That's a good point - it's quite possible that the motor axis has an effect in the storage configuration. My comment previously was only addressing effects in flight.
Mavic Mini uncommanded descent tests
- Thread starter sar104
- Start date
That's a good point - it's quite possible that the motor axis has an effect in the storage configuration. My comment previously was only addressing effects in flight.
Totally agreed. The motor axis angles are deliberately set that way for stability and response factors while in flight. I didn't mean to imply any disagreement with what you wrote.
It's the storage configuration that looks wonky to me.
I replaced the left-rear blades as instructed by the Motor Speed warning and had no problems since. But I have not again stored the Mini in its Fly More case.
Once I finally figured out how to install the Java RTE required to get CsvView to run, I plotted the motor speeds from the flight that triggered my Motor Speed warning. And I could see that the left-rear motor hit a peak of just under 16,000 rpm at one point.
There was nothing visibly wrong with the original blades. No knicks, or cracks, or blemishes. It was only upon installing the fresh new blades that I noticed the difference shown in the photos of my previous post.
Since then I did some simple indoor hover tests to eliminate any effects of wind (other than from its own prop wash). Just press auto-takeoff, let it hover for a couple of minutes, then auto-land, all without touching the control sticks.
This is with new left-rear blades, and original blades still on the other three motors. Here with fresh blades the blue left-rear motor is no longer spinning faster than the others. But the yellow right-rear motor is now the fastest.
So, even though I haven't received any more motor speed warnings, and even though there was nothing visibly wrong with the blades, I decided to also install fresh blades on the right-rear motor and repeated the hover test. Note the yellow line is now nearly equal to the blue and red. The green left-front motor speed is now the fastest.
So I also replaced the prop blades on the left-front motor and repeated the hover test. Note that now all four motors are turning at nearly the same speeds while in hover.
While it likely was not necessary to replace these other prop blades, evidently there is a very clear difference between the performance of old versus new propeller blades. And that is despite the lack of any detectable difference when visibly comparing the old vs. new.
Once I finally figured out how to install the Java RTE required to get CsvView to run, I plotted the motor speeds from the flight that triggered my Motor Speed warning. And I could see that the left-rear motor hit a peak of just under 16,000 rpm at one point.
There was nothing visibly wrong with the original blades. No knicks, or cracks, or blemishes. It was only upon installing the fresh new blades that I noticed the difference shown in the photos of my previous post.
Since then I did some simple indoor hover tests to eliminate any effects of wind (other than from its own prop wash). Just press auto-takeoff, let it hover for a couple of minutes, then auto-land, all without touching the control sticks.
This is with new left-rear blades, and original blades still on the other three motors. Here with fresh blades the blue left-rear motor is no longer spinning faster than the others. But the yellow right-rear motor is now the fastest.
So, even though I haven't received any more motor speed warnings, and even though there was nothing visibly wrong with the blades, I decided to also install fresh blades on the right-rear motor and repeated the hover test. Note the yellow line is now nearly equal to the blue and red. The green left-front motor speed is now the fastest.
So I also replaced the prop blades on the left-front motor and repeated the hover test. Note that now all four motors are turning at nearly the same speeds while in hover.
While it likely was not necessary to replace these other prop blades, evidently there is a very clear difference between the performance of old versus new propeller blades. And that is despite the lack of any detectable difference when visibly comparing the old vs. new.
Agreed. I had the exact same observation and posted it in this thread:
MM Folded Rear Prop Bending per DJI Instructions
Sorry to beat a dead horse, but even when the Mavic Mini rear props are folded per DJI's instructions to fit in the Fly More case, they get bent which could be a contributor to the uncommanded descent issues that many of us have been having. Here is how DJI's instructions in the case say to...
mavicpilots.com
Despite asking people to try this themselves to see the effect on their own, people basically responded saying that the props were being folded incorrectly by me and others when storing in the case. So I appreciate that you have also come to the same conclusion
But still frustrated with the problem.Pat
Aha! I was trying to remember where I saw this posted. It was you who first pointed this out to me.Agreed. I had the exact same observation and posted it in this thread:
MM Folded Rear Prop Bending per DJI Instructions
Sorry to beat a dead horse, but even when the Mavic Mini rear props are folded per DJI's instructions to fit in the Fly More case, they get bent which could be a contributor to the uncommanded descent issues that many of us have been having. Here is how DJI's instructions in the case say to...
Concerning carbon fibre props, over in the DJI forum someone posted a DAT file of a similar hover test using aftermarket carbon fibre props. To make a more meaningful comparison, the csv files were imported into Excel to calculate average motor speeds, not counting those under 8000 rpm while taking off or landing.
Import the csv file into Excel and average the motor speeds, not counting those under 8000rpm at takeoff & landing,
=AVERAGEIF(FL2:FL3809,">8000") In my spreadsheet the motor speeds were in columns Fi,FJ,FK, FL.
Here's a comparison of the average motor speeds in hover. It seems that fresh DJI props are slightly more efficient at hover with lower motor speeds compared to aftermarket carbon fibre props.
120ccpm's Hover Test
# New Left Back props.
Right Front: 9582 RPMs
Left Front: 9386 RPMs
Left Back: 9431 RPMs
Right Back: 9960 RPMs
Zbip57's Hover Test
With fresh L.Front, L.Rear, & R.Rear blades. R.Front remains original.
R.Front - 9756
L.Front - 9603
L.Back - 9974
R.Back - 9564
Markforrester99's Hover Test (@ Carbon props)
R.Front - 10814
L.Front - 10611
L.Back - 10820
R.Back - 10071
It remains to be seen whether the disadvantage of slightly higher motor speed is outweighed by the advantage of carbon fibre being less prone to distortion.
Import the csv file into Excel and average the motor speeds, not counting those under 8000rpm at takeoff & landing,
=AVERAGEIF(FL2:FL3809,">8000") In my spreadsheet the motor speeds were in columns Fi,FJ,FK, FL.
Here's a comparison of the average motor speeds in hover. It seems that fresh DJI props are slightly more efficient at hover with lower motor speeds compared to aftermarket carbon fibre props.
120ccpm's Hover Test
# New Left Back props.
Right Front: 9582 RPMs
Left Front: 9386 RPMs
Left Back: 9431 RPMs
Right Back: 9960 RPMs
Zbip57's Hover Test
With fresh L.Front, L.Rear, & R.Rear blades. R.Front remains original.
R.Front - 9756
L.Front - 9603
L.Back - 9974
R.Back - 9564
Markforrester99's Hover Test (@ Carbon props)
R.Front - 10814
L.Front - 10611
L.Back - 10820
R.Back - 10071
It remains to be seen whether the disadvantage of slightly higher motor speed is outweighed by the advantage of carbon fibre being less prone to distortion.
I am a newbie and the Mini is my first drone. I am hearing a lot about the forced decent and prop problems and I ordered some CF props to try. My background is in engineering and worked in a factory where I was in charge of predictive maintenance operations. One problem we ran into was resonance where at certain speeds, a fan blade would go into resonance (the natural frequency where an object will start to vibrate). Could it be that at certain speeds, the props go into resonance which could cause them to flex due to the vibration and lose it's lift? Resonance depends on speed and mass so as speed changes and other factors such as wind or angle could send the props into resonance or to start vibrating causing them to flex. If so, by changing the mass and rigidity by going with a CF prop, it would change the speed at which the props would go into resonance (hopefully above the max speed the drone is run at) and solve the problem. We would run fans with variable frequency drives from start up to full speed and check vibration levels with sensors to see if there were any speeds where the fan went into resonance then we would lock out those frequencies on the variable speed drives so they would not stay at those frequencies while trying to control a HVAC system. Just wondering.
If you follow DJI diagram for folding the rear props, the left rear prop folds under the right prop. This is the way a new drone comes packaged. I've always thought the left prop will have a greater load placed on it with this folding configuration. This may explain why a greater number of users experience rear left motor errors.
It's possible that there are resonances that have an effect. I'm hoping to do some high-speed video imaging of the props while spinning that might reveal if that is happening.
I took a few more photos to illustrate how the props become bent when stored with the arms folded and prop blades crossed.
It wasn't as obvious with the original rear props because they seemed to mesh together nicely when folded. But the newly installed fresh props won't fold together unless one side or the other is bent under/over the other side. If they're repeatedly forced to mesh together and stored for extended periods like that, then eventually the blades deform to take on the deflected shape.
Here's an image illustrating the angle at which the rear motors are canted outwards. With the blades turned outward like this, you can see how high the tips of the props rest above a flat surface. In this position the weight of the Mini is resting directly on the prop hubs, attachment screws, and motors, and the prop blades are not being bent.
You can see how far down the tips of the prop blades hang when turned inboard. If the blades from both sides are turned inboard, obviously the tips of the prop blades will cross and will not easily slide over/under each other.
If set down on the table top with the prop blades in this position, the Mini's weight is resting on the tips of the blades. The blades flex and are distorted. It's really not a good idea to store the Mini resting on a table top like this.
The Fly More case is designed with a raised section to support the belly of the Mini and a deep recess into which the props can hang freely. However, I suspect that merely forcing the rear props to mesh, crossed over/under each other, to fit into that recessed area is what eventually causes permanent distortion of the blades.
In the photo below I'm standing the Mini on its tail, allowing the props to dangle down freely along the sides of the body, rather than forcing them to cross over each other across the body.
The rear motors are canted at an angle outwards, but they are not canted forward or back. When positioned longitudinally like this, you can see that the prop blades lie flat alongside the body. In this position you could set the Mini down flat on a table top with no stress applied to the rear blades. And there is no bending stress applied to the blades because they're not meshed folded laterally across the body.
I ordered a new carry case from AliExpress in which the props are stored longitudinally, rather than crossed over/under the body of the Mini. It's still not the perfect case though, as there is no space for the charger and spare batteries. But I'm hoping this will be a solution that's friendlier to the propeller blades.
I submitted the order four weeks ago but still haven't received it...
Sunnylife Protective Storage Bag Carrying Case for Mavic Mini
TDZHDTV
Well-Known Member
I have the Sunnylife drone only case, does not trap the props if loaded vertically, good quality too.
FoxhallGH
Well-Known Member
Excellent work @Zbip57 and that's a really sensible theory. I hadn't realised that even if the rear props are stowed as DJI recommend, they are still under a distorting pressure! The alternate case makes sense, and using any alternate stowage that aligns the blades north-south as you have shown, must be a sensible way to go. Way back in the thread, I suggested that a 3rd party carbon fibre prop-set would be the best cure for all this. I still think that's the case - but it is just a matter of finding a good & reliable supplier ...
Here’s my experience with a different case:
Props: are we sure it’s just how we put it away?
So... - I’ve not been using the FMC since the 28th of April. - I have replaced all the props on the 16th of May (yes, I did crash into a bush while I was orbiting) with original parts. - It flew about 4 hours since then. - The MM has been left ”resting” every night into a shaped foam where the...
mavicpilots.com
Carbon fibre props would be stiffer, less flexible, than the stock DJI props. It's reasonable to think that should make them more resistant to being distorted.
But I wonder, is it even possible to fold the carbon props together for storage in the manner required to fit into the Fly More case?
My original props were always stored folded properly in the case, and as a result they nested together easily without complaint. It's only when I installed new props that I could see the difference. The new props could also be forced to nest together, but there is some obvious bending required. Would carbon props allow themselves to be similarly bent to force them to nest together when the props are folded across the body, or would they just crack?
The harmonics theory is a good one, but how could you tell if it was third order or maybe fifth order frequency, or a combination of several orders? I also think a smoke chamber would help determine if there's interaction between the forward and rear blade tip vortices in high speed horizontal flight. I'm sure a fair amount of testing was done by the factory. It couldn't have all been a magic SWAG. I've tried longer blades that overlap front and back and the interaction between forward and rear didn't help like I thought it would. My next test will be a shorter wider chord blade. This should increase the solidity ratio and be quieter with slower tip speeds.
My experience with Mavic carbon fiber props is while certainly not as flexible as OEM props, they do flex. Not sure about carbon fiber used for a MM as its props are so much shorter.
TDZHDTV
Well-Known Member
Be careful with 3rd party props, I bought these off AliExpress after reading good reviews. They are slightly stiffer and I didn’t suffer any body strike while aggressively flying in S mode, however they are not matched pairs and I suffered jell-O effects on video.
www.aliexpress.com
1.38US $ 25% OFF|8PCS Replacement Propellers for DJI Mavic Mini Drone Light Weight 4726 Props Blade Accessory Wing Fans Spare Parts|Propeller| - AliExpress
Smarter Shopping, Better Living! Aliexpress.com
www.aliexpress.com
FoxhallGH
Well-Known Member
I think you have hit the nail on the head! With the prop's being mounted at an angle on the motor, they are not going to 'mesh' to be stowed, without some distortion. CF prop's are simply going to highlight that fact even more. Hopefully 'Master Airscrew' will bring out a prop-set for the Mini soon - but meanwhile, it's starting to look like the FlyMore case should not be used!
chiangmaicharlies
Well-Known Member
I wish DJI would make their best intentions clearer. You can of course arrange the front blades in a north-south config in the combo case, as in my photo. But not the rear blades. Would you advise this?
Note that for the moment I have added white markers to the blade screws to quickly identify which blades should be uppermost, according to DJI.
The front blades are not nearly as critical, since they do mesh together quite easily. Storing them north-south as you've shown is possible and eliminates the potential for any bending that might be caused by meshing the blades together when crossed east-west over the body. But I'd be worried that in this position the blades are not restrained from flopping about.
There is a raised section in the Fly More case lid (red rectangle) that presses down on top off the Mini's body whenever the lid is closed. With the front blades crossed east-west they are trapped by that raised section and really can't go anywhere. Of course you always need to be careful the blades are not actually pinched between that raised section and the Mini's body. But with the blades positioned north-south they are free to swing when the case is closed and may end up wedged on top of the controller.
It's the rear blades that are more of an issue because they are forced to cross east-west. The Fly More case really needs to be modified to allow the rear propeller blades to be stored in a north-south orientation. I think that would make the biggest difference.
I wouldn't be too concerned over which blades go over or under the opposite side. On the rear blades, whichever side is being forced under will bend down, while the blades forced over will bend up. Maybe we should just alternate the way they're stored. One day up, the next day down.
In any case, DJI's label is obviously wrong (as has already been pointed out by others.) Here the label shows how DJI advises the left-rear blades are supposed to be stored, with A forced under B.
Clearly that label is inaccurate, because the leading edge of A cannot fit under the trailing edge of B. (i.e. A must go over B)
Similar threads
DJI Drone Deals
New Threads
-
3 Miami Tower and Miami Christmas presents coming soon- Started by Erk1024
- Replies: 2
-
What happened to Dark Theme on the forum?- Started by Droning on and on...
- Replies: 0
-
-
-
Forum statistics
