Two days ago finally I received my new SBUS receiver so yesterday I was able to change my previous FC (Flame F4). As I mentioned I had some video noise problem with that one. When I was removing the old FC I think I found the root cause of that issue. The VTX's positive power wire got off from the board easily without desoldering... It was just a weak soldering. But it's a different story

Anyway the change was quick and easy although I had to change all the ESCs' main power wires because all were short (the previous FC's power pads are located in the corners). I really like this board layout it's very logical. The most positive thing regarding this board is the soldering. It was easy to solder. Some AIOs are bit difficult to solder because the built-in PDB acts like a huge heat sink. I read some soldering problems with the Betaflight F3 AIO and I also had some issues with my previous Flame F4 board. Some grounding pads and the VTX power pads were a bit difficult to solder. But this FC was more easy. All the pads are large enough to solder well the cables.

Today afternoon I tried the FC in the air. It was also OK the video image is clear and everything was worked as expected. So currently I'm quite happy with this controller and I think I will use it in my future builds.

Good to hear others agrees with me. I am going to try to get camera control working on my next build. I think this is the first FC to have a pad just for this reason, and you don't have to use resistor on the OSD signal wire.

(11-Nov-2017, 07:13 PM)voodoo614 Wrote: Good to hear others agrees with me. I am going to try to get camera control working on my next build. I think this is the first FC to have a pad just for this reason, and you don't have to use resistor on the OSD signal wire.

You won't be able to use the DAC pin to control your camera's OSD until Betaflight 3.3. I just bought the Matek 405-OSD and it also has a currently useless DAC pin.

Today I had a strange issue with this FC. The first few flights were OK but later after a flip the copter fell from the sky like a stone. I was able to repeat this strange behaviour. During normal flying there were no problem but after a flip always fell down. Unfortunately I didn't have a laptop on the filed but it looked like the FC was frozen. After the crash I wasn't able to switch on the buzzer or arming the copter after a power cycle it was OK again. It was really strange.

You might want to take a Blackbox log. What is your cpu load? What loop time and gyro?

The looptime is 8/8k and CPU usage is around 6-10%. I think these are OK. I think I know what should be the problem. It's related to the capacitor and I'm going to check it this evening. After the crashes I saw that one leg of the cap removed from the soldering but I thought it happened because of the crashes. Maybe it was the cause ...

This was the reason why I thought this is a cap related issue:

https://github.com/betaflight/betaflight/issues/3319

But now I've found this:



and here is the explanation:

since the F4S have 2 Becs on board they are spec out for fast reacting to the voltage change to the input voltage, to compensate to the voltage change, to get a better and cleaner video signal on camera and VTX.here is the drawback, it also fasts reacting to the cut off voltage too. when some ESC, when it accelerates the motor, the dead time is not set properly causing the power surge down to 7v or under. the BEC react and cut off the voltage supply to the FC. causing the power outage for a split of a second. that causing the main processor to cut out or lock up. so fix for some esc have this problem is add small cap to each esc. don't need to add a big one to the battery lead, only on the each ESC.

I'm using racerstar Tatto 35A ESCs.

Where does that explanation come from?

I'm suspicious as to how correct that is. If the dead time configuration for the ESCs was incorrect, it would destroy the FETs on the ESC fairly quickly regardless of whether you add a capacitor to the ESC or not. In fact, if the dead time setting was the problem, adding more local capacitance would make the shoot through caused by the incorrect dead time worse!

I will agree that lots of noise from the ESCs could cause problems with the flight controller if the noise affected the onboard regulators badly, but that's specific to the regulator used and I have no idea which switching regulator chip the CLRacing F4 uses. Suppressing ESC noise is best done by adding capacitors at the ESC. The general rule is that the capacitor should be as close as possible to the source of the noise to do the best job.

The other thing you might like to try is to disable DMA for the onboard microSD card. Some flight controllers have a problem with this, especially if you are running the ESCs with DShot. If that's the case, use the CLI to disable DMA for the black box with:

set sdcard_dma = off

The video is posted by Cheng Lin.  If it is truly Cheng Lin, he created the flight controller.

EDIT:  I guess it is really Cheng Lin.  He also mentioned to 'avoid using Emax bullet 30A esc. and Kiss. ZTW, and little bee".

Hmmm... you'd hope he knows his electronics...

The only way to confirm this would be to use some BLHeli_S ESCs. If you get the problem, flash them with some firmware that matches the MCU type and pinouts, but with a larger dead time and see if the problem goes away.

Well I also think it's the noise generated by the ESCs during the breaking which causes the FC lockout. Which could be a good news because it can be mitigate with capacitors. But I'm a bit worrying because somebody who has the same issue and he added 220uF cap for each ESC but still has the problem.

Anyway I will try to add caps to each ESC and I keep my fingers crossed. But what do you think about his advice about not needed big capacitors just some tens of uF will do the job? It would be good because my frame is a bit thight but I afraid I need bigger to eliminate the noise...

My last question is could you recommend me a reliable FC with built in PDB and OSD? : )

This was supposed to be the board to rule all other F4 board (just my opinion). Did not have any problem with the original CL Racing F4. The only other FC I would recommend is the Matek F405 AIO. The only problem with that board is, no second uart inversion and the lipo pad has to be pointed back and can be a problem for certain configuration.

I have to say that I've not experienced this kind of problem myself so I've not been forced to get the 'scope on any of my quads and actually investigate this closely.

Would a 10uF capacitor work? I honestly don't know, but it's probably worth a try! Even a 100uF 35V capacitor is fairly small if space is limited.

I had major yaw twitch problems with the Matek F405 AIO in my talon hypebuild. Unseen took a look at it also. I'd have to suggest avoiding it also due to the 32kHz gyro. Joshua Bardwell had some twitch issues with his non AIO matek F405 in his last video where he finally flies his talon, and he also mentions problems with the Matek F405 AIO.

I'll be fitting the Matek F405 OSD and matching PDB to my Chameleon some time soon, so it will be interesting to see if I get any of these problems.