Throttle Control on ardupilot

[Dr Mike Black]

Hi all,
Not sure if im the right place or if this information is somewhere else that i havnt seen yet....but?

OK did all my testing on an easystar, got it all working sweet.....as in Tx on the floor watching my easy drift happily around the sky whilst rolling a cigarette.....height, speed, Alt, all good

transfered the whole shabang over to the Actaul UAV, ok many changes required in teh setup, but the principles are all the same...

However......

In the new UAV,

a) its much faster
b) its much more agile (In a non vague floaty easystar way)

On the ground it throttles up in WP mode..but..heres the big but..in the air, the throttle deos not work AT ALL!

Checked the .h file found the M/S setting, changed from 22 to 50 as this is closer to the cruising speed...

Get the UAV up , switch to WP mode, then nothing, just this protracted glide with no power, manual again ...fine....shes navigating & stabilising, but no throttle control.

Is my board having a spaz...?
have i knackered something in teh code?
Have i calibrated it wrong or something?
Is anyone else having this problem?

I did kinda run this by Chris, but maybe its something others are experiencing?

Any thoughts guys?

regards,

Mike.

[Dr Mike Black]

Also, just came to mind, when checking the telemetry down....

I noticed that the throttle setpoint was not fluctuating in the way that it did on the easy.....


might be nothing ..might be everything.....

ah and ....the other change was no pitot tube on the easy, on the UAV it is quite long??? mean anything?


thoughts?

[Chris Anderson]

This could be a lot of things, but the first thing I'd check is if your throttle channel is reversed on this aircraft/RC setup.

[Dr Mike Black]

Hmmm interesting point......

im using the same chinese Esc`s that we always use, its naturaly reversed in the Tx, same as the easy, so i see no reason why it would be, but def worth a check....

Sadly my 'Other' life is encroaching on doing me checks, should be back in the world tomorrow so can check in then...

Sorry chris i know you have emailed me but i cant get at it until i get home...'emails forbidden here'

LOL really irritating.....was working a treat until i swapped aircraft...


Also cant find teh thread about changing yoru ofset for your airspeed sensor( never did this but might also be worth a check)


cheers.

[Dr Mike Black]

Hi all ,

Sorry its been an eon since i have not been part of this project, sadly my 'OTHER' job has kept me abroad...

Anyway, back in the UK working on this project and the throttle is not playing ball with me at all...

OK, things i have checked...

Airspeed sensor is working....but the ardu thinks that the new UAV`s flying speed is like 20mph or there abouts..

I simply attached a couple of bicycle clips to me, and let it do its thing, the plane slowed to just over stall, i could hear the throttle keeping it flying but only just.....rather scary, given the equipment in the airframe....

So went back to the header file, changed the centre point to 72 instead of 22 for the throttle settings. Re-flew but still wants to fly at a dangeriously slow speed, I did read somewhere that there was an offset to make to teh aispeed sensor but cant find it now..

Also if i test teh one the ground, switch to auto-mode..throttle comes up , i blow at the tube adn it shuts off so its def working, what else can i change in the code to sort this? just need to change the flying speed it thinks is normal!!!

Any help would be great for this......the UAV flies great as long as i control teh throttle....
Many thanks.

[Doug Weibel]

Hi Mike,

One piece of info that may be helpful to you: The current ArduPilot code does not actually use airspeed in meters per second. The units you are working in are actually integer raw pressure data minus the initialization pressure offset. There are equations in the code to convert the pressure reading to airspeed in meters per second, but at present they are all commented out and unused.

You shoud take the speed value you want to use for AIRSPEED_CENTRAL and convert it to a pressure value. To convert meters per second to pressure as used in the program use: pressure = 0.1254 * speed**2, where speed is in meters per second and speed**2 is speed squared.

Of course this may not be your only issue but may be helpful info for debugging your problem.

[Dr Mike Black]

Hi Doug,

ok so if i understand you correctly, 22 (as set in the .h file) it claims that its meters per second...

but...

so what i need to do is take this 22 in the h file and convert as follows?


so 22 mps is something like 50mph....ish(2.23 is the factor and change)....50mph is more than fast enough for my UAV, cruising speed is something around 37-40mph...flat out is about 70mph.

so take 22 sqrd, (=484) x 0.1254 (60.6936) so the preasure is 60.6936....am i reading you right? not quite clear on the the last bit of yoru equation.....

P=0.1254 x speed(sqrd) in mps?

deos this version of the code, need teh offset manually put in to compensate for the differences in the sensors?



this is the last peice of the the puzzle....as if blow genly on teh pitot, it startes to throttle back and if i blow harder if cuts the motor.....

thanks doug,

regards,

Mike.

[Doug Weibel]

Hi Mike,

You have the calculation correct. It is unfortunate that the comments in the code are misleading about it actually being in m/s. It is not.

From your last piece of information it sounds like something is operating backwards. What mode are you testing this in? Waypoint, RTL, Fly by wire?

I think this has been asked before but have you tried changing REVERSE_THROTTLE between 0 and 1?

[Dr Mike Black]

Hi doug,

Yeah cheched the reversed throttle.....very odd though...

reversed it and at startup, got all connected then throttle went to full with bind plug attached(remove before launch) reversed it again, and it armed and remained at closed until i engaged WP mode or RTL...basically i seams as though everything is functioning as it should just the point at which it thinks its flying speed is cool really is not, needs another 15-20 mph on top..

when the uav is at rest, all connected and bind plug removed....all sits there happy, hit wp mode, servos move to desired direction throttle goes to full,...

it might be worth noting that the throttle goes to full even though it set to stop on the TX at this point, when in the air, its set probably at 2/3rds throttle to prove the point....adn it immediatly cuts to almost nothing, until the aircraft slows to a scarily slow speed...

any of this make any sence?


regards,

Mike.

[Dr Mike Black]

Ok tried this again today....

somethings def off,..


raised the centrepoint to 1000 just to see if kept trying to accelerate, but no...


no dif to teh way it performs at all....


Stuck now...

Any of you guys got any ideas?

it thinks its flying at teh right speed, but far too slow....dont know what to change..

Looking forward to any thoughts,.

regards,

Mike.

[Chris Anderson]

Is your throttle channel reversed? Have you tried the Test Throttle setting in config file (9-5) to see if it's working right?

[Guus Eldering]

Hi Mike,

If you connect the FTDI cable to the ardu and install the ground station software on your computer,
Then you can read the airspeed from the ground station when you blow in to the pitot.
This helped me one time when the throttle was not responding the right way,found that the sensor returned a airspeed of -950 m/s

Regards,
Guus

[Dr Mike Black]

Hi Guus,

Ok you pointed out my muppetry (trying to read teh ardustation and blow etc instead of putting the laptop next to the UAV and watching!!!) lol

Ok on the GS teh sensor is not far out....-3 showing...can live with that ....

but the throttle still pulls back at 22m/s on the gauge, i have changes the settings in H file from desert to mud, is there a way to make an offset on this reading to ardu? so that it adds say 10 to it so that its almost right...?

so thanks for that, its proved what i already thought..that no matter what i do to the header, it deosnt change teh throttle set point to react until later in the curve...

any thoughts?

Chris, yes friend, that caused a mightystir, as the output is somthing like 8 hp, and scared the bejusus out of us everything ended up one end of teh workshops...its def going the rigth way, just pulling back far to early!!!!

and many thanks


regards,

Mike.