100% the right way and spinning the correct direction too. Not a stupid question, I would by lying if I said I had not gotten this wrong before!
2 Likes
If you get bored in the meantime, flash 3.5.7 and leave as stock apart from the tx/rx and see if it does the same
1 Like
Can you post a vid of the receiver tab showing the stick movements?
Sure why not, only so many times I can try+fail to do a power-loop on this mob 6 before it gets old. Still not sure what PIDs I prefer, need to just stick with one and stop faffing! Tomorrow though. Thanks for the help.
Fetish for receiver tabs is it? Why notā¦
1 Like
I can see that props are on correctly BUT are you sure motor direction is correct?
The only time Iāve ever seen YAW like this is motors wrong OR a prop is loose on the shaft.
Maybe worth checking again 
Itās really hard to see what might be wrong here. I think if one of the motors was spinning the wrong way then it wouldnāt even do the hops itās doing there, it would just spin that motor right up as it tried to lift off.
One other thing that might be worth checking is that the motor numbering is correct, do this with the props off, battery connected and into the motors tab and check by spinning them up one at a time that each motor is where itās supposed to be.
While youāre there you could also perhaps try calibrating your ESCs just in case one of them is off for some reason. Other than that iām out of ideas unless itās some actual hardware issue.
Have you fitted foam over the barometer, to prevent turbulence affecting its measurements ?
This might help with your earlier problems of drift and not been able to open the throttle.
Keep us posted on progress, even if it turns out to be too daft to admit, weāve all be there plus you get a badge.
Definitely spinning the correct way and motors are correctly numbered. Props not loose.
I had already tried another clean reflash and have the same issue but Iāll try a clean back to 3.5.7 per the suggestion above by notveryprettyboy later today.
Am baffled. Originally I thought the FC was toast but considering all the sensors seem to be working properly and I can spin up all the motors perfectly well on the bench to whatever rpm I like (and they idle really well with very little juice (smooth spin at around 1030 from cold start) so the signalling seems fine. WTF 
Admittedly no, but I put o-rings above and below the FC, the FC is well spaced above the PDB, and the FC is wrapped in a single layer of Kaptop.
What kind of foam? I worry putting too much in between the FC and PDB will affect the heat dissipation of the PDB and the VTX behind can get very hot when a pack is on.
1 Like
Are you taking off in Angle mode or Acro mode?
That video was horizon I think, as thatās the default switch and I forgot at the time when I was shooting the quick video for the post. But problem still appears in acro. I get the āhoppingā behavior I was mentioning previously.
Does the quad still spin around after take-off in Acro mode?
The yaw spin isnāt common on most of the settings and I think the wind was making that worse last night. Usually I get a wobble on the one corner where the prop starts to spin erratically. But that drop is not me cutting throttle, thatās me pushing the throttle up further to say 15%, it just cuts until you reduce the throttle and throttle up again.
If you can actually see/hear one of the motors acting erratically then Iād be suspecting the ESC on that motor is having problems when under load with the prop on (bench testing without the props doesnāt really bring out those sort of issues). It could also be a dodgy motor but Iād suspect the ESC first. Either way without all 4 motors behaving properly it would make the quad unable to control yaw correctly. Iām not sure whether the FC would then detect the error (via telemetry) and disarm the quad - if it did then that would explain the falling out of the sky. If you have a spare ESC Iād try swapping it out first as everyone else has asked about the FC config which seems fine.
You need a soft OPEN CELL foam, CLOSED CELL will NOT work.
You can test it by gently putting to your lips and breathing out, if your breath passes through the foam you are good to go.
I find the foam you get in product boxes, RUNCAM cameras for example, is good.
You need a piece about 25mm square, positioned centrally over the baro (thatās the metal can component with a tiny hole on the top (must have the hole!) and thick enough to be compressed at least a third when fixed in place.
HTH
Steve 
1 Like
Great advice thanks Steve will sort when I take it apart later. For testing purposes can I just disable the barometer for now or is it integral?
Ok, lunch break so having a quick toy.
No luck with a reflash and wipe.
After a wipe I have only added the UART 5 as VTX, changed the receiver to CRSF, and assigned a channel to arm mode then I shot a new video so you can see. I think youāll understand why (likely wrong) I originally limited the throttle as you can see it goes to shoot straight up.
This is 3x arm attempts
To be clear, this is me arming the drone, and then disarming in the instance where it goes really high, but otherwise doing nothing else. Throttle is at zero (1000) the whole time. Itās hopping and taking off all on itās own hence the spotter with the camera so I have my finger ready to disarm.
I also note the motors got pretty hot even in this brief test.
SD card is due to arrive this afternoon so will get some blackbox data this evening. I donāt have any spare matching ESCs to try that.
diff
# version
# Betaflight / STM32F7X2 (S7X2) 4.2.8 Feb 15 2021 / 12:10:35 (101738d8e) MSP API: 1.43
# config: manufacturer_id: MTKS, board_name: MATEKF722, version: f173f9bc, date: 2019-10-02T14:43:13Z
# start the command batch
batch start
board_name MATEKF722
manufacturer_id MTKS
# serial
serial 4 8192 115200 57600 0 115200
# aux
aux 0 0 2 1700 2100 0 0
# master
set acc_calibration = 44,26,-3,1
set serialrx_provider = CRSF
set gyro_1_align_yaw = 1800
profile 0
rateprofile 0
# end the command batch
batch end
Are the escās on the latest firmware?
Yep. All were successfully flashed though BLHeli32 Suite before I reversed a couple of them. I did not mess with the frequencies thoughā¦ this has been nagging at me, but no one has suggested it yetā¦
1 Like
Itās not this 
1 Like