also be software related. Will keep a close eye and ear on it going forward. If it drops from the sky it might be a sign from Thor to upgrade to a Mini 2 :)
 
If your props are contacting ANYTHING at ANY portion of the flight you have a pending disaster in the making. It's not the motor but a structural issue allowing the right front arm/motor to move out of normal position.
 
  • Like
Reactions: maggior
...It started with unexpected sudden yawing during slow acceleration in cine mode...
The reasons for a uncommanded yaw movement can be many ... if the reason is motor or prop related the yaw movement is also coupled with movements in both the pitch & roll axis. If the movement is isolated only to yaw the most probable cause is instead related to ghost controller inputs ... I would firstly calibrate the sticks & in the calibration window make sure that no uncommanded inputs show up in another axis than the one you command.

If the cause is motors or props the yaw movement comes from either a motor not being able to keep the torque balance with the other motors due to not able to turn fast enough (maxed out RPM wise or ceasing) or a prop that can't produce enough thrust ... again leading to that the motor over revs & maxes out.

So if the AC didn't wobbled & lost height when this happened you can rule out motors & props as a root cause.


I also started noticing right front props touching the body...finally, I received a single brief motor error, first time ever... Am I right to suspect that the right front motor is working harder to compensate
That the front props touches the fuselage have been a very common problem with the Mini 1 ... the weak arms & the flimsy props seemed to be the main cause earlier. If it would be that a rear motor would uncommanded rotate slower the solution wouldn't be to rev up one in the front ... it would be to slow the rest of the motors down, try keeping the rotational torque in balance & over that also try to counteract rolling & pitching coming from one side that isn't performing.

That single motor error at 93sec most probably originated from that the rear right prop couldn't produce quite enough thrust & by that was rotating very near to max... have placed the marker there in the chart below, note that the FC commands the rear right to 98,9%.

1633611428158.png

At that spot you had the wind of about 6,5m/s coming in perpendicular towards the AC from a North Easterly direction (hitting the left side of the AC) ... the sat pic below show the wind direction with the yellow bar & the yaw direction with the blue bar.

1633611449352.png

...noticed a slow running right back motor. I then did a hover test indoors and yet again right back motor seems to run slower than the rest...

I have never been able to draw any firm conclusions from hover tests performed in confined spaces with a lot of turbulence hitting the craft ... if the AC hovers stable & reaches specified max ground speeds in forward flight all should be good.

Those in the know please advise if my presumption is correct in believing the right back motor needs replacement.
The de-formed prop phenomenon is not at play here since I have not stored the drone folded and I've replaced all props.
Think your conclusion is wrong ... try to calibrate your sticks (& check for ghost inputs) & if the motor error occurs periodically change the props. Just so you know ... it's earlier been several reports of brand new props being deformed directly out from the box.
 
  • Like
Reactions: MavicAir2Marc
My Mini 'refurb' is nearing the 200km mark and started displaying weird behavior. It started with unexpected sudden yawing during slow acceleration in cine mode, it will recover instantly and be happy for the rest of the flight.

I also started noticing right front props touching the body. This happens occasionally during windy flights in sport mode but now it happens during normal flight. So I changed all props and re-enforced the front arms with G10, something I've meant to do long ago. Problem persisted. Then, finally, I received a single brief motor error, first time ever. I prevented Fly App uploading the flight data and had a look at the .dat file and noticed a slow running right back motor.
The error occurred at 93 secs into the flight:
View attachment 135954
View attachment 135953

I then did a hover test indoors and yet again right back motor seems to run slower than the rest:
View attachment 135952

Am I right to suspect that the right front motor is working harder to compensate and thus prop to body contact occurs even though I re-enforced the arms?

Those in the know please advise if my presumption is correct in believing the right back motor needs replacement.

The de-formed prop phenomenon is not at play here since I have not stored the drone folded and I've replaced all props. All prop screws are properly tightened and correctly aligned. I have checked all motors for obvious mechanical blockage/fouling.

I'm tying to attach .dat file in question but get a message that the file is too big for the server to process. Will try again in separate post.
How did you reinforce the arms? I haven't found an adhesive which sticks properly to the plastic these things are made of. Tape won't give much support to the moulded area where they appear to be at risk.
I have been wanting to do this since I got mine..
Sorry doesn't answer your question.
 
If your props are contacting ANYTHING at ANY portion of the flight you have a pending disaster in the making. It's not the motor but a structural issue allowing the right front arm/motor to move out of normal position.
Two factors at play here, flimsy front arm structure AND flimsy flexible props, both rectified on the follow-up Mini models. Most obvious during hurried ascend in sport mode combined with crosswinds.
Think your conclusion is wrong ... try to calibrate your sticks (& check for ghost inputs) & if the motor error occurs periodically change the props. Just so you know ... it's earlier been several reports of brand new props being deformed directly out from the box.
Thanks for your detailed analysis Slup. It might have been distorted props out the box, I have replaced the right front props, calibrated the RC and took it for two test flights in stronger than normal wind conditions, put it through its paces and had no further issues. Touching wood :) The first ever motor warning totally threw me into panic mode but then, rather safe than sorry. I'll be sure to save logs in future when something unusual occurs.
How did you reinforce the arms? I haven't found an adhesive which sticks properly to the plastic these things are made of. Tape won't give much support to the moulded area where they appear to be at risk.
I have been wanting to do this since I got mine..
Sorry doesn't answer your question.
Tiger64 I used Bostik superglue gel and roughened and flattened the arm channel edges and the G10 for best adhesion. Also cleaned roughened surfaces with acetone. Total weight added is 1.8gram but SO worth it. The characteristic sound of blade tips hitting body is nerve wrecking! G10 is glass fiber embedded in resin, used for knife handles and also for printed circuit boards, very light but very tough. The superglue did the trick.
IMG_20210926_151954.jpg

The telltale signs of props touching aircraft:
IMG_20210926_152144.jpg
 
I have also considered using 'motor protectors' to gain some lift for front props for better clearance but decided against it for artificially extending the prop shafts will definitely put extra strain on the motor bearings. Those tiny motors work hard as is so I settled on re-enforcing the arms even though I'm no fan of 'after market' design changes on anything hi-tech.proplift.jpg
 
I have just looked at two Mavic Minis and their propellor tips are approximately 1cm clear of the top of the front arms inner ends.

The props lift the drone by 'pulling' it upwards through the air and pushing the air downwards, if they flex due to providing lift surely they would flex upwards not downwards.
Pick the drone up by the outer ends of two diagonally opposed blades, what way do they flex? Upwards for me.
I have just hand-held a MM with the props at eye level, started the motors and given them full throttle. It was a bit awkward to see the blade tips, indoors and not a very good back ground, but I am reasonably sure I saw the tips rise as I throttled up.

For the moment I would forget about raising the props and check that the motor is securely attached to the arm, there are 3 screws under a foam cover.

Aside from a loose motor I can think of only five ways in which undeformed i.e. not bent downwards, blades could strike the arm,
a) their screw has been installed askew, such that the blade tip is lower than it should be when the motor is rotating
b) their individual pivot hubs were worn and allowing the blades to wobble about the screw or
c) the screw itself was loose and wobbling.
Even then I think "b)" would need to be combined with "d)", I am not sure about "c)".
d) The drone was subjected to sudden and vicious changes of pitch and or roll which would momentarily cause the props to flex due to their attempts to retain their angular momentum, (hold a bicycle wheel by its axle, spin the wheel up then try to change the angle of the axle).
e) the arm having a crack in its bottom and sidewall and flexing/pivoting upwards when lifting the drone.

Any slop in an arm pivot would affect the whole arm and the motor and the blades so they would in effect move as one (flexing of the blades due to changes in angular momentum aside).

Unless you test it somewhere non critical first I would be very wary of using acetone as a cleaning agent, Iso Propyl Alcohol is likely much safer. I have seen, to my cost, acetone attack plastics before.
 
Last edited by a moderator:
...The props lift the drone by 'pulling' it upwards through the air and pushing the air downwards, if they flex due to providing lift surely they would flex upwards not downwards.
Unfortunately none of the movements that the props or arm do is static ... yeah, the props bends upwards when steady increasing the thrust but when decreasing it they will bounce downwards, and this together with an equal movement pattern coming from the weak arms increases the probability for the prop tips to touch the fuselage. That the props bounce up & down is normal ... have the front props on my 5" quad in view in my googles & you bet they move up ... & down depending on if I'm on or off the throttle.

When it comes to the Mini 1 this issue isn't uncommon at all ... it's been numerous of reports of this on this forum & the pics shown in this thread looked exactly the same as all others.
 
Undoubtedly the blade will move downwards, from a peak high, when the throttle is reduced but why would they move below their rest position?
Throttled off they are still producing an up thrust i.e. they are still experiencing an upwards bending force, just a reduced one. If they bend below the rest position something would have to force them to do so and I doubt they oscillate to any extent under their own elasticity and inertia.
Whether gyroscopic forces created when pitching and rolling the drone are sufficient to cause enough flex in 'healthy' blades I do not know but my main Mavic Mini has over 700 flights on it and I am none to gentle with it and I have yet to see any such markings on its arms.

All I am suggesting is that there may be other causes behind this and that they should be investigated.

Maybe DJI stiffened the arms etc. partway through the production run but the arms of my mini do not flex enough, in a vertical arc, to bring the blade tips near the drone's body or the arm. In my hands, they do however, twist but if that occurs in flight it looks as if the contact point would be on the drone's body just behind the nose.

Anyway those are my observations, make of them what you will.
 
Last edited by a moderator:
The props lift the drone by 'pulling' it upwards through the air and pushing the air downwards, if they flex due to providing lift surely they would flex upwards not downwards.
The motor and prop angle changes in relation to the drone body when the flexing of the arms take place so it's not only up and downward clearance that's affected.
When parked on the table it seems totally impossible that a prop can touch the arm or body, and beleve you me, I double checked all points on your list. But then, that's probably what the design engineers of the Tacoma suspension bridge said. I even considered mounting a high speed camera on it to solve the mystery, only, I cannot find one light enough and I don't have access to a wind tunnel.

When looking deeper into the possibilities I realized that there are very intricate forces at play while the Mini does its thing, concepts like 'aeroelastic flutter' and 'resonance oscillation torsion' popped up and I decided, you know what, I'm NOT going to do the dodgy DJI engineer's job who skimped on the original design, I only bought it to fly it, I'll simply slam 1.8grams worth of G10 on it and solve the problem, and I can since where I fly the 250 gram weight limit is not an issue.

If you have never experienced the prop on body issue with your Mini then I'm convinced that DJI made some design changes after the first Mini's were produced. Mine had the issue from day one out the box and I'm no exception.
PS: As for using acetone to prep bonding surfaces you are correct, I should have stated 'wipe with acetone', bathing any plastic compound in acetone will be looking for trouble, it dissolves PVC, polystyrene and most resins.
 
Last edited:
How did you reinforce the arms? I haven't found an adhesive which sticks properly to the plastic these things are made of. Tape won't give much support to the moulded area where they appear to be at risk.
I have been wanting to do this since I got mine..
Sorry doesn't answer your question.
The superglue bond started giving up after a weekend of heavy flying so I repeated the excesise with Pratley's epoxy, I'll report back on the outcome.
 
Undoubtedly the blade will move downwards, from a peak high, when the throttle is reduced but why would they move below their rest position? ...
The motor and prop angle changes in relation to the drone body when the flexing of the arms take place so it's not only up and downward clearance that's affected. ...
In this clip all possible arm flex movements that a quad can suffer from is nicely shown ... look at the graphics between 3:42 & 9:30.

It's not at all only prop flex that will make them hit the fuselage ...

 
  • Like
Reactions: Mynah
Just an update; the Pratleys epoxy is holding the G10 re-enforcing in place. I took the Mini down to the Wild Coast and flew in rather hectic wind conditions without any further 'funny' noises or motor errors. Thanks again to all who shared their expert knowledge and saved me from prematurely replacing Mini motors at great cost in a panic. I've passed the 200km/120mile mark and the Mini is still going strong :)
DJI_0806.jpg
 
The motor and prop angle changes in relation to the drone body when the flexing of the arms take place so it's not only up and downward clearance that's affected.
When parked on the table it seems totally impossible that a prop can touch the arm or body, and beleve you me, I double checked all points on your list. But then, that's probably what the design engineers of the Tacoma suspension bridge said. I even considered mounting a high speed camera on it to solve the mystery, only, I cannot find one light enough and I don't have access to a wind tunnel.

When looking deeper into the possibilities I realized that there are very intricate forces at play while the Mini does its thing, concepts like 'aeroelastic flutter' and 'resonance oscillation torsion' popped up and I decided, you know what, I'm NOT going to do the dodgy DJI engineer's job who skimped on the original design, I only bought it to fly it, I'll simply slam 1.8grams worth of G10 on it and solve the problem, and I can since where I fly the 250 gram weight limit is not an issue.

If you have never experienced the prop on body issue with your Mini then I'm convinced that DJI made some design changes after the first Mini's were produced. Mine had the issue from day one out the box and I'm no exception.
PS: As for using acetone to prep bonding surfaces you are correct, I should have stated 'wipe with acetone', bathing any plastic compound in acetone will be looking for trouble, it dissolves PVC, polystyrene and most resins.
Personally, I'm convinced that the problem is entirely due to prop flex, and the root of the problem is that it is designed to too tight a spec for the material used: The tip is very close to flat, and if a tip is hit with a gust of wind, I believe it can twist enough to generate a downward thrust, and if the prop doesn't recover quickly, it can bend enough to hit the body. If you liked to "fly" your hand out the car window when you were a kid (or if you still do), you know that if you twist your hand only slightly to a negative attack angle, and you don't resist the downward force generated, it might be enough to cause even more twisting, which causes a very sudden increase in both the downward force and the twisting.

It's not hard to bend the prop down until it hits the body, but I can't see any way that the arm could flex that far without serious damage. Also, if it was the arm that's bending, there would be a sudden change in its thrust vector that would drive the flight control system nuts for at least long enough to show a violent reaction in the drone. I haven't heard that mentioned as being coincident with props hitting bodies.
 
well i have just read this thread from the start ,and the prop strike is definitely not a new problem in my early days of MM ownership back when they first came out ,i suddenly had prop strike happen to my drone
after some investigation ,i came to the conclusion that the main cause was the front arms flexing
the mini came with a soft foam like insert in the front arms ,which did nothing for strength
the rear arms though had a stiff cover and were much harder to twist because of it
i found that an old credit card made a good cover for the arm, and in fact there is enough to make one for each arm,i left the foam insert where it was and super glued the shaped card to the front arms
this was over 200 flights ago and since doing my mod i have not had any more prop strikes
even flying in sport mode and some wind this fix does not prevent prop deformation when stored in the original case that the MM was supplied with that is a separate issue
 
  • Like
Reactions: Mynah
OMM, excluding twist can you remember if you could detect any significant up and down flexing of the front arms if you ever tried to bend them.
With my bought new in May 2020 I can not but they are easy to twist.
The above is a question, not an argument.
 
It's not hard to bend the prop down until it hits the body, but I can't see any way that the arm could flex that far without serious damage. Also, if it was the arm that's bending, there would be a sudden change in its thrust vector that would drive the flight control system nuts for at least long enough to show a violent reaction in the drone. I haven't heard that mentioned as being coincident with props hitting bodies.
Roger I suspect it is a combination of arm flex and prop flex. The arm twists, it is basically a miniature soft plastic u channel filled with flexible foam. The motor 'error' displayed as in the beginning of the post I suspect was just that; control system going nuts after a cross wind resulted in prop to body contact.
I have not had any further erratic behavior since I re-enforced the arms and changed the props, also no more telltale scuff marks caused by prop to body contact, nor any motor errors.
Looking at the computer simulation slup posted explains a lot as for body and arm flex while in flight.
 
  • Like
Reactions: old man mavic
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
130,593
Messages
1,554,196
Members
159,598
Latest member
fast54