You have a config issue. Your description of expectation is correct.
Agreed.
Unable to find the config issue (for now), but went to my comfort zone and did a hardware solution. I now have complimentary outputs.
Establish a 2.5V virtual ground with two 10K resistors forming a voltage divider, tie that to pin 5. 10K between pins 6 and 7, 10K between pin 6 and 1
and Pin 6 outputs an inverted analog signal.
Share your parameters and we can probably avoid all that extra stuff.
Sounds like you need to set the reverse checkbox on the servo setup
When using this output type, all the reverse option should do is reverse the polarity of the associated relay output.
Also, the last hardware hack may have unintended consequences, as your output behavior is not correct, so automated modes may not behave as intended.
Here it is
Params_July12.param (15.8 KB)
SERVO1_TRIM,1500
SERVO3_TRIM,1500
And why is RCMAP_PITCH set to channel 7? This doesn’t affect much of anything, but it does render RC7 fairly useless on Rover firmware.
Changing those trims gives me 2.5V at center stick going to 0V at the extremes
Uncheck any reverse options you have selected.
All reverse unchecked
And you get 0 - 2.5 - 0 ???
Even after a full power cycle?
I rebooted and now I have 4.3V center stick and 0.3v it boots up with outputs at 4.3V SERVO1_TRIM,1500
SERVO3_TRIM,1500 …scary!
Can you be more descriptive? I really have no idea what you are trying to convey there.
There is nothing scary. It’s a matter of applying the appropriate configuration.
Also confirm that you have removed the extra logic hardware, or else we are probably spitting in the wind.
no other hardware connected
rebooted again 0.3 - 4.3 - 0.3
SERVO1_TRIM,1500
SERVO3_TRIM,1500
Ok, that is ALMOST expected behavior, and I think ArduPilot is likely outputting proper PWM/duty cycle across the range. I wonder if the hardware you breadboarded up has an inverted polarity somewhere?
Will review tomorrow. Back to the hospital I go to fix stuff.
The good news is that we have an analog min-max-min behavior that should be “moldable” into the signal you actually need.
Did some continuity checks against John’s schematic, all good. Must be a setting somewhere I screwed-up!
Check PWM output with a scope. If it doesn’t match expectation, report back. Otherwise, it’s down to using the digital to analog hardware properly.
Show again the schematic you are using for the DAC.
What is the PWM range of the input (verified as @Yuri_Rage said)?