- Posts: 268
Devo10 Progress(2)
- suvsuv
- Topic Author
- Offline
1) Telemetry: PB is still working on it so I don't want to bring merge conflict to related codes. Once PB gives a ready signal, I will have it ported to devo10.
2) Scanner: it is not a must-have feature so I will have this page ported when I am not busy.
Currently, I am working on the following devo10 specific add-on features:
1) Auto backlight off: the back consumes about 5-10 more mA currents for each step increasing, I will add a logic to have backlight off when there is no key pressed for given period of time , get back on once a key is pressed. This feature will be nice from the perspective of power-saving .
2) Vibration alert: the sound volume of devo10 is really low even though PB has set music playing to its maximum volume. The devo10 provide vibration alert as compensation. However, I don't have any information about vibration device's hardware connection and related PWM frequency, can any hardware guys give me some help on it ?
Here are some power consumption data with current version deviation 10:
1) TX power:
tx power | Deviation 10 (Backlight: 1) | Walkera's factory firmware(Backlight:off) |
150mw | 152mA | n/a |
100mw(20db) | 138mA | 143mA |
30mw(15db) | 124mA | 131mA |
10mw(10db) | 116mA | 117mA |
3mw(5db) | 113mA | 109mA |
1mw(0db) | 111.7mA | 108mA |
300uw(-5db) | 111.0mA | 106mA |
100uw | 111.8mA | n/a |
2) Backlight
Backlight steps | TX power:150mW | TX power: 30mW |
0(Off) | 136mA | 108mA |
1 | 151 | 123 |
2 | 162 | 135 |
3 | 168 | 141 |
4 | 172 | 145 |
5 | 178 | 150 |
6 | 181 | 153 |
7 | 184 | 156 |
8 | 186 | 160 |
9(Max) | 188 | 162 |
3) In terms of Contrast, there is almost no difference from contrast level #0 to #9
Please Log in or Create an account to join the conversation.
- suvsuv
- Topic Author
- Offline
- Posts: 268
Currently it is still on alpha stage, I would suggest play with Mcpx, Minicp or 120d02
Please file bugs in this forum
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Offline
- Posts: 4402
b) I agree the scanner isn't very useful.
c) If you are going to implement power off, do it for both models. Even if you can't test it is fine.
Please Log in or Create an account to join the conversation.
- suvsuv
- Topic Author
- Offline
- Posts: 268
Will do a) and c)PhracturedBlue wrote: a) You can pull the telemetry pages. While the logic itself still needs some work, the pages are not going to change much. Go ahead and get it in.
b) I agree the scanner isn't very useful.
c) If you are going to implement power off, do it for both models. Even if you can't test it is fine.
Please Log in or Create an account to join the conversation.
- rangebound
- Offline
- Posts: 4
I didnt really want to raise this as an issue (I seem to be making a fair few)
It seems the AIL output on the monitor page is opposite to stick movement.
@PB or anyone
Is there a way of closing /deleting resolved issue tickets. I seem to have opened a fair few which have been dealt with already. So it would be good to be able to clean up the open issues list.
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Offline
- Posts: 4402
You can resolve tickets yourself by pressingthe Resolve button in bitbucket
Please Log in or Create an account to join the conversation.
- suvsuv
- Topic Author
- Offline
- Posts: 268
Didn't notice that. Pls assign a bug to me, I will get it fixedrangebound wrote: @SuvSuv :
I didnt really want to raise this as an issue (I seem to be making a fair few)
It seems the AIL output on the monitor page is opposite to stick movement.
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Offline
- Posts: 4402
Your changes do not change the power for the reasons you think they do. There is something going on with the compilation of the main loop. I've found various changes that affect the power, and it has nothing to do with the frequency of update of the 'low priority' things. It is either ~174mA or ~132mA (for backlight=1, power=100mW). There is no middle ground between the 2 . I can cause it to go from 174->132 by either adding or removing code to the main loop or by changing the order code is executed, or by changing compile options.
What I don't yet know is what the underlying cause is. But finding and fixing that is the right solution. Your 'low-priority' movement will likely break if we modify the event loop in any significant manner. Interestingly, there is no difference in the number of iterations-per-second when the Tx is consuming 174mW or 132mW. It is very odd, but I will continue to work on it.
Please Log in or Create an account to join the conversation.
- dc59
- Offline
- Posts: 799
suvsuv wrote: Attachment is the latest FW with corresponding filesystem, you are welcome to play with them. Of coz on you own risk You can also pull codes from PB's or my repo and make dfu by yourself
Currently it is still on alpha stage, I would suggest play with Mcpx, Minicp or 120d02
Please file bugs in this forum
It's really cool~~~~~~~~~~~~~~~~
Thanks!
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Offline
- Posts: 4402
Please Log in or Create an account to join the conversation.
- suvsuv
- Topic Author
- Offline
- Posts: 268
Cool, this is a good signal for us to figure out the root-cause soon.PhracturedBlue wrote: After a couple hours tweaking and rebooting, I've found that the location of the Crc() routine in the dfu matters. if it is moved slightly, it causes the power to spike by ~40mW. This happens even if all calls to Crc() are commented out shuch that it is never called. This, of course, makes no sense, so it is likely indicative of a stack issue somewhere. I have not yet been able to further zero in on the real issue as yet.
Please Log in or Create an account to join the conversation.
- suvsuv
- Topic Author
- Offline
- Posts: 268
Feature notes:
1) All GUI pages, except the scanner, are ported from devo8 to devo10. However, the telemetry feature is not verified with any real RXs
2) Replace 12narrow font new 12 normal font generated by PB, the font looks better in English and Chinese, don't know how it looks in other language
3)Built with optimized setting without debug info, the dfu size is much smaller and should run a little faster
4) Fix most of the known bugs
5) Be able to change TX power on the fly
Major known bugs:
The current consumption rises a lot after manual binding, or interrupting the auto-binding
Please Log in or Create an account to join the conversation.
- Home
- Forum
- General
- General Discussions
- Devo10 Progress(2)