Wishlist for a custom Transmitter board

FWIW other similar ideas now discontinued but some of the ideas, pitfalls could be used - Gruvin9x gruvinandco.com/shop/ and Sky9x openrcforums.com/forum/viewtopic.php?t=1497 and www.ersky9x.net
I actually have a Sky9x board in a 9x and was happy with it until Deviation. If Deviation had a board I wouldn't hesitate to get one. If it would help I would ship the tx anywhere for someone to poke around at.

Hmm from the Gruvin website ...
<Quote>You can still build your own G9X board!
The G9X v4.x boards are fully supported by Open-TX. Full design data (KiCAD) for the final version (v4.2d) is available here.
PCB's can be ordered ecconmically from ITead Studio (or your preferred prototype PCB supplier) by supplying them the Gerber files from within this zip file.</Quote>

Does this have the space for a serial TX line for an audio processor, or do we use an ADC and do the audio processing on the CPU? With internal processing I suppose we could use a bluetooth connection to a speaker or headset in addition to an internal speaker. With an audio processor we require limited CPU resources, and may impact on power and space requirements.

PhracturedBlue wrote:

Support for 2nd LCD/OLED display - Is this actually useful?

After getting my X9E I'd say YES IT'S USEFUL! A second display like an OLED is a great feature (actually it's the reason I ponied up for the E over an X9D)! I only wish the one on my Taranis E was more useful than it already is (OpenTX devs promise it's coming). I wish it would display the model's battery level, currently the only telemetry data it displays is the RSSI (the number value and a pretty little bar graph), the rest of the data is internal transmitter data (tx battery, a useless lifetime counter and a resettable countdown timer).


My concern is the form factor. Is the idea here to make an AR9X style board that fits perfectly in a transmitter body or will it be a small square (actual shape irreverent) board that could fit in any transmitter? I believe one of the biggest advantages of this system would be the ability to deviate ANY transmitter, even a toy grade RTF packaged tx.Is that kind of where this idea is going?


Edited to add:
Here's an idea for the USB connectivity


This would allow the board to be plugged in directly or a port to be easily wired to the pads (I dont know if you guys have ever used this type of full-size USB connection before but it works fine on it's own and is also easy to solder to).

PhracturedBlue wrote: {snip}It does has the benefit of making initial firmware installation very fast though (assuming no usb plug is on the board itself)

That was my thinking. I like to use boards with those connectors for all my embedded projects, for the initial programming I just plug the entire thing into my computer and then during install I wire a small USB micro port to the easy to access pads with small (30AWG) wire that I can place where ever I want. Super simple and the pinout is impossible to get wrong as it's all straight across (D- to D- and D+ to D+)


Just an idea, I wont be sad at all if it's not used...

My only input on this would be to increase digital inputs from 16 to 32. 16 Digital inputs amounts to only 8 two-position trims, or 4 trims and 8 pushbuttons or switches. This number seems extremely limiting.

For me the CPU board should be future ready, so there should definitively be enough spare ports pins connections memory calculation power plenty of, IO, ADC and more than what you'll now ever can think of. Doing that it means there is always room for creativity and implementation of ideas later.

Size should be small enough to fit in devo 7e.
Power not higher than 5V. 3V3 would be best.
It should be fully modular. No custom pcb needed! Only wiring to be done, to build the TX.
No need to have all features onboard, but use of standard on the market available modules for, display, RF, wifi,, Bluetooth, GPS, V regulator, sound, buzz, vibr, etc etc,
Price of the total hardware for the build <50Usd, to attract max of people.

Doing this leaves room for future development, some may like a big color touchscreen, some other may prefer monochrome small Display, should be possible. Flexibility, to have room for customization changes that's a good thing!

On extra ideas, not to be implemented by default, but to have just the plenty of io available, or possibly some mux?

What we could think off out of the box, for later, to have the pins?
WIFI/Bluetooth
GPS (for range to model and clock logging sessions etc )
MEMS IMU
RF module at least space for 5, many also like to add an UHF LRS module over SPI.
Outputs: Trainer port, PPM port, and digital port like SBUS.
Audio module
Force Feedback

i think this would be an awesome project, and it would give me something to finally get some good out of my pcm9xii internals again lol. though i think trying to cover many different combo's of parts may be a headache, perhaps settle on a base model and get it working first, then start expanding outwards. ie walking before running with that in mind are there any dev boards that fit our bill? or would be looking at a custom board from the get go? unfortunately my coding and electronics skills will limit my input to testing so i'd be fully reliant on others bringing this to fruition

For the average person this will be way to complicated which will limit and eliminate the number of Deviation users..The great thing about porting to a already built and functioning tx is what makes deviation such a great success, making it to where a person will have to build boards ect will kill it...

+FLightlesBird, you miscounted. 16 digital inputs is 16 two-way switches, as each input is either on or off, not 8, or 8 three or four-way switches (a three-way switches uses two pins, and ignores one of the four possible states).

The comments about "a dev board" are spot on. Someone has already posted a video to youtube of a proof-of-concept port of deviation to the STM32F429I board. These are relatively cheap - you can get them four around $20, include a (small) touch-screen color LCD, multiple USB ports, all pins broken out except maybe some tied to onboard devices. These are STM32F4 boards, so more CPU than in anything that deviation currently runs on. You can go cheaper and drop the LCD and get a smaller/slower STM32F4, or more expensive for a larger screen, on-board SD cards, audio processors,. They're all roughly the same form factor, which I wouldn't expect to fit in a 7E, and might even have issues fitting in a 10 case.

The STM32F3 "gumstick" boards should fit in a 7E easily, but you don't get the CPU boost or a complete set of pin breakouts. But they're cheap, costing around $3 in onesies.

One of the things on my todo list is moving as much of the hardware connections info into hardware.ini as possible. The primary goal is to make it easy for hardware modders to add inputs - the 7E & 6S extra switches, for instance, the "ultimate 7E" extra switches, the guy on rcgroups wanting one 2-way and one 3-way on his 7E, etc., but it should also make setting up a tx based on such dev boards reasonable without having to have a custom build. Supporting those might be a first step towards a custom board like this?

If you see how succesfull Arduino became and how many modules boards are available, as long as modules are standard parts can be stacked toghether easily. I think many will follow us, especial RC modellers, who are technical guys who like to build and learn from that. It is part of the hobby.

The lazy guy, or the sponsored flyer, etc they will never switch to such platform......
They stick to a nice futaba or buy a Spektrum etc and it works out of the box. Those people anyway will never switch to deviation.

If modular Deviation Tx become a success, it is not unthinkable some Chinese, will just bring on the market the full deviation board, as you see that with many DIY projects.........For the young generation, I see open source from hardware to source is widely becoming accepted.


.

the stm32f429i dev board mentioned by mwm seems like a really great candidate , the board is ~119mm x 66mm and also has a built in stlink v2. the touch screen is situated off center so if used in a d7e (if it fits) or a d10 chassis the screen will need pulled from it's header and some interlinks made up but that's nothing too difficult to accomplish.

another alternative is the stm3210e-eval of which the newer version uses an stm32f103zgt6 mcu, which is a 144pin version with 96k of ram and 1mb of flash, it also has support for much the same peripherals as the f4 version like external sd etc too, it is also in the same family of existing devo processors which should make porting to it a little easier I would think?

This sounds similar to what I was trying to do a couple of years ago. I tore apart some old TXs and built my own interface boards. Mine didn't have LCDs, and were much simpler, but it was fun to experiment with. That said, it's not fun to dig out the old board and resolder all the wires, etc. I don't see this being useful for a large audience. Deviation is what drew me away from building my own control board...

I just received a couple of STM32F7 discovery boards, and thought that would be a nice board to build a TX around. I wish it was cost effective to build a case with high-quality gimples and switches, etc around that.

Just came across the top model Transmitter from Jr, it is an android controlled Tx. Price 2600usd!!!! OughF www.jramericas.com/234048/JRP01050/?pcat=2

But the idea could be interesting also for the future deviation, basically use a small ARM development board to connect the hardware Tx modules and all inputs, switches and then use a cheap android tablet or phone as control and display for Tx. There are already chinese models available from 50bucks..... We get then a lot of features directly and can keep the hardware setup more simple.

Speakers, Mp3
Vibrator
GPS
Touchscreen color
Bluettooth/Wifi/3G
Plenty of storage memory for logging etc.

Only downside is to start your tx, must, boot start the app etc, might take bit longer, complicated, less easy etc.

I remember seeing a video on y-tube were people mounted tablets in the dash of there cars replacing the radio and it worked quite well..There's actually a company that makes custom install kits ..

OK first, to be clear this IS NOT INTENDED as a solution for this project! This is something I've wanted for a while now and the work by pb, slipstream and HappyHarry have made it now worth it for me to put this together.

Goal is to build old crappie useless tx's into ultimate 7E's.
Zif socket for 7e display and flash chip still to be added as well as devo u7e specific labeling in place of stm32 labeling, this is just the first draft (ok that's a lie, this is actually v06 lol)