[vgwit]

After couple of dozens good flights got unexpected crash into a pond. Was in Rate mode and did nothing in terms of throttle or pick/roll. Did not switch to RTH mode but think it hardly helped.

https://drive.google.com/file/d/1miXLvf2txF9bNK41G-43IVWP8fpEQ1Ri/view?usp=sharing

The quad was in water about 5 min. All props got broken - it make me think they were rotating right after diving. There were no more mechanical damages. I disconneted battery, and after couple of hours blowed the quad with hot air.

Switching on showed not working camera and video transmitter. But I've managed to fix them replacing some electronic parts. So the quad looks healthy now, not tried to fly though.

My setup:
FC - Revo mini with Next FW and IBus connection
ESC - DYS BL20A with BlHeli firmware
Motors - BG2204-2300kV

So the question is what's happened?
The most probable I guess is a ESC-motors interaction. On CC3D I used PWM/490, but after upgrade to Revo got some motor stuttering with sudden crashes. Switching to OneShot125 cured the problem but obviously not completely.
Will appreciate your ideas.

[jdl]

I see tiny vibrations and occasional jerks in the video footage that post-processing stabilization could not cure completely. Seems no wind gusts, at least not significant (trees stay still).

Could this be a badly ballanced or damaged propeller that eventually broke at the end of that flight?

[vgwit]

Yes, no wind, no birds (I hope). And yes, the footage was passed through deshaker.

Props were fine. I made only 2 successful flights before that with take-off and landing from/into may hand. And one more thing. Before I switched from PWM/490 to OneShot125 I saw just abrupt stops of props spin and crash. But this time a strange shivering can be seen on last seconds.

[TheOtherCliff]

To me, crash looked like prop broke in the air or that motor screws came out.  Possibly even just that a prop came loose but didn't come off.

[vgwit]

All motor screws are in place. When I was removing prop remainings all their nuts were in place also.
A prop breakage could be. It would be good to know how often it happens. Props were Gemfan 6030.

Have looked once more over last seconds of flight and at 0:02:38:000 seen smth that possibly is a part of prop (crash2.png). On previous frame (crash1.png) and next frames (crash3.png) it also can be seen. Frame rate is 60 f/s. Could it be a bird?

[jdl]

... It would be good to know how often it happens. Props were Gemfan 6030.

I prefer using KingKong (LDARC) 6040 props instead (2204 2300kv motors). Better thrust, efficiency and overall feeling in the air and had never failed in flight (4-5 years so far).

Of course, I'm replacing ones showing signs of wear or damage.
Also always checking for cracks on the hubs or where blades attach the prop hub. These are usually caused by overtightening the prop nuts.

https://www.banggood.com/10-Pairs-KINGKONG-6040-Propellers-CW-CCW-For-QAV250-RC-Multirotors-p-988179.html?rmmds=search&ID=224&cur_warehouse=CN

[vgwit]

I also use KK 6040 and like them. But GF6030 shows slightly better efficiency (by ~10-15%) with slightly less thrust ( ~5%) in my prop tests. So I tried them.

[jdl]

I admit I'm maybe wrong about the efficiency, and charts say so.

However, I've once tested GemFans 6030 and was really disappointed with them. If I recall correct, I had to use more throttle to hover and the quad was noticeably more unstable in the air. Could not confirm longer flight time, compared to KK6040, if any, so I've ditched the GF6030 away and never used them again...

Btw, few months ago I've bought and tested APC 6040 Quad props (https://hyperion-world.com/en/p2766954-lp-b06040e-b4), hoping to get longer filght time on my ZMR250 builds. I've got this idea while playing in ecalc with different props. Sadly, I did not see any improvement compared to KK6040.

The quad PIDs had been auto-tuned for every different kind of props before flight tests.