Protocol Stacks

You may have noticed that I added documentation for the DSM2 protocol. Hammer22 provided all of the SPI data, and most of the protocol info I needed, but the entire spec wasn't available anywhere to my knowledge.

As I have a cheap HK Orange DSM2 Rx, I'll probably code up an interface to it and see if I can get that working.

After that I'm not sure if I'll go back to working on the interface or if I'll start on the J6 protocol.

Cool! DSM2 is essential.

What's next is up to you. I can help you making the interface, at least with ideas and testing, but with the protocols you are alone...

I'm out on holiday next week, I don't know if I will have web access at all...

A quick note that I now have DSM2 binding too
I don't seem to have the channel data correct yet, since I can only affect one channel, and there is a huge amount of noise on that channel.
Also, the code is incompatible with the DEVO protocol (I can't currently compile code that would allow selecting the protocol in software). This is because the DSM traces I have initialize in MSB 1st mode, whereas the DEVO uses LSB 1st mode. When I tried to use LSB-1st mode, I can't get theDSM2 receiver to bind (but it should work, as the CYRF6936 supports inputs in both syntaxes). I assume I can get that fixed with some more experimentation though.

Anyhow, even though it still needs work, it is another big milestone.

Well, I found most of the problems with the DSM2 protocol and got that fixed up and now I've moved on to the Nine-Eagles J6Pro protocol. This one is important to me because I have an SP-100 an SP-125, and I hate the J6Pro Tx. I was able to figure out 90% of it in only a day, so I'll probably code it up next. At that point I'm out of 6936 protocols to decode. I'm not going to move onto the Flysky or FrSky right now as that requires hooking up new hardware, so once the J6Pro is done, I'll likely move back to working on the firmware until I can get my hands on a DSMX receiver and 2801 Tx/Rx (or equivalent)

Do you think it is important to keep this topic in the "restricted" forum, or should I move it to the general discussions, so simple users could see?

FDR wrote: Do you think it is important to keep this topic in the "restricted" forum, or should I move it to the general discussions, so simple users could see?

I need to go through the posts 1st. The main parts that I want to keep restricted are the parts relating to reverse-engineering. I don't recall exactly what is in this thread at the moment.

BTW: Were rch4x0r's missing files, that I sent you, useful?

Yes, although not enough to actually implement the protocol. I also see several difference between the 4ch and 8ch protocols, which is annoying, but not too surprising.

No big surprise there.
I guess they changed the protocol every time when it was limiting. It was some bad design I guess...

The 2801 knows all the three protocols, so renatopub can supply you with the missing data.

I hope you will succeed, because I have a CB100 that I would like to control with my DEVO 8.
The old Walkera users will raise you to heaven...

Well, It is already largely done. If I can figure out one additional byte, I can probably bind and fly an rx2801 (though I don't have such a receiver, so I can't test). I need more data to do the 6 and 4ch though (as well as to fully flush out the 8ch).

And now the last byte is done. So I think I have enough info to fly an rx2801 now. I'm missing a few things like failsafes, but I think I could bind and fly with what I have
I'll need to start working on the code itself I guess.

Cool!
Actually I have a RX2801 too, but have nothing to control with. All my servos are with smaller plugs...

MatCat wrote: Do we know what protocol Esky uses?

No, but I'd bet the nrf 24l01 chip. That has been the most common choice for the 4channel stuff. If someone has one and wants to rip it apart, and take pictures, we could know for sure.

FYI, I moved this topic to the general section

wuselfuzz wrote: Heya. Just wanted to say hi, but not without any info at all.

I could do testing with quite a bunch of Walkera stuff. I have a TX2402 and TX2801Pro as well as a V100 (240x compatible), rx2609 (2801 compatible, from a 4G6), a bunch of rx2801, a couple of rx2435 (UFLY receiver) and an rx2610s.

I currently don't have any SPI sniffing device, though. A quick fix might be an AVR running at 3.3V dumping SPI data to a UART, though.

Glad to have you aboard.I have an rx2801Pro, as well as a RX2609 that I plan to use for testing. I do not have anything using the TX2401 protocol though, and it is somewhat different than the other 2.
My hope is that once I debug the 2801 and 2601 protocols, that the 2401 will be easy, but one difference is that I've mixed the trims into the Channel data already, and the 2401 broadcasts trim separately from channel data. I will be interested to see if this causes issues or not.

Over the past 2 weeks, I've been solely focused on GUI capabilities, but once I release the emulator, I'll get back to the Tx and debugging the protocols. Once I have the 2801 working, I'll definitely need someone to test the 2401 stuff. I am somewhat concerned about the 6ch too, as I'm not really sure which connections on the RX2609 are supposed to be used.

I'll also just generally need brave souls to test out the firmware. I'm a much better coder than pilot (which is not to say my coding skills are great), and I spent most of my free time this summer coding up Deviation as opposed to improving my skills. It means that I amnot prepared to test things like the 6ch CCPM controls, except with the bird on the ground. I also don't have a 6ch Walkera heli, so that is also a limitation.

I have a CB100, which is 2401 compatible, so I can test it too.
But I have to figure out how to compile for the tx and make a dfu from that...
I hasn't implemented the encoding in my dfu tool yet.