[anthony.debiase]

I live these nice f3 and F4 FC's but hate the cost. I'm putting my ideas and knowledge into making a rippin fpv fc that doesn't cost too much and doesn't die in a week.  Here is the GoFundMe link. https://www.gofundme.com/41yfnbc

Sent from my LGL62VL using Tapatalk

[TheOtherCliff]

It would be best if it was compatible with an FC that we support or plan to support, otherwise it might be a long time or never for LP to support it. 

[anthony.debiase]

It would be best if it was compatible with an FC that we support or plan to support, otherwise it might be a long time or never for LP to support it. 

Could the board be designed as to ease compatibility or would I need to be developing lp and firmware for the board?

Sent from my LGL62VL using Tapatalk

[Mateusz]

I am not developer so I speak only for myself, but I guess there are things you should consider

• There are already many very good flight controllers on the market that are cheap (see attached picture taken from http://www.gearbest.com/flight-controller-c_12223/)
• Developers may want to contribute if it has interesting features and large user base or potentially large user base

What price range would be affordable ? Assembling Revo on your own should not cost more than $90 for components from digikey (see BOM prices when buying NOT in bulk) and $20 on 3 PCB boards from OSHPark, so pretty affordable to make prototype nowadays. It's VERY affordable to make an interesting prototype, hence such competitive market. Good FC are ~$40-50 on the market and some like this one for $52 includes ESC as well in that price https://www.banggood.com/20x20mm-XJB-F425-Dshot-Flytower-Micro-F4-AIO-OSD-BEC-Flight-Control-25A-Blheli_S-2-4S-4-In-1-ESC-p-1148284.html

It's hard to beat that price, they are so cheap and good.
 
Do you have a prototype to show ?

[anthony.debiase]

No prototypes yet. I already have a decent surface mount rework setup with bunches of resistors, capacitors and inductors of various values in both 402 and 804 sizes.  I have done custom pcb design and fabrication as well.  Plenty of work with computer programming, hardware and the evil little arduinos are nothing new.  When I say affordable I mean probably $20-25.  The revolution I loved was ~$50 and the GPS for it was around $10.  But I would buy many more if they were only $30 for the FC. I would even go as far to say doesn't need oplink modem nor barometer.  I believe airbot makes an F4 that's those specs.   So I guess the goal is a decent FC at half of what's normal cost on market. I am satisfied plenty with a cheap C300 atom on eBay for $12 but they seem to die often. One good impact and gyros stop sending data and no more ability to arm or fly after that.

Sent from my LGL62VL using Tapatalk

[Mateusz]

How about these ? They look decent, first one I have and I am flying, really happy with it has neat bootloader button instead of pads.

• HGLRC PBF3 EVO F3 Flight Controller  - 100 SEK is $11 after conversion
• Mini Omnibus F3 Brushless FC with Integrated OSD - 164 SEK is  $19.40

I mean on BG you also get Omnibus from Airbot that includes OSD for $22 https://www.banggood.com/Betaflight-3_1-F3-6DOF-Mini-Flight-Controller-20x20mm-Integrated-OSD-Built-in-5V-BEC-p-1119503.html
Airbot also makes F3, F4 like you noticed and even F7 which I have in a package (see attached photo). Some are stackable with ESCs so no need for connectors.

All these MEMS gyros use the same physics, they all go bad if you hit really hard, in some manual from InveSense they state how much banging they can take I remember.

[anthony.debiase]

The datasheet PDF for the 6000 and 6050 chips are actually on my desktop computer now. The weird thing is, the cheap eBay cc3d died after pretty minimal impacts. The revolution however was on my spedix s250 frame and I had insane impacts with it and it kept going with no problems.  I've also killed the cheap cc3d mounted the same way on that same frame.  It has to be the board, solder connections or some other non-gyro related defect that ends up killing that chip.  That revo boards was piloted by my uncle into the side of his hi house at great speed.  Fthe little board that the battery strap connected through was on small short stand offs and those standoffs sheared off sending the little board and battery flying. It was a crazy impact. How the frame didn't explode is a mystery. But the revoboard lived on. 

Sent from my LGL62VL using Tapatalk

[anthony.debiase]

Well since you showed me some devices I might like in a pleasing price range I am starting to wonder if I need to make my own boards at all.  To sell them cheap I would have tiny tiny profit and this technology is developing fast enough to make it futile to do the project alone. 

Sent from my LGL62VL using Tapatalk

[Mateusz]

I certainly didn't want to discourage you from the exciting project !
Only wanted to guide you what is available on the market and some of these already supported in next development branch.

Having said that, I am pretty sure there is and will be demand for new flight controllers that are more unique and have interesting features!
Some integrate other modules, have soft-mounted gyro, sdcard, multiple gyros, integrated radio, even integrated vtx.
Range of possibilities and new ideas is growing. For example InvSense is planning chips with gyro+accel+baro in one package (no one uses these yet, but samples are not yet available). There are interesting modules like Murata's CMWX1ZZABZ-078 integrating 898Mhz RF module and F0 mcu so whole OPlink in 11x10mm package. There are nicer sdcard sockets, connectors which can be used for stacking boards and so on.

[anthony.debiase]

I just built a quad using the kiss cc which has 4esc built in and is only 11g. I like all in one designs not price tags.  I often wondered why there isn't a board already made that has no gyro/Accel onboard but has that chip on a separate pcb that uses flexiport or something allowing for optimum gyro positioning. These pcbs with sensors might also be flavored so that some only do Accel gyro data but other have compus or barometer. That would be useful beyond imagination .  Integrating vtx is cool but useless in some frames.   I love the simplicity and size of combining different electronics into the FC, however when the FC dies it isn't all dead.  So adding more things into it creates more ways to fail or croak, this opposed to a system where it's somewhat module based (like it started out as) and when modules die they are independently replaced while keeping the working stuff .   The FC really should only be tasked with taking rx and sensor data and running it through the pid voodoo and outputting something to the esc's.  The sensors, rx, and esc's should be separate.  That's years of university science lab experience talking but logic says that's the way we should take things

Sent from my LGL62VL using Tapatalk

[TheOtherCliff]

optimum gyro positioning

It doesn't matter where the gyro is mounted, other than it can't stand excessive vibrations.

For accels, they need to be at CG.  For mags, they need to be away from high current wiring.  For baro, it needs to be away from prop blast.

[Mateusz]

For baro, it needs to be away from prop blast.

And away from blinking LEDs as baro is light sensitive

[xfce]

i wonder why so many copter controller use the STM32 chip,it is boring. could someone develop a controller use ADI 、MAXIM，or Freescale's chips. i think

[Mateusz]

i wonder why so many copter controller use the STM32 chip,it is boring. could someone develop a controller use ADI 、MAXIM，or Freescale's chips. i think

I also like Freescale chips, the reason for STM32 is the community around it I think and documentation to start with. Another thing is that there are already many firmwares/drivers written for STM32 which serve as a good example. All FC I've seen are STM32 based on the market, so writing firmware and maintaining it for another one, might not be optimal. It is kind of boring, but it's well proven to work

We should team up and make nice hardware for LP that stands out

[anthony.debiase]

 so freescale chipsets is what the people want. alrighty then.  and perhaps a group effort is better. i like different perspectives on projects.  keeps them from becoming dulled with a single persons biases and knowledge.  unless the groups is fighting and killing eachother instead of producing something, a a group should output better quality than just a single dude. 

Sent from my LGL62VL using Tapatalk