Important note on capacitors, they have an effective voltage range. A 25v capacitor for example, is usually only rated for say 5-25v, Putting it on a 3.3v rail would not be effective. Using a 6 or 10v one would be a better choice. usually listed in the datasheet.
I imagine it's possible that we don't have Blackbox on many of the "cheaper," AIO FC's, because it might expose a lot more slightly defective units. This is just pure speculation, of course, but it aligns with what would likely be considered good business practice, from the perspective of a manufacturer anyway. Take for example the JHEMCU SP24S/PP24S ELRS receivers (which by nature, rx's get examined more closely, since you can't fly *at all* with a rx that wont bind,) that have been shown to have small issues preventing binding on certain firmware versions in some cases, and preventing binding entirely, on any firmware version, in others. From this example, it's clear JHEMCU are more than willing to ship large volumes of hardware with essentially zero QC, at least on the less expensive hardware. I wonder if they QC the boards that have BB included, but do not QC those without BB. Pardon my tangent here, but it was my first thought when I got to 2:24 in the video.
I have several quads on the same frame but a few with different flight controllers/ESC’s but same frame motor combo. I thought some of them flew better than others and figured it was a Gyro issue because same pids would get terrible results on identical builds and great on others but when I went to 4.4 and tried presets for the first time with your 5” preset it confirmed I’m just terrible at tuning 😅 thanks for your presets and tip videos
Very interesting Mark. I've used the Matek bec in this case. It has really nice filtration too. I would be really interested in your thoughts if you use one - don't forget to run a redundant ground.. *Like button smashed...👍🏻🍻👊🏻
So, had an issue with a Radix 2 v1.2 FC. Finally figured it out for the last six months until as of recently. Have v1.0 boards on most of my rigs that work great with no issues at all. Got a few of the v1.2 boards and on my 5" freestyle rig the quad would just flip out and the ESC would reboot. Turns out for whatever reason, the v1.2 boards have bad filtering on the FC. The only fix I was able to come up with was to direct solder a 35v 220uF Rubycon ZLH series low ESR capacitor directly up to the FC's VBat and ground pads on the underside of the board. That fixed it but really shouldn't have to be doing this period.
Have you tried to put an anti-interference shield between FC and ESCs to see if it can make any difference ? Maybe a spike aborber might also help a little bit, it is mainly supposed to protect electronic but reducing spikes amplitude might also reduce electronic noise. I have never made any mesurements, but I usually use both (shield and spike aborber) in most of my builds.
@@joshualiang2377 I use "DIATONE MAMBA INTERFERENCE ISOLATION BOARD V2", you should find it easily (I don't want my message to be deleted because of a link ;)
CAUTION powering FC with separate voltage source on a Cinelifter as you loose VBAT in OSD. Not being able to monitor your main LIPO is INSANE IMO. You might be able to use ESC telemetery. In my case I replaced the Matek F722-HD with a T-Motor F7 Pro Full Function FC and problem was solved.
For the two FCs you used, did you install them with the F7 MCU side facing down to the ESC? I heard if you isolate the FC MCU with the 4-in-1 ESC by using a piece of copper tape, this will reduce the noise from the ESC.
@@joshualiang2377 - the FC was adjacent to the ESC stack. The Matek was unflyable with VBAT from ESC stack. It was flyable with clean Pololu power but no OSD. The TMotor FC is fine with ESC VBAT.
@@pippifpv - timers are ok for freestyle quads IMO. BUT for an X8 with a $7K RED Komodo and lens... I must insist on being able to monitor VBAT. Especially if a cell drops all of a sudden. Just too much risk.
I felt for ya when you 'fessed up to breakin' stuff, Mark. Sux when it happens, coz we only got our own meathooks to blame. Upside: most replacement parts are cheap, and they get here pretty quick. Downside: I always feel kinda sheepish for a spell, so I curl up in a corner, muttering softly, and slowly shake my head until the feels go away - dunno about you. :)
I managed to eliminate the pitch noise almost completely on SpeedyBeeF7 V2, by just powering FC through an LC filter (220uF, 35V capacitor and 47uH inductor)
Hope my terraplane just aggressive pid tune. Not gyro. After a dive it hold altitude. I go zero throttle holds altitude till I tilt nose to get it flying again. Like a runaway but doesnt go up. I lowered D to betaflight default. Lowered P a lil lowered I a lil try keep ratios a lil
Ive had this issue with the HGLRC BMI270 flight controllers. I am going to try using a voltage regulator to power the flight controller and see if that helps.
Im having this issue on a Axisflying Argus F7 pro stack combo. 4 in 1 ESC, 65amp. Im not sure how to fix it. Im wondering if just swapping tye FC with a new one same model would make a difference?
I have a funny issue with my tmotor F7HD flight controllers with the BMI270, they both oscillate on take off and then calm down, I’m convinced it’s an electrical issue.
It is not really clear to me where would you conect the black battery wire to? The ground esc-FC stays the same then what about the battery ground cable? Thanks
I do not 100% understand your question, but know ground is ground. The only difference in ground pads is their current rating capabilities. Like the big ground pad on the ESC is made for big current. The rest, smaller current.
I'm having a weird issue and kinda think it's my gyro but maybe someone with more knowledge knows. I take off in angle, it flys level. Switch to acro for fun stuff, then coming in to land I switched back to angle and it goes sideways and backwards. Maxing out my stick for correction wasn't enough, had to switch back to acro. Any ideas?
I've got a iFlight F7 FC with a mean yaw twitch/drift I could send you if you wanted. I just wired up a cap to the 5v rail but I'm not hopeful its going to fix it.
I recently had an issue when I changed an ESC and started get shakes. Switched out the flight controller and the problem went away. Do you think there's any chance one of those diatone ground plane isolators that go between the FC and ESC would help?
You need to match the capacitance with the freq of the noise. You have low freq covered with the big cap on the battery leads. That is why adding a 330uF didn't do anything. You were on the right track with the tantalum. Adding a 10uf, 1uF, and a .01uF very close to the gyro will cover a wide range of freq. Good board mfrs should have done that already though.
No. It was still there when doing the roll. Noted the spikes on pitch when motors spin up to start and stop a roll. Finally, same moves on different quad and when FC is powered by a battery don't produce the issue.
I have better method how to test electic gyro noise. Open cli and write this set blackbox_mode = MOTOR_TEST ... save ... Then you remove the props, put the quad on something soft, and turn the motors with the keyboard arrows from 0% to 100% to 0% through the motors tab of the betaflight ... Uload the log to the pid toolbox-> spectral analyzer -> feq x time .The graph will show noise in quad control area 0-100hz if there is electrical noise exists. I have test like this zeus 35 "with bad gyro" ...
I like to pid tool box too. On my quad now I have noise around 60hz in a vertical line but only on like 30-60% throttle. And only on the dterm chart. What do you think about it?
It was a new build with tmotor FC stack. Been the cleanest one on logs I have done so far. After watching this video, I started thinking about the capacitor placement. And I’m gonna try a small one on the flight controller somewhere.
@@carbendi @Hurcan Emre Yılmazer I used a 35V 330uf cap on my speedybee f7 V2. It worked real well at clearing up noise in the unfiltered range between 50-100hz.
Important note on capacitors, they have an effective voltage range. A 25v capacitor for example, is usually only rated for say 5-25v, Putting it on a 3.3v rail would not be effective. Using a 6 or 10v one would be a better choice. usually listed in the datasheet.
What I learned from all this. Speedybee needs to give you money for doing their troubleshooting. Good job sensei.
Thanks for all the tips, Mark! 😃
Stay safe there with your family! 🖖😊
I imagine it's possible that we don't have Blackbox on many of the "cheaper," AIO FC's, because it might expose a lot more slightly defective units. This is just pure speculation, of course, but it aligns with what would likely be considered good business practice, from the perspective of a manufacturer anyway. Take for example the JHEMCU SP24S/PP24S ELRS receivers (which by nature, rx's get examined more closely, since you can't fly *at all* with a rx that wont bind,) that have been shown to have small issues preventing binding on certain firmware versions in some cases, and preventing binding entirely, on any firmware version, in others. From this example, it's clear JHEMCU are more than willing to ship large volumes of hardware with essentially zero QC, at least on the less expensive hardware. I wonder if they QC the boards that have BB included, but do not QC those without BB. Pardon my tangent here, but it was my first thought when I got to 2:24 in the video.
I have several quads on the same frame but a few with different flight controllers/ESC’s but same frame motor combo. I thought some of them flew better than others and figured it was a Gyro issue because same pids would get terrible results on identical builds and great on others but when I went to 4.4 and tried presets for the first time with your 5” preset it confirmed I’m just terrible at tuning 😅 thanks for your presets and tip videos
You were talking about a separate power supply for the FC, on e.g. a cinelifter, but what about a LC filter? any good? thanks
you are rocking them sweet t-motor props with you could find these 5143R can only get the updated S model
Very interesting Mark. I've used the Matek bec in this case. It has really nice filtration too. I would be really interested in your thoughts if you use one - don't forget to run a redundant ground..
*Like button smashed...👍🏻🍻👊🏻
So, had an issue with a Radix 2 v1.2 FC. Finally figured it out for the last six months until as of recently. Have v1.0 boards on most of my rigs that work great with no issues at all. Got a few of the v1.2 boards and on my 5" freestyle rig the quad would just flip out and the ESC would reboot. Turns out for whatever reason, the v1.2 boards have bad filtering on the FC. The only fix I was able to come up with was to direct solder a 35v 220uF Rubycon ZLH series low ESR capacitor directly up to the FC's VBat and ground pads on the underside of the board. That fixed it but really shouldn't have to be doing this period.
Have you tried to put an anti-interference shield between FC and ESCs to see if it can make any difference ? Maybe a spike aborber might also help a little bit, it is mainly supposed to protect electronic but reducing spikes amplitude might also reduce electronic noise. I have never made any mesurements, but I usually use both (shield and spike aborber) in most of my builds.
What shield are you using? Would you please share the link?
@@joshualiang2377 I use "DIATONE MAMBA INTERFERENCE ISOLATION BOARD V2", you should find it easily (I don't want my message to be deleted because of a link ;)
CAUTION powering FC with separate voltage source on a Cinelifter as you loose VBAT in OSD. Not being able to monitor your main LIPO is INSANE IMO. You might be able to use ESC telemetery. In my case I replaced the Matek F722-HD with a T-Motor F7 Pro Full Function FC and problem was solved.
For the two FCs you used, did you install them with the F7 MCU side facing down to the ESC? I heard if you isolate the FC MCU with the 4-in-1 ESC by using a piece of copper tape, this will reduce the noise from the ESC.
@@joshualiang2377 - the FC was adjacent to the ESC stack. The Matek was unflyable with VBAT from ESC stack. It was flyable with clean Pololu power but no OSD. The TMotor FC is fine with ESC VBAT.
Just setup a timer, either in the osd or following throttle position on the radio. Works perfectly on any opentx radio.
@@pippifpv - timers are ok for freestyle quads IMO. BUT for an X8 with a $7K RED Komodo and lens... I must insist on being able to monitor VBAT. Especially if a cell drops all of a sudden. Just too much risk.
I felt for ya when you 'fessed up to breakin' stuff, Mark. Sux when it happens, coz we only got our own meathooks to blame. Upside: most replacement parts are cheap, and they get here pretty quick. Downside: I always feel kinda sheepish for a spell, so I curl up in a corner, muttering softly, and slowly shake my head until the feels go away - dunno about you. :)
Yeah, there was lots of swearing.
@@uavtech U a professional engineer; this ex-Navy type can imagine. . . musta been 24-karat.
I managed to eliminate the pitch noise almost completely on SpeedyBeeF7 V2, by just powering FC through an LC filter (220uF, 35V capacitor and 47uH inductor)
Hope my terraplane just aggressive pid tune. Not gyro. After a dive it hold altitude. I go zero throttle holds altitude till I tilt nose to get it flying again. Like a runaway but doesnt go up. I lowered D to betaflight default. Lowered P a lil lowered I a lil try keep ratios a lil
As far as a FC with issues try and find a lumenier lux. That thing was maddeningly horrible with gyro issues.
Brilliant video thanks
Thanks for those great tips!
Ive had this issue with the HGLRC BMI270 flight controllers. I am going to try using a voltage regulator to power the flight controller and see if that helps.
Im having this issue on a Axisflying Argus F7 pro stack combo. 4 in 1 ESC, 65amp. Im not sure how to fix it. Im wondering if just swapping tye FC with a new one same model would make a difference?
I have a funny issue with my tmotor F7HD flight controllers with the BMI270, they both oscillate on take off and then calm down, I’m convinced it’s an electrical issue.
That or PIDs just a touch to high or Anti-Gravity to high.
Try lowing both, if still pretty much the same, yeah, personally I would log it.
It is not really clear to me where would you conect the black battery wire to? The ground esc-FC stays the same then what about the battery ground cable? Thanks
I do not 100% understand your question, but know ground is ground. The only difference in ground pads is their current rating capabilities. Like the big ground pad on the ESC is made for big current. The rest, smaller current.
Thanks, good info!
I'm having a weird issue and kinda think it's my gyro but maybe someone with more knowledge knows. I take off in angle, it flys level. Switch to acro for fun stuff, then coming in to land I switched back to angle and it goes sideways and backwards. Maxing out my stick for correction wasn't enough, had to switch back to acro. Any ideas?
I've got a iFlight F7 FC with a mean yaw twitch/drift I could send you if you wanted. I just wired up a cap to the 5v rail but I'm not hopeful its going to fix it.
Sure. Trying to figure out if I should get a V2 again to continue the quest.
117 South State Street
Box 504
Clarks Summit, PA 18411
I recently had an issue when I changed an ESC and started get shakes. Switched out the flight controller and the problem went away. Do you think there's any chance one of those diatone ground plane isolators that go between the FC and ESC would help?
You need to match the capacitance with the freq of the noise. You have low freq covered with the big cap on the battery leads. That is why adding a 330uF didn't do anything. You were on the right track with the tantalum. Adding a 10uf, 1uF, and a .01uF very close to the gyro will cover a wide range of freq. Good board mfrs should have done that already though.
Is there any way to fix the speedybee V2 pitch noise or should I replace it with the V3?
Maybe give this a try? www.diatone.us/products/mamba-interference-shielding-fpc
The fact that you pitched forward wouldn’t cause more noise on the pitch axis?
No. It was still there when doing the roll. Noted the spikes on pitch when motors spin up to start and stop a roll. Finally, same moves on different quad and when FC is powered by a battery don't produce the issue.
@@uavtech appreciate the explanation!
I have better method how to test electic gyro noise. Open cli and write this set blackbox_mode = MOTOR_TEST ... save ... Then you remove the props, put the quad on something soft, and turn the motors with the keyboard arrows from 0% to 100% to 0% through the motors tab of the betaflight ...
Uload the log to the pid toolbox-> spectral analyzer -> feq x time .The graph will show noise in quad control area 0-100hz if there is electrical noise exists. I have test like this zeus 35 "with bad gyro" ...
drive.google.com/drive/folders/1wFC_NM3vhBoUfGM0ziqPx__MBea8ImWe
I like to pid tool box too. On my quad now I have noise around 60hz in a vertical line but only on like 30-60% throttle. And only on the dterm chart. What do you think about it?
Prop wash?
It was a new build with tmotor FC stack. Been the cleanest one on logs I have done so far. After watching this video, I started thinking about the capacitor placement. And I’m gonna try a small one on the flight controller somewhere.
Adding a cap to vbat works best. This video has a great oscilloscope demo of the difference - th-cam.com/video/D9oVshtFcBc/w-d-xo.html
what cap size would be best on vbat for a 6s build?
@@carbendi @Hurcan Emre Yılmazer I used a 35V 330uf cap on my speedybee f7 V2. It worked real well at clearing up noise in the unfiltered range between 50-100hz.
@@UnmannedSky Thank you very much for the information i will try that :)
@@carbendi hope it works for you. You haven't lost much if it doesn't except for a few cents on a cap
We always do that. Filtration on the 3.3v rail is an internal problem with the FC in this case.