Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

1.11.beta: Fixed Wing Flaps PWM output only on ch6, but not on ch7 #14743

Closed
jstrebel opened this issue Apr 23, 2020 · 24 comments
Closed

1.11.beta: Fixed Wing Flaps PWM output only on ch6, but not on ch7 #14743

jstrebel opened this issue Apr 23, 2020 · 24 comments
Assignees
Labels

Comments

@jstrebel
Copy link

jstrebel commented Apr 23, 2020

On my FW with 1.10.1 I have the flaps on channel 6 & 7, works fine.
Not with 1.11.beta1. Only flap on channel 6 is moving. Flap on Channel 7 is dead.
Reverted back to 1.10.1 on the same pixhawk, both Flaps working again.
QGC is on 4.0.6
HW: pixhawk4
Please let me know If you need futher info

Mixer file is on ./microsd/etc/mixers

### AAERTWF.main.mix
#### Skywalker 2000
##  10.4 16:30
###
#### servo 1 Quer rechts
M: 1
O:      10000  10000      0 -10000  10000
S: 0 0  10000  10000      0 -10000  10000
## S: 0 6  10000  10000      0 -10000  10000
#### servo2 Quer links
M: 1
O:      10000 10000    0 -10000  10000
S: 0 0  10000  10000   0 -10000  10000
### S: 0 6 -10000 -10000     0 -10000  10000
### servo 3 Hoehe
M: 1
O:       10000   10000      0 -10000  10000
S: 0 1  -10000  -10000      0 -10000  10000
### servo4 Seite
M: 1
O:      10000  10000      0 -10000  10000
S: 0 2  10000  10000      0 -10000  10000
### servo 5 Motor
M: 1
O:      10000  10000      0 -10000  10000
S: 0 3      0  20000 -10000 -10000  10000
#####
##### zusaetze
### servo 6 flaps links
M: 1
O:      -10000  -10000      0 -10000  10000
S: 0 4  0  20000  -10000 -10000  10000
###
### servo 7 Flaps rechts
M: 1
O:      10000  10000      0 -10000  10000
S: 0 4  0  20000     -10000 -10000  10000
##### s8
M: 1
O:      10000  10000      0 -10000  10000
S: 0 4  10000  10000      0 -10000  10000
#####
M: 1
O:      10000  10000      0 -10000  10000
S: 0 4  10000  10000      0 -10000  10000
@jstrebel jstrebel changed the title Fixed Wing Flaps channel 6 provides PWM output channel 7 not 1.11.beta Fixed Wing Flaps channel 6 provides PWM output channel 7 not Apr 24, 2020
@jstrebel jstrebel changed the title 1.11.beta Fixed Wing Flaps channel 6 provides PWM output channel 7 not 1.11.beta: Fixed Wing Flaps channel 6 provides PWM output channel 7 not Apr 24, 2020
@jstrebel jstrebel changed the title 1.11.beta: Fixed Wing Flaps channel 6 provides PWM output channel 7 not 1.11.beta: Fixed Wing Flaps PWM output only on ch6, but not on ch7 Apr 24, 2020
@sfuhrer
Copy link
Contributor

sfuhrer commented Apr 25, 2020

@jstrebel could you share a log from this? Even from some bench testing would be useful, so just arm the vehicle on the ground and apply the flaps. Thanks!

@jstrebel
Copy link
Author

jstrebel commented Apr 26, 2020

the first is the log 1.10.1
https://review.px4.io/plot_app?log=23e498af-9db1-4794-9464-35c3a2bd529e
this is the log with to 1.11.beta.
https://review.px4.io/plot_app?log=7fde38d8-10eb-4020-a6f4-4c0b7517d28e

Hope this helps. ( I did downgrade to 1.10.1 und upgrade to 1.11.beta several times. NO change with the hardware nor parameter changes

@sfuhrer
Copy link
Contributor

sfuhrer commented Apr 26, 2020

Thanks for the logs. So far I haven't found the anything, beside that they confirm your findings (output 5 and 6 are fixed to 0).

@sfuhrer
Copy link
Contributor

sfuhrer commented Apr 26, 2020

@jstrebel could you maybe run a test with all the other stuff beside the settings for the Flap channels removed in the mixer, so only:

servo 6 flaps links
M: 1
O: -10000 -10000 0 -10000 10000
S: 0 4 0 20000 -10000 -10000 10000

servo 7 Flaps rechts

M: 1
O: 10000 10000 0 -10000 10000
S: 0 4 0 20000 -10000 -10000 10000

If you want to check if that makes the flaps move you would then need to connect them to channel 1 and 2 (in the log it would though be anyway noticeable).
If it works with the reduced mixer, try to, one by one, add the remaining channels again, until you have all you need or it fails again.
Or maybe somebody else has an idea in the meantime what the issue could be.

@LorenzMeier
Copy link
Member

@jstrebel I would test removing all comment lines. Multi-line comments and alike things are not officially supported in mixers.

@jstrebel
Copy link
Author

@LorenzMeier, I deleted the comment lines. No change in behavior.
@sfuhrer If I put only the Flap entries in the mixer file, then as expected the two ailerons (channel 1 and 2) move as flaps in the correct direction.

  • If I put the two entries of the ailerons as the first two entries, then the ailerons work correct. The Flaps move now at channel 3 and 4 (yaw and elevon)
    In case I go back to the complete mixer File with all channels then the Flaps problem is back.
    I do not see any strange character in the mixer file.
    Strange to me is the identical mixer file works correct with 1.10.1 but not in 1.11.beta. Should I upload the mixer file?

@jstrebel
Copy link
Author

@sfuhrer, did you get the required info you requested from me?
Is there a reason that this issue is not marked as a bug?

@sfuhrer sfuhrer added the bug label Apr 28, 2020
@sfuhrer
Copy link
Contributor

sfuhrer commented Apr 28, 2020

@dagar do you have any idea what could have changed on the mixer level lately?

@jstrebel
Copy link
Author

jstrebel commented Apr 28, 2020

I made some further experiments as suggested by sfuhrer.

Put the following mixer file on the sd card then I get:

AAERTWF.main.mix

Skywalker 2000

10.4 16:30

servo 1 Quer rechts

M: 1
O: 10000 10000 0 -10000 10000
S: 0 0 10000 10000 0 -10000 10000

servo2 Quer links

M: 1
O: 10000 10000 0 -10000 10000
S: 0 0 10000 10000 0 -10000 10000

servo 3 Hoehe

M: 1
O: 10000 10000 0 -10000 10000
S: 0 1 -10000 -10000 0 -10000 10000

servo4 Seite

M: 1
O: 10000 10000 0 -10000 10000
S: 0 2 10000 10000 0 -10000 10000

servo 6 flaps links

M: 1
O: -10000 -10000 0 -10000 10000
S: 0 4 0 20000 -10000 -10000 10000

servo 7 Flaps rechts

M: 1
O: 10000 10000 0 -10000 10000
S: 0 4 0 20000 -10000 -10000 10000
Servo 1&2 roll working correct
Servo 3&4 pitch and yaw woring correct
Servo 5&6 Flaps working correct

Test with a changed test mixer file:
If I put duplicate now channel 4 (channel 4&5 now the same mixer setting. Servo 5 moves like servo 4, as expected)
Flap on channel 6 work ok but Flap on channel 7 does not move.

AAERTWF.main.mix

Skywalker 2000

10.4 16:30

servo 1 Quer rechts

M: 1
O: 10000 10000 0 -10000 10000
S: 0 0 10000 10000 0 -10000 10000

servo2 Quer links

M: 1
O: 10000 10000 0 -10000 10000
S: 0 0 10000 10000 0 -10000 10000

servo 3 Hoehe

M: 1
O: 10000 10000 0 -10000 10000
S: 0 1 -10000 -10000 0 -10000 10000

servo4 Seite

M: 1
O: 10000 10000 0 -10000 10000
S: 0 2 10000 10000 0 -10000 10000

servo4 Seite

M: 1
O: 10000 10000 0 -10000 10000
S: 0 2 10000 10000 0 -10000 10000

servo 6 flaps links

M: 1
O: -10000 -10000 0 -10000 10000
S: 0 4 0 20000 -10000 -10000 10000

servo 7 Flaps rechts

M: 1
O: 10000 10000 0 -10000 10000
S: 0 4 0 20000 -10000 -10000 10000

@jstrebel
Copy link
Author

Observation: If I remove the sd card, forcing px4 starts with the built in mixer (Bormatec Maya) channel 4 Yaw and channel 6 in this case wheel ( show the same value if I do pwm info). There is no pwm output value on channel 7
pwm_info_maya

@RomanBapst
Copy link
Contributor

@jstrebel Really weird. I assume that you are using IO for pwm, correct?

@jstrebel
Copy link
Author

Correct I/O PWM OUT on the pixhawk4. Its really weird. Going back to 1.10.1 everything works again.

@RomanBapst
Copy link
Contributor

@jstrebel Can you try to use the pwm test command on that channel?

@jstrebel
Copy link
Author

I did the following test with the standard Bormatec mixer no SD CARD inserted
pwm test -c 6 -p 1000 and 1800
pwm test -c 7 -p 1000 and 1800
Result: Both Flap Servo move

@ryanjAA
Copy link
Contributor

ryanjAA commented May 1, 2020

I’ll load up a fresh plane with 11 beta (and the Albatross standard mixer) and report back the results on the flaps. We have flaps on ch 7 and 8.

@jstrebel
Copy link
Author

jstrebel commented May 2, 2020

Just did another test putting my mixer file back to the SD Card.
pwm test -c 6 -p 1000 (and 1800)
pwm test -c 7 -p 1000 (and 1800)
Result: Both Flap Servos move
But stil no change via RC: only Flap on Ch 6 move

@ryanjAA
Copy link
Contributor

ryanjAA commented May 2, 2020

I loaded up 11 beta 1 and tested as well. We have 2 channel flap output, main 7 and 8 but triggered by RC channel 9 when done manually. It also doesn't work in our case. Looking at it a little closer, we have no output on anything from output 6 and up (servo_output_raw) shows this as well.

Taking from above, running pwm test -c 7 -p 1000 and we have movement but otherwise output shows 0 us.

This was done on 11 beta 1 with the Albatross aircraft selected and mixer that goes with it, nothing on the SD card or mixer to pull in, all standard and pulled from QGC.

@mcsauder
Copy link
Contributor

mcsauder commented May 4, 2020

Hi @jstrebel and @ryanjAA. If you add the following to your airframe config files does it change the behavior?
set PWM_OUT 12345678

@jstrebel
Copy link
Author

jstrebel commented May 4, 2020

@mcsauder : No change

@mcsauder
Copy link
Contributor

mcsauder commented May 5, 2020

@jstrebel can confirm with his tests, but this change appears to have made this issue show up.

Hi @dagar , you might be able to explain what the intention of this change was and if there is a better way to address the current behavior observed in this issue.

(@jstrebel , in the meantime, if you want, you can explicitly place set PWM_OUT fmu in your aircraft config file as well and that should take care of things without modifying the code.)

Updated Current master behaves differently than v1.11.0-beta1 and what I wrote above is likely incorrect.

(Edited, sorry @bkueng , I was looking at a different commit when I tagged you.)

@dagar dagar self-assigned this May 5, 2020
@dagar
Copy link
Member

dagar commented May 5, 2020

Hi @dagar , you might be able to explain what the intention of this change was and if there is a better way to address the current behavior observed in this issue.

I don't think I'm following the issue here. That change shouldn't have been more than a rename, is that not the case?

I'll take this issue and try to reproduce.

@dagar dagar added this to the Release v1.11.0 milestone May 5, 2020
@jstrebel
Copy link
Author

jstrebel commented May 5, 2020

@dagar, happy to help and do more beta testing.

@mcsauder
Copy link
Contributor

mcsauder commented May 6, 2020

@dagar , this issue appears fixed in current master but not in v1.11.0-beta1 as reported back by @jstrebel .... I'll poke around on this a bit more tomorrow.

@jstrebel
Copy link
Author

Just tested it on Beta2. Its ok now

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

7 participants