Re: [Paparazzi-devel] Building hardware / Paparazzi Hardware CVS contents

[antoine drouin]

Hi

> My experience has been with OrCAD, but will be getting familiar with Eagle
> for this since that appears to be the group's choice of tool. For future
> changes I'm interested in using geda as those tools seem to be quite capable
> and completely open source.

I'm ready to give another try at GEDA - II had tried it some years ago
but didn't find it usable

> For now I'll work to get up to speed with Eagle
> and then will determine whether I can handle finishing the tiny4l_18_1_99
> design or if I should just fab the standard v2.11 design to get going.
>
> Is there any sort of bug tracker setup for paparazzi? Seems like it might be
> useful for keeping up with sw/hw issues.

there's one at savannah, but not really used

>
> On the IMU, panelizing the various little PCBs should let you have the board
> house only build one PCB (which you can then cut yourself) and reduce build
> cost. Maybe still the cost isn't much less than the sparkfun and other
> solutions though, I haven't looked.

I've setup at panelization solution in the cvs using gerbmerge - go to
the hw/panel directory and type "make _panel"

> One thing that isn't apparent is how many folks are using thermopiles vs.
> IMU, if both are supported well in sw, and the
> advantages/disadvantages/tradeoffs (beyond the obvious indoor/outdoor
> limitations). The thermopiles aren't cheap (neither are the gyros) so
> curious what which provides what in terms of overall price/performance. I'd
> be interested to hear the accounts of users on that. Also, does it ever make
> sense to have thermo and IMU on an aircraft or is that overkill?

>From what I know everyone flying fixed wing uses thermopiles. IMUs are
complicated - for example you'll need a turntable to calibrate your
gyros. Aso the only difficulty about using inertial is not only the
hardware, but the filtering software. The traditional AHRS won't work
very well on a fixed wing

> The booz project seems to have a functioning IMU, the design is in SVN and
> the videos do seem to show it functioning (on a quadrotor). Adding a few
> things back to it (like servo driver) might turn it into Tiny with IMU (and
> quad rotor support -- or dual/tilt rotor a la Osprey?).

the booz project is really part of the paparazzi project. The booz
control board already has the capability
 to drive two servos directly, or a 4017 if you want more

Poine