PPM issues (Was: It's time for a 2.1 release?)

quique65
Topic Author
Offline

09 Dec 2012 13:54 #3819 by quique65

It's time for a 2.1 release? was created by quique65

Hello PB,

are the issues found on 2.0 solved? When you plan to publish a new release?. Than you.

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

suvsuv
Offline

09 Dec 2012 16:24 #3823 by suvsuv

Replied by suvsuv on topic It's time for a 2.1 release?

I would suggest to wait several more days until we don't get any feedback about reboot. Also, I will add bug fixes for the following 2 issues
1) The TX still generates telem alarm when minicp's battery is removed
2) Current ppm doesn't actually work if ppm channel is set to 10, and the mid-point of the ppm pulse is 1.4ms instead of 1.5ms.

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

domcars0
Offline

09 Dec 2012 17:13 - 09 Dec 2012 18:00 #3825 by domcars0

Replied by domcars0 on topic It's time for a 2.1 release?

Hi suvsuv,
I remember a old pre 2 release, with which, on my Devo 10, I heard a bip when I long pushed the ENT button to 'quick save' curves or channel mixers... After the 2.0, for me, this bip disapeard? Is'nt it possible to have it again?
Also I would like the feature which is on the Devention devo 10 firmware , a bip
when trims, AUX4 or AUX5 reach 0.
Thanks

[EDIT] Sorry I made a mistake and forgot to copy the good sound.ini file in the media directory

.. Now I've the bip when I save curves/mixer

Devo 10 (+7e) owner. It's mine, please don't touch it with your big fingers

Last edit: 09 Dec 2012 18:00 by domcars0.

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

PhracturedBlue
Offline

09 Dec 2012 20:32 - 09 Dec 2012 20:34 #3832 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

suvsuv wrote: I
2) Current ppm doesn't actually work if ppm channel is set to 10, and the mid-point of the ppm pulse is 1.4ms instead of 1.5ms.

This is on purpose. the extra 400msec is the notch. What is the issue you are having? Note that the period is basically irrelevant for PPM, the only thing being measured is the high pulse-width (which is between .500 and 1.500 with a center at 1msec).
I think the reason 10ch doesn't work is that there needs to be more of a delay between frames. Increasing PERIOD may help, but As I recall, there are some situations where the maximum frame-time is 20msec, so increasing it may break compatibility. We could easily reduce the max channels to 9, but any changes to the PPM will need a lot of testing.

Last edit: 09 Dec 2012 20:34 by PhracturedBlue.

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

suvsuv
Offline

10 Dec 2012 00:44 #3834 by suvsuv

Replied by suvsuv on topic It's time for a 2.1 release?

PhracturedBlue wrote:
suvsuv wrote: I
2) Current ppm doesn't actually work if ppm channel is set to 10, and the mid-point of the ppm pulse is 1.4ms instead of 1.5ms.
This is on purpose. the extra 400msec is the notch. What is the issue you are having? Note that the period is basically irrelevant for PPM, the only thing being measured is the high pulse-width (which is between .500 and 1.500 with a center at 1msec).
I think the reason 10ch doesn't work is that there needs to be more of a delay between frames. Increasing PERIOD may help, but As I recall, there are some situations where the maximum frame-time is 20msec, so increasing it may break compatibility. We could easily reduce the max channels to 9, but any changes to the PPM will need a lot of testing.

You are right, we need to either increase PERIOD to 22000ms or reduce the max-channel to 9. Below is a test result when setting channel to 10, you can see that the sync pulse overlaps with channels' pulse

And from the following partial enlarged image, you will see the max pulse-width is 1.9ms instead of 2.0ms, so the NOTCH might need to set to 500 as a result

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

PhracturedBlue
Offline

10 Dec 2012 02:45 #3836 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

but what problem are you seeing? there is no reason to increase the pulse-width to 2ms that I know of. The current implementation is based off analysis of a real Tx. I think reducing number of channels to 9 is better than increasing the period. Unless you can capture a real Tx and see how it works. For instance, captures from a Devo10 or 12 connected to a simulator would be useful.

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

suvsuv
Offline

10 Dec 2012 04:30 #3838 by suvsuv

Replied by suvsuv on topic It's time for a 2.1 release?

Problem #1, there is no way to squeeze 10 channel of PPM into 20ms standard frequency frame. So the max-channel must be changed to 9.
Problem #2, the neutral position of a ppm pulse should be 1.5ms, and devo10 with factory FW does follow it according to test result. However, the neutral position of a ppm pulse in devo10 with deviation is 1.4ms, which means that users need to adjust subtrim/trim if they want to connect a deviation TX to a PPM-FM module

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

rbe2012
Offline
So much to do, so little time...

10 Dec 2012 06:35 #3841 by rbe2012

Replied by rbe2012 on topic It's time for a 2.1 release?

#1: Would it be possible to do some interlacing like DSM2 does? Sending the important channels in every frame and the rest divided in to frames (e.g. 1-4,5,6,7 and 1-4,8,9,10)? Will a sim recognize them all correct?

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

PhracturedBlue
Offline

10 Dec 2012 06:39 #3842 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

No the PPM protocol sends channels in series there is no numbering as in DSM2. you can make the packet as big as you like, but my fear is that some sims won't like the long period. It probably makes sense to adjust the perido based on the number of channels, so in the case of >9 channels, the period is larger than 20msec. then we could be sure to keep compatibility while allowing more channels where supported.

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

Hexperience
Offline

10 Dec 2012 11:17 - 10 Dec 2012 11:20 #3845 by Hexperience

Replied by Hexperience on topic It's time for a 2.1 release?

I tried the following in ppmout.c

#define MIN_PPM_PW 700
#define MAX_PPM_PW 1700
#define NOTCH      300


#ifdef EMULATOR
#define PERIOD 3000
#else
#define PERIOD 22500
#define BITBANG_PPM
#endif

suvsuv, where does this show the neutral position?

It's working perfectly fine for Phoenix. Will try it with a PPM TX module later today.

There are 10 types of people in this world. Those that understand binary and those that don't.

Last edit: 10 Dec 2012 11:20 by Hexperience.

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

PhracturedBlue
Offline

10 Dec 2012 11:23 #3846 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

you chnaged the throw to be significantly more than a typical tx has I believe.
justchangethe notch to 500, and you'll get the pulse width suvsuv wants.
you also didn't change PERIOD enough given the changes you made.
basically PERIOD should be (I think) the larger of 20000 and (MAX_PPM_PW + NOTCH) * (num_channels + 1)

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

Hexperience
Offline

10 Dec 2012 11:53 #3849 by Hexperience

Replied by Hexperience on topic It's time for a 2.1 release?

I'm just playing around with it based on this:
www.mftech.de/ppm_en.htm

There are 10 types of people in this world. Those that understand binary and those that don't.

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

PhracturedBlue
Offline

10 Dec 2012 14:24 #3851 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

ok. well that is certainly different than what I've seen. It is very possible that not all transmitters use the same definitions.
Some of these parameters may ned to be configurable in the advanced options if we need to be able to plug in other PPM modules.

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

Hexperience
Offline

10 Dec 2012 17:22 #3855 by Hexperience

Replied by Hexperience on topic It's time for a 2.1 release?

Those settings are configurable in "another popular custom firmware" and default to 22.5ms and 300us.

But please let me say, I'm in no way trying to compare deViation to "another popular custom firmware". All the work you've done is fantastic and I thank you for it. I simply started looking into it when the topic came up in this thread.

There are 10 types of people in this world. Those that understand binary and those that don't.

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

FDR
Offline

10 Dec 2012 18:12 - 10 Dec 2012 18:18 #3856 by FDR

Replied by FDR on topic It's time for a 2.1 release?

You don't have to be so mysterious about the "you-know-which" firmware!
It is named in the introduction:

While Deviation is heavily influenced both by the Walkera DEVO8 firmware as well as by the Flysky/Turnighy based ER9X firmware, it has been written completely from scratch to be easily portable and extensible.

Last edit: 10 Dec 2012 18:18 by FDR.

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

Hexperience
Offline

10 Dec 2012 18:18 #3857 by Hexperience

Replied by Hexperience on topic It's time for a 2.1 release?

I know, but I don't want anyone to think I'm disrespecting deViation in anyway.

There are 10 types of people in this world. Those that understand binary and those that don't.

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

PhracturedBlue
Offline

11 Dec 2012 03:15 #3864 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

I checked the Devo10 with stock firmware:
# channels sent: 8
min PW 0.7ms
max PW 1.5ms
center PW: 1.1msec
notch: 0.4msec
frame time: 22msec

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

vlad_vy
Offline

11 Dec 2012 03:43 #3865 by vlad_vy

Replied by vlad_vy on topic It's time for a 2.1 release?

Can you rename this topic to "PPM Issues"???

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

suvsuv
Offline

11 Dec 2012 05:26 #3866 by suvsuv

Replied by suvsuv on topic It's time for a 2.1 release?

PhracturedBlue wrote: I checked the Devo10 with stock firmware:
# channels sent: 8
min PW 0.7ms
max PW 1.5ms
center PW: 1.1msec
notch: 0.4msec
frame time: 22msec

by changing notch to 500, the pulse width becomes normal, its neutral position is just 1.5ms

With such a change, deviation's ppm output is the same as 2801's ppm output

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

PhracturedBlue
Offline

11 Dec 2012 05:30 #3867 by PhracturedBlue

Replied by PhracturedBlue on topic It's time for a 2.1 release?

I implemented a config option in the code.
the default is set to match the devo10's normal output.
I also added support for more than 10 channels, though I'm notsure it will actually work.

Let me know if you have further issues. It works fine in Phoenix

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

Time to create page: 0.129 seconds

Force desktop display

PPM issues (Was: It's time for a 2.1 release?)

Attachments:

Attachments: