Protocol Stacks

wuselfuzz
Offline

09 Aug 2012 18:52 #1023 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

PhracturedBlue wrote: Right...4-channel doesn't have pitch control... . But the 4channel does have heading hold? how do you control it? That would need to be a 5th channel (even if it is not proportional) right?

There's pots on the rx for gyro gain.

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

wuselfuzz
Offline

09 Aug 2012 18:54 #1024 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

PhracturedBlue wrote:
wuselfuzz wrote:
Throttle 50%, Throttle trim 100%:
df 00 2f ff a6 5f 40 c8 ff 93 86 62 a0 62 34 38 

Throttle 50%, Throttle trim 0%:
df 00 8f ff a6 00 40 c8 ff 91 86 62 40 02 49 6b
Can you check these again?
The 1st looks like 100% throttle, 0% trim
The 2nd looks like 25% throttle, 0% trim

Just to be clear:

0% = stick all the way down, 100% = stick all the way up, 50% = stick centered. Same for trims.

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

wuselfuzz
Offline

09 Aug 2012 19:19 - 09 Aug 2012 20:50 #1025 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

Data for centered stick, Trim varied:

removed to avoid confusion

Last edit: 09 Aug 2012 20:50 by wuselfuzz.

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

wuselfuzz
Offline

09 Aug 2012 19:25 - 09 Aug 2012 20:50 #1026 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

Elevator data, Stick Centered, Trim varied

removed to avoid confusion

Last edit: 09 Aug 2012 20:50 by wuselfuzz.

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

wuselfuzz
Offline

09 Aug 2012 19:30 - 09 Aug 2012 19:45 #1027 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

d'oh.

Sniffer set to LSB first, CYRF datasheet specifies MSB first. -.-

/edit: Here's some elevator trim data with MSB first:

 0% fb 02 f1 ff 45 00 14 13 ff 29 c7 18 15 42 eb b7 
 1% fb 0c f1 ff 45 00 14 13 ff 29 c7 18 16 42 e6 c2
48% fb ec f4 ff 55 00 14 13 ff 29 c7 18 12 4c 19 bb
49% fb f6 f4 ff 55 00 14 13 ff 29 c7 18 11 4c 00 c4
50% fb 00 f4 ff 65 00 00 13 ff 29 c7 18 15 42 d8 c4
51% fb 0a f4 ff 65 00 00 13 ff 29 c7 18 16 42 d1 cf
52% fb 0a f4 ff 65 00 14 13 ff 29 c7 18 15 42 c6 e2
53% fb 1e f4 ff 65 00 14 13 ff 29 c7 18 19 46 da fe
54% fb 28 f4 ff 65 00 14 13 ff 29 c7 18 1a 46 ef 09
55% fb 32 f4 ff 65 00 14 13 ff 29 c7 18 1a 46 f5 13

Packet counters make sense now, too.

Last edit: 09 Aug 2012 19:45 by wuselfuzz.

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

wuselfuzz
Offline

09 Aug 2012 19:51 - 09 Aug 2012 19:55 #1028 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

Throttle again:

Stick/Trim

 0%/ 0% fb 0c f1 ff 45 00 02 13 ff 09 c7 18 11 4c d9 95 
 0%/100%fb 0c f1 ff 45 00 fe 13 ff 39 c7 18 1a 46 14 c4 

 0%/49% fb 0c f1 ff 45 00 f6 13 ff 19 c1 1a 09 36 5b 77 
 0%/50% fb 0c f1 ff 45 00 00 13 ff 29 c7 18 16 42 f2 ae 
 0%/51% fb 0c f1 ff 45 00 0a 13 ff 29 c1 1a 0a 36 94 9c 

50%/49% fb 0c f1 ff 45 01 f6 13 ff 99 c1 1a 0a 36 d9 f9 
50%/50% fb 0c f4 ff 45 ff 00 13 ff 69 c7 18 11 4c 41 f5 
50%/51% fb 0c f4 ff 45 03 0a 13 ff a9 c1 1a 05 32 19 19 


 0%/50% fb 0c f1 ff 45 00 00 13 ff 29 c7 18 16 42 f2 ae 
25%/50% fb 0c f1 ff 45 fa 00 13 ff 29 c7 18 12 4c 02 ae 
49%/50% fb 0c f1 ff 45 f6 00 13 ff 69 c7 18 11 4c 4d e9 
50%/50% fb 0c f4 ff 45 ff 00 13 ff 69 c7 18 11 4c 41 f5 
51%/50% fb 0c f4 ff 45 08 00 13 ff a9 c7 18 11 4c 76 3e 
75%/50% fb 0c f4 ff 45 05 00 13 ff e9 c7 18 15 42 31 75 
100%/50%fb 0c f4 ff 45 fa 00 13 ff e9 c7 18 15 42 ce 6a

Last edit: 09 Aug 2012 19:55 by wuselfuzz.

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

wuselfuzz
Offline

09 Aug 2012 20:25 #1029 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

This is what happens when I bind the devo8 to the rx2435:

f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6

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

PhracturedBlue
Offline

09 Aug 2012 20:39 #1030 by PhracturedBlue

Replied by PhracturedBlue on topic Protocol Stacks

wuselfuzz wrote: This is what happens when I bind the devo8 to the rx2435:
f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 

Where do you see that?
I thought binding was working ok a couple of days ago, and I didn't change anything releated to binding.

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

wuselfuzz
Offline

09 Aug 2012 20:41 #1031 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

PhracturedBlue wrote:
wuselfuzz wrote: This is what happens when I bind the devo8 to the rx2435:
f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 f6 
Where do you see that?
I thought binding was working ok a couple of days ago, and I didn't change anything releated to binding.

This happened after the rx bound, but now it's not reproducible...

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

wuselfuzz
Offline

09 Aug 2012 21:01 - 09 Aug 2012 21:43 #1032 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

A few (proper) bind sequences from the tx2402 and the tx2801pro and their corresponding data packets:

2402

a5 23 2c 24 d0 28   00 13 ff 29   74 e6 00   2c 0d d1
fb 0c f4 ff 45 00   00 13 ff 29   64 e6 05   24 df ed

a5 23 42 46 d0 3a   00 13 ff 29   dc 82 00   42 b1 35
fb 0c f4 ff 45 00   00 13 ff 29   cc 82 05   46 71 13

a5 23 18 0e d0 12   00 13 ff 29   36 04 24   0e 8f 77
fb 0c f4 ff 45 00   00 13 ff 29   26 04 20   18 66 dc

a5 23 4e 48 d0 20   00 13 ff 29   b4 4a 08   20 63 af
fb 0c f4 ff 45 00   00 13 ff 29   a4 4a 00   4e dc b6

a5 23 48 4c d0 40   00 13 ff 29   d4 02 27   4c 6a f0
fb 0c f4 ff 45 00   00 13 ff 29   c4 02 26   4c d0 b2

2801 Pro

a5 23 48 22 d0 29   ff 00 00 32   15 30 93   f8 96 2c
df 00 0f 00 6a 70   00 f6 fc 25   e8 30 95   f8 50 84

a5 23 31 1c d0 21   ff 00 00 32   18 03 06   f8 72 50
df 00 0f 00 6a 70   00 f6 fc 25   e8 03 01   f8 f7 c3

a5 23 2e 32 d0 4f   ff 00 00 32   18 87 91   f8 3e a0
e0 00 10 00 6a 70   00 f6 fc 25   e8 87 9a   f8 c8 e2

a5 23 24 45 d0 28   ff 00 00 32   1b 9a 66   f8 cd 6d
e0 00 0f 00 6a 70   00 f6 fc 25   e8 9a 65   f8 35 bf

Last edit: 09 Aug 2012 21:43 by wuselfuzz.

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

wuselfuzz
Offline

09 Aug 2012 22:03 - 09 Aug 2012 22:13 #1033 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

Bit 0 in Byte 10 (count from 0) seems to be a "bind packet"-bit in 2401 mode and TxID seems to be only 12 bit.

Finally got a servo moving with the change in my repo, but it probably broke 2601/2801 stuff (quick hack for testing, need to go to bed).

/edit: I just hovered my V100D01 for a short time using the Devo 8.

Last edit: 09 Aug 2012 22:13 by wuselfuzz.

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

PhracturedBlue
Offline

10 Aug 2012 01:47 #1034 by PhracturedBlue

Replied by PhracturedBlue on topic Protocol Stacks

wuselfuzz wrote: Bit 0 in Byte 10 (count from 0) seems to be a "bind packet"-bit in 2401 mode and TxID seems to be only 12 bit.

Finally got a servo moving with the change in my repo, but it probably broke 2601/2801 stuff (quick hack for testing, need to go to bed).

/edit: I just hovered my V100D01 for a short time using the Devo 8.

Ok, I implemented something that should work with the 2401, and 2801. Let me know how it goes.
I will test the 2601 later and see if I can get it working reliably too.

I also updated the servo travel such that it should give full range to the 2401.
Note that as I've said before, I'm sending the trim/channel-mixed data in the 'channel' fields as opposed to sending the trim data separately. Your data indicates that the trims do not affect the maximum servo travel, so I remain unconvinced that there is any benefit to letting the RX mix the trims.
FDR is concerned that it may affect some flybarless models, but someone will have to show me an actual issue before I agree to change it. Did Walkera even make a flybarless FP compatible with the 2402?

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

PhracturedBlue
Offline

10 Aug 2012 04:33 #1035 by PhracturedBlue

Replied by PhracturedBlue on topic Protocol Stacks

Ok, I've played with the 2801 and 2601 settings and I think I have 2601 and 2801 coded properly now. Hopefully I didn't break the 2401 in the process.

The 2601 channel behavior is odd. servo-throw is different for each channel for the same input, and the throttle channel is non-linear. There is also a Gyro channel, but it doesn't seem to be exported to a servo control wen bound to the Rx2801.

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

FDR
Offline

10 Aug 2012 04:35 #1036 by FDR

Replied by FDR on topic Protocol Stacks

PhracturedBlue wrote:
wuselfuzz wrote: /edit: I just hovered my V100D01 for a short time using the Devo 8.
Did Walkera even make a flybarless FP compatible with the 2402?

Yep, for example the aforementioned V100D01, but there were quite a few:
V100D01, V120D01, V200D01, V200DQ01, V200D02, V370D01, 4F180, 4F200LM, V100D06...

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

wuselfuzz
Offline

10 Aug 2012 06:12 - 10 Aug 2012 06:14 #1037 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

PhracturedBlue wrote: Note that as I've said before, I'm sending the trim/channel-mixed data in the 'channel' fields as opposed to sending the trim data separately. Your data indicates that the trims do not affect the maximum servo travel, so I remain unconvinced that there is any benefit to letting the RX mix the trims.

Actually I wrote earlier that I grounded the ADC input on the RX2435 from the "ELE-AIL-EXT" pot. This is a scale factor for servo throws on the rx. I noticed that there's a dead zone at maximum stick deflections, even with the original TX. There might be oversaturation in the rx going on - stick and trims are mixed, the result is multiplied by a too large "EXT" factor and subsequentially clipped to a value corresponding to 910us/1910us before the PWM output is generated.

I have another unmodified RX2435 that I could check later with the EXT pot set to a more sane value.

/edit: The V100D01 was the best FP micro heli I got from Walkera yet, and yes, it's flybarless.

Last edit: 10 Aug 2012 06:14 by wuselfuzz.

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

wuselfuzz
Offline

10 Aug 2012 07:13 #1038 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

PhracturedBlue wrote: There is also a Gyro channel, but it doesn't seem to be exported to a servo control wen bound to the Rx2801.

It's news to me that a TX260x can bind to an RX2801.

The TX260x is weird anyways. On the original TX2601, it requires flipping DIP switches to set some values. On the 2801 Pro, I ran into some strange behaviour when trying to configure the RX2610S.

I suspect that an RX in 2601 mode can enter different states for programming certain values that are stored on the RX.

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

FDR
Offline

10 Aug 2012 09:38 #1040 by FDR

Replied by FDR on topic Protocol Stacks

PhracturedBlue wrote:
FDR wrote: I would like to say that "it is not even close to the real 100%"...
With the fix I checked in an hour ago that is still true?

It is better now, but the range is still too narrow.
The CB100 could barely lift off to about 4 inches.
There was a strong left drift too, I must use aileron trim...

But it binds, and the directions are all right.
However we should indicate somehow, when the automatic binding period ends...

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

wuselfuzz
Offline

10 Aug 2012 10:58 #1043 by wuselfuzz

Replied by wuselfuzz on topic Protocol Stacks

FDR wrote: It is better now, but the range is still too narrow.
The CB100 could barely lift off to about 4 inches.
There was a strong left drift too, I must use aileron trim...

Hmm. With the latest code, I think I get full servo throws. Hovering the V100 was quite twitchy, too. Are you sure that limiting doesn't happen in the mixer, does the output monitor show full ranges?

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

FDR
Offline

10 Aug 2012 11:04 - 10 Aug 2012 13:40 #1044 by FDR

Replied by FDR on topic Protocol Stacks

wuselfuzz wrote:
FDR wrote: It is better now, but the range is still too narrow.
The CB100 could barely lift off to about 4 inches.
There was a strong left drift too, I must use aileron trim...

Hmm. With the latest code, I think I get full servo throws. Hovering the V100 was quite twitchy, too. Are you sure that limiting doesn't happen in the mixer, does the output monitor show full ranges?

Yep, I have a simple 1-to-1 for all the channels without limiting them with scale, min/max or anything else...
The only thing I've done is a throttle hold with the safety value feature...

It is with the latest build of today morning: 6c984eceee15

EDIT: Well, I have to check if the battery was fresh enough...

Last edit: 10 Aug 2012 13:40 by FDR.

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

FDR
Offline

12 Aug 2012 20:20 #1062 by FDR

Replied by FDR on topic Protocol Stacks

Looks like it was a flat battery...

I've flown my CB100 and Genius CP with the new fw!
BTW looks like the gyro gain is sent on channel 7. I thought it is the 5th, but the rx led doesn't light, which is the case of too low gyro gain, and it flew horrible that way...

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

Time to create page: 0.075 seconds

Force desktop display