Just Another Fatshark Receiver Module (JAFaR) - DIY OpenSource 48ch receiver OSD integrated

This is not the support thread for the realacc module, so, please, stop asking for feature or complaining about the problems you have with the modules you bought already assembled by 3rd party which has "cloned" the project (it's actually a different board, as far as I understand, so I don't even know how it works).


there a lot of different project out there to build an internal Fatshark receiver module... so why am I here? Ok so, let's start from the very beginning...

As far as I know, there are no a FREE “perfect fit project” with my rationales:

• Opensource (both software and hardware)
• easy to develop, upgrade and maintain
• community driven
• cheap

technical details:

• 48ch (8 channels for 6 bands, including race band 1 & 2)
• auto search feature
• no external switch, or hall effect sensor, or buzzer, it must use *only* the goggles buttons (up-down)
• it should include the “scanner” feature (see my recent project here https://github.com/MikyM0use/OLED-scanner)
• frequency selection via OSD inside the goggles
• i want to keep the “door”, so it must fit just inside the goggles! (this feature is not implemented at the moment)

I based all my project on the “rx5808” receiver module, many people just insert the module inside the goggles, and use it out-of-the-box with the limitation of 8 “boscam” channels.
I tried it, it works. Nice, lets move on.

I discovered that the RX5808 receiver module can be software programmed by the “SPI” option (the module requires a small modification, more details later).

So what I needed is a micro controller with an easy and well-know software IDE… nothing better than Arduino. So I decided to use an Atmega 328p as micro controller.

To select the frequency I used a series of menus on the screen inside the goggles, the useful TVOUT Arduino library is perfect and requires just a couple of resistors (http://playground.arduino.cc/Main/TVout). A digital switch (TS5A3359) allow to switch from OSD to the normal receiving video from the quadcopter.

What I’ve done so far:

PCB version 0 (just a proof of concept, no micro controller involved):


bug/corrections:
I discovered (I should say “I read the fu**ing manual of the RX5808 module and I see) that a capacitor is required to eliminate the DC component of the output video signal, and one capacitor is required to filter the input power source. It’s a pity that I ordered the first version of the PCB before see the “capacitor” problem -.- so, this is the first version on PCB:



and inside the goggles



this is the very first version of the software, recordered directly from the Fatshark DVR (I use the HD version)

OSD menu - DIY 40ch Fatshark rx module (0 min 35 sec)

the percentage shows the RSSI signal for each band/frequency, so you can guess what channels are used and what are free.

bug/corrections:
I forget to check the labels on the solder mask, so some resistor are “not named”. And the LED is on the wrong side of the PCB and wrongly connected to the ADC7 pin of the Atmega, which I noticed after it’s only an analog input.


I’m waiting for a 1.1 version, which will have the same LED issue, but with the integrated capacitor.

---------------------------------------------UPDATE march 18th 2016----------------------------------------------

I tested the 1.1 version!



it seems ok, so I created a repo on github:

https://github.com/MikyM0use/JAFaR

Please notice that this version of PCB still has the "LED ISSUE" (I used a wrong pin on the microcontroller). it's still a beta version, so use it at your own risk and only if you know what you are doing.

I also discover that the BIG capacitor on the video output is useless, so use the series resistor (10ohm) insted!

I decided to develop a diversity version of the module: the base module is a simple rx, but you can connect a second rx module on top of it (even when the goggles are already working) and the second module is automagically detected and used.


If you are in a hurry you can use the PCB v1.1, but if you have just a little patience I will share the diversity version as soon as it has been shipped and tested (and the source code developed ). I already ordered it from OSH, it should be shipped on march 25 taking at least one week to reach me in Italy.

---------------------------------------------UPDATE april 15th 2016----------------------------------------------

diversity PCBs are here!



yesterday I tested the base module (only one module without the diversity "daughter board"), it works!!

I written some lines of arduino code, you can find it on github. I upload also the new PCBs on github.
my daughter board has a little problem with the antenna connector (it's too near to the module) but in the version uploaded on github the problem should be solved.

let me just check tomorrow the module with the diversity board to be sure all is working. Also, you need some source of external power (I put the module as sandwich just to develop the code, DO NOT insert the modules with the internal power of the goggles).

---------------------------------------------UPDATE april 19th 2016----------------------------------------------

the diversity module needed a major revision, this is what I added:

1. integrated linear 5v regulator with ENABLE feature connected to the VCC of the main module
2. a buffer for RSSI signal. Maybe the new RX modules are more delicate? don't know, btw I inserted a pad to bypass the buffer if you don't want to use it
3. a jst connector
4. moved a bit away the antenna connector

please notice that because of the linear regulator it's not possibile to use a 3s battery to power the diversity module at the moment.

MAIN MODULE BOM
https://docs.google.com/spreadsheets...it?usp=sharing

DIVERSITY MODULE BOM
https://docs.google.com/spreadsheets...it?usp=sharing

I'm pretty sheptic on the jst connector, I think I will rework it as soon as possible... found a compatible connector in BOM list

---------------------------------------------UPDATE may 11st 2016----------------------------------------------

ooook, it's a pleasure to say the diversity module PCBs are here, and they're TOTALLY WRONG I've never seen so much errors in a such small board, really sorry guys, I designed them too quickly so now I need to rest and start to fix all this mess

---------------------------------------------UPDATE may 27th 2016----------------------------------------------

I received the PCBs from dityPCB, the "testing version" seems to work, the only thing changed in the base module is the add of the buffer (optional) and I also soldered the diversity module, it seems to work and it's back compatible with the older versions of the base module ("stable" version for sure). BUT there's a problem with the regulator, I needed to add a resistor to the "enable" signal, btw I already put the resistor on the pcb V2.1 of the diversity.

V4.0 of the base module and V2.1 of the diversity could be the release candidates!! NOW TESTING PHASE!

---------------------------------------------UPDATE june 28th 2016----------------------------------------------

I added the support for an external OLED SPI module in case you have problems with OSD or to use the module as standalone

it works for the band/channel selection. It's just a matter of pins connection between the base module and the OLED, so another pcb is not needed (even if I think I'll make a pcb for this module as well).

more datails here and on github readme too:
https://www.rcgroups.com/forums/show...&postcount=113

---------------------------------------------UPDATE july 9th 2016----------------------------------------------

It's now possible to use the module as standalone (without the goggles). More infos on github (https://github.com/MikyM0use/JAFaR)

---------------------------------------------UPDATE December 7th 2016----------------------------------------------

thanks to the great job of a friend Alessio Molteni, here you can find the 3d-printable doors for this project: https://www.thingiverse.com/thing:1946290


----------------------------------------------------------------------


FAQ:

why the OSD is on the blank screen?
it’s simpler because one cannot have to recognize the vsync and the hsync pulses. This is why you cannot see images from quad in the background.

Would it be possible to add these new band/frequency in the next firmware update?
there's only 8 position/channel in the goggles, you cannot recognize the "pression" of a switch, just the selected channel, and there's no more RAM to implement another sub-menu. So if you want a $INSERT_NEW_FEATURE_HERE menu item you have to delete another item (there must be always 8 items for each menu).

please note that you try and use this project AT YOUR OWN RISK!

Great job! Tuned ...

this looks really kool if you can keep the documentation for dummys id appreciate it

will try to do this

Very interested. Will build as soon as PCB is shared.

first post updated, you can find hw and sw on the github repo, but if you can wait a little more I will share a diversity version as soon as possible (estimated ship on march 25th)

https://github.com/MikyM0use/JAFaR

awesome! temping to build now, but i'll wait for diversity since I have 2 rx5808s. maybe ill order the parts i don't have instead anything special on the diversity board besides another 5808 and SMA ?

on the diversity board I'm trying to use the TS5A3359 switch instead of TS5A3157 since I now need to switch over three different video source (the two different modules and the OSD to select the channel/scanner/autoscan)...



I think in the future release to add a linear regulator to the "second" rx module, since I don't want to put two modules on the internal regulator of the goggles.
I'm thinking to use the balance connector of the main goggles battery...

This is the idea for the rx:




and this is the rx diversity module...

Cool, looking forward to more from this project.

Subbed

nice

any luck getting the boards from oshpark?

don't ask me -.- I thought they lost my shipment, so I wrote to OSH and they confirmed "yes, we lost it", but when I was replying to them, the shipment magically appeared in the italian "post carrier" tracking site... now it should be in Milan (north of Italy) for customs...

it's a pity because as far as I know all the other diversity modules has been experiencing troubles with the internal regulator of the goggles and this module should be the first that allows the use of an external regulator for the second module.
I wrote in a previous post that the next version will have a linear regulator on it, but I want a proof of concept in the first place using a switching regulator to be connect to the balance plug of the battery.

Originally Posted by miky mouse

don't ask me -.- I thought they lost my shipment, so I wrote to OSH and they confirmed "yes, we lost it", but when I was replying to them, the shipment magically appeared in the italian "post carrier" tracking site... now it should be in Milan (north of Italy) for customs...

it's a pity because as far as I know all the other diversity modules has been experiencing troubles with the internal regulator of the goggles and this module should be the first that allows the use of an external regulator for the second module.
I wrote in a previous post that the next version will have a linear regulator on it, but I want a proof of concept in the first place using a switching regulator to be connect to the balance plug of the battery.

Wow, slow! Was the white prototype you posted from osh or elsewhere? Anxious to build one of these things, hopefully it comes soon and there's no major issue

Also the balance lead on the fatshark dominator v3 and v2's with the faceplate upgrade connect to the fan, maybe other models too? Granted, I rarely need the fan and could always make a Y-Cable. Small price to pay for diversity

great news! very unexpectedly, yesterday the PCBs arrived!
So I started the build of a first prototype of the "base module" with the new TS5A3359 switch. It seems to have no major issues (just my SMA connector has 2 pins bigger than the other two, unlucky me).



I lost almost 1 hour trying to burn the bootloader with another arduino as ISP before remember to put a 10uf cap between RST and GND of the programmer -.-

btw, hope to find time to fix some sw bugs around and give a try to the "base" module (without the second module for diversity) this evening, then I have to commit all the code, the new PCBs and some explanations of how to build the modules.

sorry for the delays stay tuned

family grows!



ps: the white pcb come from dirtypcb