SPI capture guide

More
06 Apr 2016 22:17 #45969 by 92zb
Replied by 92zb on topic SPI capture guide
let me try again, photos are only 72 and 67kb


Attachments:

Please Log in or Create an account to join the conversation.

More
06 Apr 2016 22:22 #45970 by 92zb
Replied by 92zb on topic SPI capture guide
One more thing, what are those red dots appering on CLK, what does it mean ?

Here is the data file :-
www.mediafire.com/download/maw5lll0031r4cw/TX-On-Off.csv

Please Log in or Create an account to join the conversation.

More
07 Apr 2016 00:05 #45984 by C0ckpitvue 777
Replied by C0ckpitvue 777 on topic SPI capture guide
Which model is this?

Please Log in or Create an account to join the conversation.

More
07 Apr 2016 02:01 #45988 by victzh
Replied by victzh on topic SPI capture guide
Don't export to CSV, just upload the whole session - .logicdata file. It can be opened and zoomed in while CSV needs to be looked by bare eyes.

Please Log in or Create an account to join the conversation.

More
07 Apr 2016 10:52 #46007 by 92zb
Replied by 92zb on topic SPI capture guide
It is for LIANSHENG LS114 4CH 2.4GHz RTF UFO Drone. I will open a new thread for it when every thing captured properly.

www.ebay.co.uk/itm/LIANSHENG-LS114-4CH-2...?hash=item3cfe669d8f

Please Log in or Create an account to join the conversation.

More
07 Apr 2016 13:53 #46029 by 92zb
Replied by 92zb on topic SPI capture guide
Here is the first capture. There is a problem though. When I connect GND to the analyzer I do not get any signal from CSN, without it error of "Mismatch" and red dots on clock. Any idia ?

www.mediafire.com/download/kbrfkd3q3r2v0fz/TX-ON-BIND-OFF.7z

Please Log in or Create an account to join the conversation.

More
07 Apr 2016 15:13 #46033 by victzh
Replied by victzh on topic SPI capture guide
No, your capture is invalid. While you have CS (it's positive, not negative as it is usually happens, but SPI has many options), your clock is invalid. Clock pulses mark the bits of a byte transferred, so there should be packs of 8 clock pulses at least. Your clock repeats CS edges mainly, so there is high intertalk between the signals. Another option is weakly connected ground - all your signals look weird, that's one sign of bad ground connection. Also, the shorter the wires between your device and logic analyzer the better.

Please Log in or Create an account to join the conversation.

More
08 Apr 2016 20:35 #46128 by 92zb
Replied by 92zb on topic SPI capture guide
This is a new capture. Please have a look and let me know which channel is right. Thanks

www.mediafire.com/download/gz81inmxhtz0d6f/TX-ON-BIND-OFF.7z

Please Log in or Create an account to join the conversation.

More
09 Apr 2016 04:10 #46151 by victzh
Replied by victzh on topic SPI capture guide
Now, we are talking! It's better, your signals are in this order: ENABLE, MISO, MOSI, CLOCK. But one thing bothers me - it's low complexity. While it's well formed - multiple of 8 clock pulses per one CS (ENABLE in Saleae parlance) pulse, some changes on MISO and MOSI which do not contradict other signals - they still are changing very little, not as if it was a real transmission. Can you read markings on a radio chip?

Please Log in or Create an account to join the conversation.

More
09 Apr 2016 15:23 #46181 by 92zb
Replied by 92zb on topic SPI capture guide
The chip is XN297 1452, the configuration is like this :-
( 1) 100mV
( 2) 3.30V
( 3) .5V
( 4) GND
( 5) 2.35V
( 6) 1.35V
(7) 2.35V
[8 ] 1.5V

1) = CLK
3) = CSN
4) = GND
6) = MISO
7) = MOSI

The data extracted @ 24MHz. Please let me know if any thing has to be changed. After that I will go ahead the extract the rest of the Data.

Please Log in or Create an account to join the conversation.

More
11 Apr 2016 23:01 #46332 by victzh
Replied by victzh on topic SPI capture guide
It's a known chip, but so far no real data is present in your capture. As I said - it's low complexity, you have CS and clock right but MOSI and MISO do not look real. They change very little and do not form sensible commands. Check your soldering on MISO and MOSI.

Please Log in or Create an account to join the conversation.

More
12 Apr 2016 14:51 #46365 by 92zb

Please Log in or Create an account to join the conversation.

  • goebish
  • goebish's Avatar
  • Away
  • I Void Warranties
More
12 Apr 2016 14:54 #46366 by goebish
Replied by goebish on topic SPI capture guide
It's perfect :)

Please Log in or Create an account to join the conversation.

  • goebish
  • goebish's Avatar
  • Away
  • I Void Warranties
More
12 Apr 2016 17:48 - 12 Apr 2016 17:51 #46373 by goebish
Replied by goebish on topic SPI capture guide
At first glance it looks a lot like MT99xx / H7 protocol.

I think this proto is not in nightly yet because I've to size optimize xn297 emulation layer for one of its sub-format (it's on my priority todo list), but there's a test build here:
www.deviationtx.com/forum/protocol-devel...in-h7?start=60#42624
Last edit: 12 Apr 2016 17:51 by goebish.

Please Log in or Create an account to join the conversation.

More
12 Apr 2016 19:41 #46379 by 92zb
Replied by 92zb on topic SPI capture guide
If you mean the following test built

deviation-devo7e-v4.0.1-37035f6.zip

I tried but no joy.

Please Log in or Create an account to join the conversation.

  • goebish
  • goebish's Avatar
  • Away
  • I Void Warranties
More
12 Apr 2016 20:08 #46380 by goebish
Replied by goebish on topic SPI capture guide
Ok, probably a protocol option is required, please add your captures to the H7 thread, I'll have a further look :)

Please Log in or Create an account to join the conversation.

  • goebish
  • goebish's Avatar
  • Away
  • I Void Warranties
More
12 Apr 2016 22:56 - 12 Apr 2016 23:14 #46392 by goebish
Replied by goebish on topic SPI capture guide
Ah OK, I've already seen this kind of protocol some time ago in another forum but never implemented it, it's pretty similar to H7 protocol but has something weird in the sense that it has a channel hopping sequence, but it is not used actually because it always switches to the same channel before transmitting the packets :blink:

eg:
25 3D ; set RF channel 3D
27 70 ; clear interrupts
E1 00 ; flush TX fifo
25 2D ; set RF channel 2D
A0 E1 88 70 70 20 1F 00 27 B0 ; fill TX fifo then transmit packet
; wait packet period
25 0E ; set RF channel 0E
27 70 ; clear interrupts
E1 00; flush TX fifo
25 2D; set RF channel 2D (!!!!)
A0 E1 88 70 70 20 1F 00 00 89 ; fill TX fifo then transmit packet
; wait packet period

... and so on ... so in the end it always transmits on channel 2D.
It shouldn't be very hard to implement, I've to check a few stuffs then I'll try to do it.

Have you opened a dedicated thread yet ?
Last edit: 12 Apr 2016 23:14 by goebish.

Please Log in or Create an account to join the conversation.

More
12 Apr 2016 23:33 #46394 by victzh
Replied by victzh on topic SPI capture guide
Interesting strategy!

Please Log in or Create an account to join the conversation.

  • goebish
  • goebish's Avatar
  • Away
  • I Void Warranties
More
12 Apr 2016 23:55 #46396 by goebish
Replied by goebish on topic SPI capture guide
Smells like debug firmware sent to production ;)

Please Log in or Create an account to join the conversation.

More
13 Apr 2016 04:40 #46400 by 92zb
Replied by 92zb on topic SPI capture guide
[quote="goebish" post=46392
Have you opened a dedicated thread yet ?[/quote]

I will open a dedicated trread for it and let you know.

Please Log in or Create an account to join the conversation.

Time to create page: 0.346 seconds
Powered by Kunena Forum