First sucessfull flight

Yesterday I went for a second attempt to fly. The weather was still bad, but we got a little break and my testing pilot has a lot of Bixler experience and felt confident.

This time everything worked incredibly well. The plane flew like on rails, in AUTO and in MANUAL. It managed to fly some very hard turns very safely. I have great confidence. And it does not look like any tuning will be required.
Except for cruise throttle. The flight duration was very short. That was probably mainly due to a unnecessarily high cruise throttle speed.

I flew here
http://binged.it/NCYHBS

Resumee of first flight

First of all, my initial assumption that the elevator servo was inverted was wrong.

I found several issues though.

• Elevator neutral in the Servo section was biased towards diving.
• Gyro sensitivity was set to 1. Which is absolutely wrong.

Stabilisation is mainly a Gyro thing. And my gyros did not work thanks to my idiotic default value. It took more than five seconds for the IMU to 'sense' new headings.

First flight

Yesterday we had the first flight of the Bixler airframe (flies very nice) and the first flight in AUTO (not so nice).

First I noticed that I had the aileron serovs reversed. Quite idiotic that I never realized that earlier.
Then I noticed the same for the elevator. But I never checked if that was the radio or the autopilot. I just reversed them in the <servos> section.

My pilot approached the test area nice and stable but a little bit low. Too low as it turned out.
Soon as it was in Launch mode it started diving full throttle into the ground. The pilot had no chance to catch that.
In retrospect it's clear that we should have expected that and should have done that at high altitude.

I still have to investigate the cause. But I think the elevator channel was reversed in AUTO.

Damage report:

• Belly landing at high speed.
• Cockpit ripped into pieces.
• Motor and motor mount ripped off.
• GPS flew off and the helical antenna broke off.
• Camera mount snapped.
• Lost tail wing
• Front foam severely bent and snapped.

All in all it's fixable. She will be flying again in a few days and a few pounds of hot glue later...

Recent Progress - Cockpit wiring [With pictures]

I finished my cockpit and wiring based on this neat HobbyKing product http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=22731

I'm quite pleased with the result. I just had to cut a window in the back firewall and some mounting holes for the Lisa board to make it UAV ready.

I also added a super cheap mini keychain camera which currently sells for £10 on amazon.
No live FPV with that. But I should get some nice recordings.

Recent Progress - Bixler Airframe tuning

A post from me asking for some kind of clean starting point for my aircraft file turned into some nice effort for a Bixler based standard template for others to use.

It still needs tuning. But this is great news for beginners.
The Bixler is cheap to buy and versatile.

This is the location of this file https://github.com/paparazzi/paparazzi/blob/v3.9/conf/airframes/examples/bixler_lisa_m_2.xml
Thanks Felix.

Recent Progress - Possibly stable XBee Telemetry

I still have great issues connecting XBee.
But it seems that if I reset the ground station radio shortly before I power up the airborne ratio I have a chance of getting a sync.
If it holds for more than a few seconds it will hold forever.

Still not happy with it. But at least there seems to be a chance that the radio itself is not broken.

Recent Progress - Stable Spektrum satellite receiver

My Lisa/m always crashed when I tried to bind my radio.
I recently found the solution. The interrupt handler for the URART which the second Satellite is connected to needs to be overwritten.

<define name="RADIO_CONTROL_SPEKTRUM_SECONDARY_PORT" value="UART5"/>

Felix Reuss came up with a way to make this better
http://paparazzi.enac.fr/wiki/Subsystem/radio_control#Spektrum
Now only this is needed.

<configure name="USE_SECONDARY_SPEKTRUM_RECEIVER" value="1"/>

Recent Progress - Spektrum DX5e mode switch

My existing RC transmitter is a Spektrum DX5e which has no three way switch.
Paparazzi expects one channel with such a switch for mode selection MANUAL/AUTO1/AUTO2

I had to come up with a solution, since I don't have the means and the nerves right now to buy a pimped up Turnigy 9x or other flexible TX.

My TX actually has 6 channels, the bind switch is another channel.
This witch returns to it's original position, I will have to hack that to be bi-stable. But I can live with that for now.
I thought I should be able to do something with that.
Someone on the Paparazzi Mailing List gave me the right hint and I came up with this.
 https://github.com/tbaumann/paparazzi/commit/6398ff7833aadf94ad6a9b6234f947d43cef503f

 One switch selects MANUAL and AUTO. The other toggles between the two AUTO modes.

I'm quite pleased with his.

Recent progress - RC Flying

I improved my RC flying foo significantly together with friends.

We found some nice flying areas around Guildford.
Hardly perfect locations, but they will have to do
http://www.gmapgis.com/index.htm?4PXITx7HrG.gmp

Need more RC practice

Here the picture, before I crash it.

I'm too afraid of crashing my Bixler which I want to fly wit paparazzi.
This is a Slipstream 1m wing from flyingwings.co.uk. Should be fun, when the wind dies down.

Serial mode telemetry

I tried with transparent serial mode telemetry through the FTDI adapter on the FlossJTAG. Works just fine.
Perhaps my modems have a problem...

Exercise your hinges

I learned a important RC foam flyer lesson.

Exercise the control surface hinges!

Those foam hinges are very stiff. Your servo will have a very hard time deflecting them. And sometimes they warp rather than actually deflecting the whole surface.
So, wiggle them until they get a bit floppy.

Telemetry fuuuuu

XBee telemetry is super unreliable. When it works, it's a dream.
But so often I just get nothing. And neither combination of resetting XBee, Autopilot, usb-port or link-server help.
I'm frustrated.

[Update]
I think I know what happened. I had for some reason set  NO_XBEE_API_INIT. Stupid me. So it ran effectively in transparent mode.

Spektrum

Spektrum receivers work. I actually forgot how to bind the damn thing.

Lot's of tiny things

Little exciting progress. But here is what happened so far.

• I got indoor GPS Fix.
• I can bind both Spektrum satellites.
• I have severe Telemetry problems. The modules often don't sync up.
• And I hooked up a lighting system to my model. Looks very flashy, literally.

RC Experience

Today I flew my shitty trainer aircraft for some rc practice.
Crashed quickly after the second flight of course. But it is recoverable.
But I'm afraid the same can't be said for the speedcontroller. It lost its magic smoke.
The one I bought was too small for this engine, and when the prop stalled at the crash it must have had enough.
It nearly split the little FET in half. :)
More simulator flights from now on I suppose.

GPS Pinout

I just got a response from csgshop about the pinout of the MAX-5Q GPS Eval board.

VCC is 5V and logic level is 3.3V

[Edit] UBLOX MAX-6Q of course. Not MAX-5Q

Simulator flying

I finally accepted the necessity to do some simulator flying before I do the real thing™.

I choose CCRCsim and a Esky Hobby RC Simulator from a colleague.
I'm very pleased with the results. I think this will help avoid some very costly flight errors.
CCRCsim is opensource and runs under Linux, OSX and Windows.

The graphics are not modern gaming standard, but pretty frickin amazing actually

Fiat Lux

I will be sharing progress for my Paparazzi UAV project here.

I will be flying a hobbyking Bixler with Lisa/m 2.0 and 2.4Ghz Xbee telemetry.