C-17 Transport

@planger looks good...I'm sure goebish will update the test build for Fer tonight

@Fer One more thing you could test for us: go in to protocol settings for xn297dump and set "Get Intvl" to 20. Then with the c-17 original tx already bound go back in to the xn297 Dump menu and start another scan. After hitting the first channel it will show "Measuring packet intervals...", this will take about half a minute in which the tx looks like it hangs. After that, tell us the result that is displayed. This is just a confirmation for something we already know, just want to make sure this works right. I suggest turning the option off again after that.

@planger which packet does payload xor go? forked goebishs branch on my GitHub

packet[3]=(packet[0]+packet[1]+packet[2]+packet[4])^(rx_tx_addr[0]^rx_tx_addr[1]^rx_tx_addr[2]);

@planger Thanks!

@Fer please try if the attached test build works with your c-17. Protocol GD00X and in protocol options select "Format" "V2"

There is a check somewhere which makes the TX ID valid. So not all IDs work with what I've provided to calculate the xor and channels.

Moeder wrote: @planger Thanks!

@Fer please try if the attached test build works with your c-17. Protocol GD00X and in protocol options select "Format" "V2"

I installed the new firmware you posted, still didn't bind with the V2 option selected. Also, there isn't there the 6th option anymore (for XNdump), but i guess that is not needed anymore.

Let me know if you want me to do something else.

planger wrote: There is a check somewhere which makes the TX ID valid. So not all IDs work with what I've provided to calculate the xor and channels.

Hmmm... I don't know if this makes any difference, but since i have 3 DEVO 10s, i have assigned one TX ID to all of them, so my models bind to all of them, no matter which one i use. If this is the problem, the ID is 1971.

There is another mistake line 200:
if (phase == GD00X_DATA)
{
NRF24L01_WriteReg(NRF24L01_05_RF_CH, hopping_frequency[hopping_frequency_no++]);
hopping_frequency_no &= 3; // 4 RF channels
}
That should be somethink like:
if (phase == GD00X_DATA)
{
NRF24L01_WriteReg(NRF24L01_05_RF_CH, hopping_frequency[hopping_frequency_no]);
if(Model.proto_opts[PROTOOPTS_FORMAT]==FORMAT_V1)
{
hopping_frequency_no++;
hopping_frequency_no &= 3; // 4 RF channels
}
}

Fer wrote:

planger wrote: There is a check somewhere which makes the TX ID valid. So not all IDs work with what I've provided to calculate the xor and channels.

Hmmm... I don't know if this makes any difference, but since i have 3 DEVO 10s, i have assigned one TX ID to all of them, so my models bind to all of them, no matter which one i use. If this is the problem, the ID is 1971.

The TX ID is fixed in the code for now.
It's important to state the RX LED behavior which points where in the code there is an issue.
You see if the bind works when the RX stops blinking fast, after it can do one of the 2:
- If the RX does not see a valid signal it blinks slowly.
- If the RX receives a valid signal the led is solid.

fixed hopping channels and hard coded Fer's fixed id for testing purpose

Moeder wrote: fixed hopping channels and hard coded Fer's fixed id for testing purpose

I installed this, but it doesn't work correcty with my TX. I can't select anything, the ENT button doesn't work the way it should. I can't select model, nor select subprotocol or anything. I don't know why... Any ideas, or is it something i did wrong?

That's odd, there are really only minor changes. Try reflashing

Moeder wrote: That's odd, there are really only minor changes. Try reflashing

I reflashed, had the same result. I installed the previous version posted above (that is c6ba262), and everything went back to normal. It must be something with this file.

We are getting somwhere, but still no success. It tries to bind, but the light after very short period of time, flashes slowly (like planger mentioned).

Ok, as soon as i am back home, i will try with another DEVO, which has an original NRF in it, and will let you know.

Finally SUCCESS!!! The DEVO i used this time, is one with original CYRF plus XL7105 and NRF (CC2500 not installed in this DEVO).

I used the exact same file as with the other DEVO. The other DEVO has a CYRF and a 3in1 Module installed, and has never failed before to bind to anything it should. This is weird. I will have to try to bind it to some NRF small quads i have and see if there is something wrong with the 3in1 in this DEVO.

Also, i will upgrade my main DEVO with 4in1 to this version, and see if that one binds ok. Thank you all