×

Notice

The forum is in read only mode.
  1. Forum
  3. Development
  5. Protocol Development
  7. New Hubsan Upgraded Version on the way

New Hubsan Upgraded Version on the way

More
28 Dec 2013 22:45 #17361 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way

Xermalk wrote: And you have to do it Every time you bind the quad.

What ?
The Hubsan TX doesn't remember the last selected vTX freq ?
More
28 Dec 2013 22:54 - 28 Dec 2013 22:55 #17362 by Xermalk
Replied by Xermalk on topic New Hubsan Upgraded Version on the way
Nope :dry: It randomizes a new one every time in the 5725-5800 range.
Last edit: 28 Dec 2013 22:55 by Xermalk.
More
28 Dec 2013 22:56 - 28 Dec 2013 22:58 #17363 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
But why did they do that ? The TX has some eeprom space because it saves sticks calibration, radio mode and probably a few other things.

Stupid.
Last edit: 28 Dec 2013 22:58 by goebish.
More
28 Dec 2013 23:05 - 28 Dec 2013 23:06 #17364 by Daryoon
Replied by Daryoon on topic New Hubsan Upgraded Version on the way
Looks good goebish.

I used the protocol option to set 5945MHz. And the video displayed on the Hubsan TX just fine.

Then I shut off Devo and X4.

Turn back on Devo and X4. And right after the X4 reconnected to the Devo, the Hubsan TX displayed the video at the last frequency, 5945MHz.


Good job goebish. Very fast work.
Last edit: 28 Dec 2013 23:06 by Daryoon.
More
28 Dec 2013 23:08 #17365 by Daryoon
Replied by Daryoon on topic New Hubsan Upgraded Version on the way
Will it be possible to use the Hubsan's VTX in other quads? Anybody know?
More
28 Dec 2013 23:10 - 28 Dec 2013 23:11 #17366 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
This will require to hack the com between the vTX module and the H107D mcu... I bet there's i2c or serial there.
I'll have a look when I have one :)
Last edit: 28 Dec 2013 23:11 by goebish.
More
28 Dec 2013 23:13 #17367 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
Would you mind sharing HD pictures of the H107D pcb and vtx module ? Or can you link to existing pictures ?
More
28 Dec 2013 23:18 - 28 Dec 2013 23:19 #17369 by Xermalk
Replied by Xermalk on topic New Hubsan Upgraded Version on the way

goebish wrote: Would you mind sharing HD pictures of the H107D pcb and vtx module ? Or can you link to existing pictures ?


www.rcgroups.com/forums/showthread.php?t=1842561

Check the pictures attached, thers high res of everything.

But you likley going to have to fake the frequency selection from the hubsan gyro/rx. Seems its not outputting any video unless the quad is bound.

Edit, interesting the tx is using a biquad antenna, i doubt my cloverleaf will ever compare to that in range.
Last edit: 28 Dec 2013 23:19 by Xermalk.
More
28 Dec 2013 23:19 - 28 Dec 2013 23:21 #17370 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way

Xermalk wrote: But you likley going to have to fake the frequency selection from the hubsan gyro/rx. Seems its not outputting any video unless the quad is bound.


Of course, that's where all the fun is !
Last edit: 28 Dec 2013 23:21 by goebish.
More
28 Dec 2013 23:24 - 28 Dec 2013 23:30 #17371 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
I just created a pull request ... should be in the nightly soon.

Now to the telemetry...

Found anything interesting Midelic ?
Last edit: 28 Dec 2013 23:30 by goebish.
More
29 Dec 2013 06:52 - 29 Dec 2013 06:53 #17382 by midelic
Replied by midelic on topic New Hubsan Upgraded Version on the way
For the moment there are 2 distinct telemetry frames one starts with E0 and the next one with E1...alternates each other.and in between 2 tx frames..for example:

RX:(0xE0),(0xFF),(0xF7),(0x00),(0x020,(0x00),(0x13),(0x00),(0x00),(0xFF),(0xFF),(0x00),(0x00),(0x00),(0x00),(0x17)
TX:(0x20),(0x00),(0x00),(0x00),(0x80),(0x00),(0x80),(0x00),(0x80),(0x0E),(0x19),(0x00),(0x33),(0x00),(0x2A),(0xDC),
TX:(0x20),(0x00),(0x00),(0x00),(0x80),(0x00),(0x80),(0x00),(0x80),(0x0E),(0x19),(0x00),(0x34),(0x00),(0x2B),(0xDA),
RX:(0xE1),(0x00),(0x00),(0x00),(0x00),(0x00),(0x30),(0xFE),(0xBD),(0x00),(0x62),(0x03),(0xDF),(0x2A),(0x00),(0xC6),

SO every 2 TX packet frames there is 1 telemetry frame..only that the second telemetry one is different than previous

Another thing ,when switching channels in the menu ..there are no telemetry packets received.That's it for the moment.
Last edit: 29 Dec 2013 06:53 by midelic.
More
29 Dec 2013 09:45 - 29 Dec 2013 09:46 #17387 by midelic
Replied by midelic on topic New Hubsan Upgraded Version on the way
It looks like every frame sent every 10ms from which
3 ms in TX mode and 7 ms in rx(listening) mode with timeout.Telemetry packets received in this 7 ms (aprox. 4 ms received packets) continue listening for another 3ms polling mode register with timeout).
Last edit: 29 Dec 2013 09:46 by midelic.
More
29 Dec 2013 10:37 - 29 Dec 2013 10:37 #17388 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
I'll try to understand the format.

Daryoon, do you remember what was your battery level when you made the capture ?
Last edit: 29 Dec 2013 10:37 by goebish.
More
29 Dec 2013 10:50 #17389 by midelic
Replied by midelic on topic New Hubsan Upgraded Version on the way
maybe you can see in his video...For some reason I cannot load the video.

www.deviationtx.com/forum/protocol-devel...e-way?start=80#17134
More
29 Dec 2013 10:54 #17391 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
I wouldn't have thought of that, well done Midelic it starts with 4.1 Volts then quickly drops to 4.0
More
29 Dec 2013 13:23 - 29 Dec 2013 13:27 #17394 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
I think I found the voltage value in the telemetry frames :)
    aa bb cc dd ee ff gg hh ii jj kk ll mm nn oo pp
Rx: e0 ff 4c 00 37 00 14 00 00 ff ff 00 00 2a 00 62 
Rx: e1 ff 4c 00 37 00 31 fe bd 00 61 03 de 29 00 46 
Rx: e0 ff 4c 00 37 00 14 ff ff ff ff 00 01 29 00 64 
Rx: e1 ff 4c 00 37 00 31 fe bd 00 61 03 df 29 00 45 
.
.
.
Rx: e1 ff 4a 00 36 00 32 fe b9 00 5f 03 dc 28 00 51 
Rx: e0 ff 4a 00 36 00 13 00 00 ff ff 00 00 28 00 68 
Rx: e1 ff 4a 00 36 00 33 fe ba 00 5f 03 dc 28 00 4f

See nn ? This is the current voltage value, eg:
0x2a = 42 = 4.2V
0x29 = 41 = 4.1V
0x28 = 40 = 4.0V

Now to see how it could be implemented into Deviation...

Thanks for the timings analysis Midelic.
Last edit: 29 Dec 2013 13:27 by goebish.
More
29 Dec 2013 13:28 #17396 by PhracturedBlue
Replied by PhracturedBlue on topic New Hubsan Upgraded Version on the way
look at the Devo code. You should be able to use a similar method for the hubsan telemtery.
specifically the state-machine loop in devo_telemetry_cb() and the parse_telemetry_packet() function.
you only need to set a single value (Telemetry.p.devo.volt[0]) unless there is more interesting stuff being sent back.
More
29 Dec 2013 14:12 - 29 Dec 2013 14:18 #17398 by midelic
Replied by midelic on topic New Hubsan Upgraded Version on the way
After each frame sent, tx is going to rx mode (listening)..poling the mode register till receiving 18(if not receiving that, is going on time out) ..then start reading the fifo register extracting each telemetry packet.
What I want to say the order of telemetry(2+1)frames is coded on hubsan rx not tx. We just simple going in rx after every single tx packet...and poling mode register.
I hope is clear what I want to say.
Last edit: 29 Dec 2013 14:18 by midelic.
More
29 Dec 2013 14:18 - 29 Dec 2013 14:19 #17399 by goebish
Replied by goebish on topic New Hubsan Upgraded Version on the way
I think it is and it shouldn't be very difficult: set the a7105 to rx mode after each frame sent, then on the following frame, before sending a new frame, verify if the mode register is 0x18 and if that's the case read 16 bytes in the fifo register (the 14th one being the voltage value).
Last edit: 29 Dec 2013 14:19 by goebish.
More
29 Dec 2013 14:25 #17400 by midelic
Replied by midelic on topic New Hubsan Upgraded Version on the way
Exactly..it is easy to implement...."peanuts".Hope that "nn" is the real one.
  1. Forum
  3. Development
  5. Protocol Development
  7. New Hubsan Upgraded Version on the way
Time to create page: 1.510 seconds
Powered by Kunena Forum
Re-enable mobile display