Replies to This Discussion

Permalink Reply by Jeff Smith on August 5, 2012 at 1:59pm

Hi Guy's :-) Add 1 to the team. I knew this was a long term project when I signed on.

I have available a 600efl (flybarless), 250 flybarred, and 250 flybarless, a pair of APM 2.0's (one mounted and flying in the 600, the other on the desk currently). I also have on the way once shipping a APM 2.5, ArduIM V3, and a full array of sensors plus the Ublox GPS.

I get a few hours each evening and some time each weekend for flying.

I can mount my second APM 2.0 to my 600 just to capture data if that is what you are looking for at this time?

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on August 10, 2012 at 6:21am

So you have your 600 flying?  Can you show your mounting, and which code are you running?  Which heli?  Did you do any radical balancing procedures?

I've just purchase an electronic scale, 500g, 0.01g resolution, so I can try and balance my blades better.

• ▶ Reply

Permalink Reply by Jeff Smith on August 10, 2012 at 7:55am

Yup, been burning off packs just in stabilize mode.
It's an Align 600efl with a Neu 1915 motor, Phoenix ICE HV80A, 12S 2650mah nano-tech 45/90C packs, TH9X (standard software), Align BEC with 5V regulator.

I have set the blade tracking but never balanced anything.

The APM 2.0 is running 2.7.1 software and mounted about 6 inches back on the tail boom. I have wired it on 2 foot of HDMI cable to test different locations.

I'll get a few pictures of the layout tonight.

• ▶ Reply

DeveloperPermalink Reply by R_Lefebvre on August 10, 2012 at 8:00am

Very interesting, you have it running on 2.7.1.  So how well does it fly?

Previously, my flybarred 600 flew like a giant Blade CX.  Totally stable, predictable, easy to fly.  Do you still have that?  Or do you feel like you have to chase it around a bit?

• ▶ Reply

Permalink Reply by Jeff Smith on August 10, 2012 at 8:15am

I'll throw up a video if I can isolate one that illustrates the stability, for me I would say it is very stable, though I do seem to have a high frequency oscillation that is present in windy conditions that is absent in the calm.

P.s. I have no vibration isolation in place, the APM is mounted to Velcro and foam cut to the radius of the tail boom to help retain level.

• ▶ Reply

Permalink Reply by Vernon Barry on August 10, 2012 at 8:57am

Just a tip about tuning PIDS. When  you tune one super tight (Values near the point of oscillation in a neutral environment) adding load, mass, or outside influence can cause the oscillation requiring you to lower the PID values.

Keys about the values to remember

P is gain, in it's purest form. This value is what you always adjust first (raise it) until you get oscillation and then back off a little bit

I is additive to P but works in the time domain. The longer the sensed position does not equal target position, this value adds to P. Thus, pure P tuning may require a backoff setting when I is tuned because I adds forcing a high value of P to reach oscillation.

That last thing you touch is D and only after extensive testing of I an P. It can remain at a lower value in most cases.

Again, be cause the over all response is rather slow, P and I are the largest factors here.

 

• ▶ Reply

Permalink Reply by Jeff Smith on August 10, 2012 at 10:27am

PID's for the tail are VERY low, I'm teetering between oscillations and a mushy tail. I strongly believe it is due to the fact that the MPU-6050 is some 10 inches off the center of gravity causing amplified yaw signals combined with lateral acceleration. Like putting a standard HH gyro out on the tail boom next to the tail rotor :-)

The current location was ONLY to get the magnometer away from the monster inrunner I'm using. On Tuesday I will have my 2.5 with external sensors, so I'll move the MPU as close to the CG as possible and mount the mag,optical flow, sonar, and Lea-6 accordingly... Should make a good difference for PID tuning.

• ▶ Reply

Permalink Reply by Vernon Barry on August 10, 2012 at 10:48am

Ok, that makes good sense and would explain the behavior. You basicaly have a mechanical gain adding to any of the software gains, and thus even with good tuning, may be dificult to control.

• ▶ Reply

Permalink Reply by Jeff Smith on August 10, 2012 at 10:56am

Only good pic I've got on my phone is the HDMI routing to the bottom of the APM, I think the wall-o-servo wires around the inboard GPS is partially to blame for the poor reception issues. After the mod the GPS works significantly better.
http://i217.photobucket.com/albums/cc187/DrKnow65/8ac3be02.jpg

• ▶ Reply

Permalink Reply by Vernon Barry on August 10, 2012 at 11:05am

Nice work!!

• ▶ Reply

Permalink Reply by Jeff Smith on August 10, 2012 at 1:18pm

Thank you, it was sketchy as the insulation retracts significantly when heated, the shields like to fray (insulated with sections of yellow servo wire), and I had to be very careful not to create a ground loop in the system. It all worked out well in the end :-)

My issues with the APM are mostly with Loiter and getting the hover spot on.
Loiter seems to make the heli drift backwards tail down even if I set the hover (auto trim) intentionally nose down.

• ▶ Reply

Permalink Reply by Jeff Smith on August 10, 2012 at 5:52pm

Some more photo's of the mounting. Video is all on the wife's phone, will get that up later.

• ▶ Reply

• ‹ Previous
• 1
• …
• 140
• 141
• 142
• 143
• 144
• …
• 161
• Next ›
• Page