HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Dec 18, 2012, 10:22 AM
Registered User
Joined Jan 2012
682 Posts
Quote:
Originally Posted by RW9UAO View Post
look, i set one value for CRC_SEED, correct value for SOP/DATA (based on channel num and manufacturerID). i receive good packet info, not garbage =) if channel is empty, you have not interrupt (IRQ) for a time-out. check RX_IRQ_STATUS_ADR and RX_STATUS_ADR.
Thanks, that is good to know. Though, since I'm not implementing an Rx, it is not so relevant to me.
I got Deviation working with DSMX, so I'm going to start playing around with telemetry next. Thanks again for all your great work.
PhracturedBlue is offline Find More Posts by PhracturedBlue
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 18, 2012, 10:31 AM
Registered User
RW9UAO's Avatar
Russian Federation, Kemerovskaya oblast, Kemerovo
Joined Dec 2012
61 Posts
have you spektrum telemetry block?
RW9UAO is offline Find More Posts by RW9UAO
Reply With Quote
Old Dec 18, 2012, 11:24 AM
Registered User
Joined May 2011
657 Posts
I think I've lost...

Quote:
Originally Posted by David T View Post
Do you have a firm understanding of what the Tx is sending out? If not you need to work that out paying attention to the freq order, timing and whether the CRC is normal or inverted. Because 23 is an odd number, you need 2 repeats of the 23 freq to have the whole pattern.
Regards, David.
Do you ask me?
Because then the answer is no, I do not understand everything, that's why i'm asking questions...

Quote:
Originally Posted by PhracturedBlue View Post
So you can either sit on one channel waiting for data (46 packets at 1 packet per 11msec (or 2 packets per 11msec for >7 channels => 506/253msec to rebind)
I think (but didn't do the math) it ends up taking at least the same time if you scan through the channels (probably even longer since you could miss the channel while changing), but the benefit is that you are much less susceptible to collisions from another radio.
But the point is, that according to:
Quote:
Originally Posted by RW9UAO View Post
next, dsmX lost signal i see on the SPI bus between CYRF and PSoC.
if channel list indexed from 0 to 22, channels scan on index:
unsigned char jump_table = [9, 10, 8, 11, 7, 12, 6, 13, 5, 14, 4, 15, 3, 16, 2, 17, 1, 18, 0, 19, 22, 20, 21];
scan in a both directions. from 0 to 22, and back from 22 to 0 in jump_table. try.
...the Spektrum rx doesn't sit on one channel, but tries to scan the channels up and down by turns.
At least if that was about an actual Spektrum receiver...

So, since I presumed, that the tx continues to send out data in the original channel order, knowing nothing about the receiver's problem, I tried to discover in theory, how much time the rx needs to at least listen to the same frequency, because otherwise it has no chance to recover.

I see, David, that even then the validation may fail, I'm just not yet there...
I think even the time needed to find the same channel seems too long...
FDR_ is offline Find More Posts by FDR_
Last edited by FDR_; Dec 18, 2012 at 11:30 AM. Reason: typo
Reply With Quote
Old Dec 18, 2012, 12:18 PM
Registered User
UK
Joined Aug 2000
1,082 Posts
FDR,
I think it is like in the spreadsheet. I've put two sequences of 23 frequencies to get a complete set of possibilities. Each frequency appears twice.

If you switch the Rx on at time '4' on freq1 with a non-inverted CRC (0), you should get a hit at time '246' but with a CRC error because the Tx inverted the CRC (1). So instead of waiting another 246ms for freq1 to be used again, why not try freq2 without inverting the crc (0)? If that works at time '264' then you know where you are in the sequence. Obviously you need some rules/plan B to try something else if that has errors. You probably need to try different frequencies after every 246ms in case the 'first' freq you try cannot be detected. OK?
Regards, David.
David T is offline Find More Posts by David T
Reply With Quote
Old Dec 18, 2012, 04:11 PM
Registered User
Joined May 2011
657 Posts
Quote:
Originally Posted by David T View Post
FDR,
I think it is like in the spreadsheet. I've put two sequences of 23 frequencies to get a complete set of possibilities. Each frequency appears twice.

If you switch the Rx on at time '4' on freq1 with a non-inverted CRC (0), you should get a hit at time '246' but with a CRC error because the Tx inverted the CRC (1). So instead of waiting another 246ms for freq1 to be used again, why not try freq2 without inverting the crc (0)? If that works at time '264' then you know where you are in the sequence. Obviously you need some rules/plan B to try something else if that has errors. You probably need to try different frequencies after every 246ms in case the 'first' freq you try cannot be detected. OK?
Regards, David.
Thanks!

I've forgotten about the double sending. However I calculated with >7ch, which seems more forgiving, since it uses up the sequence faster...

Actually I don't want to make a DSMX receiver myself, but simply interested in how Spektrum implemented their own, so there is no point to make up an algorythm myself.
These properties are missing from the spec sheets...
FDR_ is offline Find More Posts by FDR_
Reply With Quote
Old Dec 18, 2012, 04:22 PM
Registered User
Joined Jan 2012
682 Posts
Quote:
Originally Posted by RW9UAO View Post
have you spektrum telemetry block?
No. I went to my hobby store but they didn't have it. Ordered one which should be here in 2 days though. Hopefully I can get it going, as I don't have any Spektrum Tx to test with.
PhracturedBlue is offline Find More Posts by PhracturedBlue
Reply With Quote
Old Dec 19, 2012, 09:07 AM
Registered User
RW9UAO's Avatar
Russian Federation, Kemerovskaya oblast, Kemerovo
Joined Dec 2012
61 Posts
PhracturedBlue do you _now_ receive answer from RX when you wait end of BINDing cylce? and show RX type on a screen (dsm2/X 11/22 msec). after you make normal transmit for a few seconds, but do not exit from BIND screen, you need to wait answer from telemetry. it contain a zeroes or normal. add a string to a screen - we have a telemetry answer now =)
note. in dsmX my AR8000 use only 11 bit channel mode, in dsm2 only 10 bit resolution.
RW9UAO is offline Find More Posts by RW9UAO
Reply With Quote
Old Dec 19, 2012, 09:09 AM
Registered User
Joined Dec 2011
292 Posts
From real Spektrum receivers (AR6010 and AR8000) I notice that AR6010 (6 and 7 channels mode) rebind faster then AR8000 (8 and 9 channels mode) (after switch off and switch on Deviation Tx).
vlad_vy is offline Find More Posts by vlad_vy
Reply With Quote
Old Dec 19, 2012, 09:15 AM
Registered User
Joined Jan 2012
682 Posts
Quote:
Originally Posted by RW9UAO View Post
PhracturedBlue do you _now_ receive answer from RX when you wait end of BINDing cylce? and show RX type on a screen (dsm2/X 11/22 msec). after you make normal transmit for a few seconds, but do not exit from BIND screen, you need to wait answer from telemetry. it contain a zeroes or normal. add a string to a screen - we have a telemetry answer now =)
note. in dsmX my AR8000 use only 11 bit channel mode, in dsm2 only 10 bit resolution.
I don't actually listen for the response from the Rx. There are a couple of reasons:
1) I don't need to. Since the user has full control, we don't need to ask the Rx for feedback.
2) Some devo radios don't have a telemetry capable module. They wouldn't be able to receive the feedback signal. Rather than having 2 different modes, I chose the more inclusive one.

At the moment, in Deviation, I use 11bit for DSMX and 10bit fro DSM2. I'll probably add a configuration option for this eventually. the same goes for 22msec/11msec modes. But so far, no complaints about compatibility which is what is most important.

Deviation does have support for telemetry for the Devo protocol already. Adding DSM telemetry support should be as easy as adding a couple of new states to the state machine, and populating the relevant structures. Once I have a telemetry module in hand, I'll give it a shot.
PhracturedBlue is offline Find More Posts by PhracturedBlue
Reply With Quote
Old Dec 19, 2012, 09:24 AM
Registered User
RW9UAO's Avatar
Russian Federation, Kemerovskaya oblast, Kemerovo
Joined Dec 2012
61 Posts
in a future, maybe, if i write own software for T-SIX or DX6i clone i will make a menu with all choices: dsm2/X, 10/11 bit, 11/22 msec. let the user headache.

about Devo module w/o telemetry. have you both? or a photo? the different is big anoth?
RW9UAO is offline Find More Posts by RW9UAO
Reply With Quote
Old Dec 19, 2012, 09:40 AM
Registered User
Joined Jan 2012
682 Posts
Quote:
Originally Posted by RW9UAO View Post
in a future, maybe, if i write own software for T-SIX or DX6i clone i will make a menu with all choices: dsm2/X, 10/11 bit, 11/22 msec. let the user headache.

about Devo module w/o telemetry. have you both? or a photo? the different is big anoth?
I have both, though I've not taken off the shield to see what is inside. I believe they did not include a LNA and/or did not hook the antenna to the Rx pin though. Previous experiments resulted in 0 signal even at very close range. Otherwise the boards are identical. They all have exactly the same markings.
PhracturedBlue is offline Find More Posts by PhracturedBlue
Reply With Quote
Old Dec 19, 2012, 09:40 AM
Registered User
Joined May 2011
657 Posts
Quote:
Originally Posted by RW9UAO View Post
about Devo module w/o telemetry. have you both? or a photo? the different is big anoth?
Those moduls, which you can buy separately are all telemetry capable.
You can distinct them by the small rhombus left to the pins:
FDR_ is offline Find More Posts by FDR_
Reply With Quote
Old Dec 19, 2012, 09:45 AM
Registered User
RW9UAO's Avatar
Russian Federation, Kemerovskaya oblast, Kemerovo
Joined Dec 2012
61 Posts
maybe different is a 2 switch - bypass of power amplifier. need a both photo to compare =)
RW9UAO is offline Find More Posts by RW9UAO
Reply With Quote
Old Dec 19, 2012, 09:49 AM
Registered User
Joined Jan 2012
682 Posts
Here is a pic of the telemetry board with the shield removed:
http://9xforums.com/forum/viewtopic.php?f=5&t=362
I don't know of a pic of the non-telemetry module.
PhracturedBlue is offline Find More Posts by PhracturedBlue
Reply With Quote
Old Dec 19, 2012, 09:52 AM
Registered User
Joined May 2011
657 Posts
RW9UAO, could you tell me what receiver gave you the following channel searching sequence:
Quote:
Originally Posted by RW9UAO View Post
if channel list indexed from 0 to 22, channels scan on index:
unsigned char jump_table = [9, 10, 8, 11, 7, 12, 6, 13, 5, 14, 4, 15, 3, 16, 2, 17, 1, 18, 0, 19, 22, 20, 21];
...and in what circumstances? Did it do on the initial channel synchronization after power on, or just when the signal is lost somehow?
If the latter, how did you make it to loose signal? Did you simply turn off the tx?

It's still bothering me...
FDR_ is offline Find More Posts by FDR_
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Sold DSMX DX-8 and (2) DSMX Sats for Sale laughingstill Aircraft - General - Radio Equipment (FS/W) 3 Jan 20, 2012 12:24 PM
Wanted JR9503 DSMX or JR 11X DSMX and AR9200/AR9210 Receiver w/sats patrick21x Aircraft - General - Radio Equipment (FS/W) 0 Nov 17, 2011 11:22 PM
Discussion WALKERA 2.4 TX hack -can YOU hack it? aaronstomfoolery Radios 2 Oct 29, 2011 05:44 AM
Sold NNIB AR6210 DSMX w/sat + AR600 DSMX GeetarJoe Aircraft - General - Radio Equipment (FS/W) 0 Oct 19, 2011 02:50 AM