Re: [Paparazzi-devel] RC signal lost not clean and tiny reset issue

[Jerry Van Baren]

Gareth Roberts wrote:
> miles wrote:
> > On Fri, Apr 3, 2009 at 17:19, Paul Cox <address@hidden 
> > <mailto:address@hidden>> wrote:
> >
> >     2) In manual mode, moving all servos to their extremes
> >     simultaneously (rudder, ailerons, elevator) I can get the Tiny to
> >     reset. I am using a charged 3S pack and the voltage reported in
> >     the GCS looks really good (12.3), but obviously there must be some
> >     dip or spike that's affecting the processor enough to reset it.
> >     What are your suggestions?
> >
> >
> > My group has run into that also.  As a temporary work-around, we power 
> > the receiver and servos from a separate battery (not connected to 
> > Tiny2.11).  Be careful to disconnect the power line between the 
> > Tiny2.11 and receiver -- if not you will *damage* the Tiny2.11, at 
> > least the voltage regulator.
> >
> > For us, the problem was even worse than reseting the Tiny2.11, it put 
> > the Tiny2.11 in a strange mode where the servos would twitch 
> > periodically and not respond to Manual RC commands.
> >
> > Does anyone know of a nice way to determine the total power drawn by 
> > all the servos actuating a full strength simultaneously?  I guess 
> > there should also be a calculation to compute a rough upper bound on 
> > this.
> >
> > Cheers,
> > --
> > miles
> > ------------------------------------------------------------------------
> Couldn't you put a multimeter (on dc current mode) between the tiny and 
> the battery?  Measure the baseline current draw, then actuate one servo 
> to full and so on, measuring each one.  I'd be interested in the result 
> as well, we only use two servos (aeriolons) + motor controller so 
> haven't hit that particular problem. The other thing you could do is 
> feed the PPM signal back to the radio, as someone was discussing on here 
> a few weeks ago, and have that run the servos.
>
> Hope this helps
> --G

Since servos are motor-driven, they will draw a *lot* more current when 
loaded and slewing than when they are standing still.  It is going to be 
hard to measure peak current - that will likely be when the servo 
accelerates from stopped to full speed (max slew rate) while fully 
loaded.  For a give servo part number, I would expect the max current 
draw to be reasonably similar.  It will vary widely between different 
types and models of servos (different motor sizes and efficiencies).

Problems:
1) A multimeter isn't going to measure this since it is going to be a 
short transient current draw.  A digital (storage) scope set up for 
current measurement (voltage drop across a small resistor) theoretically 
will work.
2) It is going to be hard to implement "fully loaded" or even simulating 
a representative load.  Perhaps set up the servo on a bench with a 
spring providing the load?

If you figure out who to ask and get lucky, perhaps your servo supplier 
can tell you what the max current draw is for the servo you are using.

Good luck,
gvb