Remote controlled flying things-Quads, Hex, Octo or anything

[Marschy]

Hello Chris, Great video.

I was just reading on rcgroups about yaw wobble, then I see that you have some in your video. Apparently if the motor is wobbling while it is sitting on the ground, this indicate that the yaw P gain is too high.

Cheers, Marschy

[Spartan]

Hi Spartan,

Hang in there mate, just on the Tx and the antenna I would suggest you have fried the TX Module, if there is no antenna on it then that power is reflected back into the TX and it cant handle it as another X amounts of milli watts are being pumped into it.  So it cooks itself.  It would explain why you get really bad transmission range.  But they normally fry so bad they output nothing.

I would suggest you leave the antenna attached in future, they may not look it but they are pretty robust.

On the vibrations I cant remember what frame you were using, I think you were doing a new version of David's Tricopter?  If so is it the one with the CF arms?  Also what have you used to isolate the KK2 from the frame (do you have foam between it and the board?).

Chris

Thanks mate.

Yeah I'm hoping it's the Tx module - if it's not then I'm not sure what to look at next but I'll cross that bridge when I come to it. Half the reason I bought the case is so I can leave the Tx sitting somewhere with the antenna on and not worry about it falling over or damaging the antenna.

Yep using Davids V3 with the carbon fibre booms. I've mounted the KK board using some double sided sticky foam from HK - it's about 5-6mm thick.

I bought a different model Orange Rx too because I've always been a little bit nervous about the length of the 2 short antennae on the 615 rx I'm using now. I don't think the Rx is the problem because it flew perfectly to the limits of sight when I had it running well, so I'm definitely trying the Tx Module first.

A mate of mine has an Orange T-six Tx that I was going to try out to eliminate the Rx as being a problem but he's away at work for a couple of weeks so I'll have the new module before I'll catch up with him again.

[CBRK]

Well I got to go flying today and this time the action cam worked, amazing what happens when you have a charged battery 

I also tried the some Yaw D tuning, when I had Yaw D at 0.003 at start up I would get a wicked side to side motion on the yaw, so I dialed it back to 0 and it stabilised, once stable again I turned it back on it seemed to remain stable.  I took off and the Yaw D set at 0.003 did wonders.  I dont like fact I need to have 0 at start but then 0.003 once I'm ready for take off.  At 0.003 it did seem to reduce the slow wiggle I get.  I will keep trying with some changes, I will dial it back to 0.001 and see if that helps.  I'll also try 0.005 and see what that does on start up :-)

I took her up to 30m today and got a good view of Chatswood from my local oval.  Here is the video of that part:
http://youtu.be/iXRTzZkQQHE

I did a bit of loiter mode flying and on occasions it would want to veer off so I'll need to review the flight logs to find out what it thought was going on.

Chris

I think I need to recalibrate the gimbal, seems to have a slight lean on it.....  It now weighs AUW 1.7kg and I now only get around 13-14 mins flight time on the 3s 5000mah battery.  Now that I've got around an hour of flight with the gimbal mount I'm pretty confident it works I'm happy to try even higher, it would be better if I could trust loiter mode fully.

I have received my new programmer and I've partially restored the APM2.5 board.  I can load on a firmware (ac 3.2) via the programmer but I cant seem to get the 2nd chip which handles the PPM signal and shares the USB port to work properly.  I cant load firmware via Mission Planner yet but I can connect with MP when I load it via the programmer.  I've tried programming the 2nd chip with the bootloader from the ardupilot site but it hasn't worked. 

I'll need to do some more work with it, I'll try it when I return from my trip.  There is a regulator issue that I'll check, just in case it has happened at the same time.  I want this board to be my test board that I can load custom versions of firmware to try new things but it mainly stay on the ground or go in test machines.

Tonight I will be hooking in Ch 7 to do a manual tilt on the gimbal.  And I'll try to fit the sonar onto the Tricopter,

I should get in one more set of flights tomorrow.

Chris

[CBRK]

Hello Chris, Great video.

I was just reading on rcgroups about yaw wobble, then I see that you have some in your video. Apparently if the motor is wobbling while it is sitting on the ground, this indicate that the yaw P gain is too high.

Cheers, Marschy

I missed this post until now, I'll turn Yaw P down a little and see what happens, it would be good to avoid having to change Yaw D on the fly all the time.

Was this Stab Yaw P or Rate Yaw P?

Chris

[Marschy]

And I'll try to fit the sonar onto the Tricopter,

I haven't tried the method successfully that I described earlier that I had great success with my LV-EZ4 sonar on the HC-SR04 sonar yet. Getting a reference voltage reading from the SR04 sonar proved to be problematic as it was jumping around quite a bit. Unless you can get a reasonably stable voltage reading when the copter is sitting on the ground as your reference zero altitude reading, setting up the scaling will prove difficult.

I'm going to try setting up the SR04 sonar again before I do the pitch and roll tuning.

I also thought of a way I can do the pitch and roll tuning. I am going to use some 3mm ply which I will elongate at the pitch and roll points where I will be tethering it and double up that area with some scrap ply to strengthen it at the point where it will be tied off. Then I just have to have a means of mounting my quads/tri onto the ply so that it sits level and can be tied down onto the ply without doing any damage to the frame. This way I am not trying to stretch the quad frame between the posts on my pergola where I will be tethering it, and the ply will be what is under tension. I'll get some photos once I get it sorted out.

[CBRK]

I also had a good crack at acro without the trainer mode, it was brilliant.  No flips or tricks with the tri - too heavy and not the platform for it but I sure was able to to set it for the wind and let it hover there.  I can see myself using this mode more and more in the future.

I think I'll turn the trainer mode off on the mini Quad too (after I master flips, etc with trainer mode on  ).

I've reviewed the logs quickly and nothing jumps out why I had a sudden 20 deg bank in loiter after a minute of sitting in loiter.  I thought it might be a sudden loss of HDOP due to the sat count dropping, but it doesnt appear to be it.  I can see where I put input in that is ROLLIN, but that is after the waypoint distance has moved - note that prior to the waypoint distance changing there was no ROLLIN, but then suddenly the waypoint distance goes up and my roll occurs which then see's a ROLLIN shortly after as I try to move the waypoint location (you see it dropping down).

What is interesting is I gave it an extra second before trying to reset the position, i should have waited a bit longer.....  It appears that the waypoint distance is only 2 to 3 m so I should just learn to let it do it's thing.  I'll try that again tomorrow as 2m is well within 1.8 HDOP scope.  I was impressed that 85% of my flight was done with 12 sat's locked on, I love this GPS (Neo-7N).  I have another one coming with the Pixhawk, I cant wait for the Ublox-M8N to drop in price.

Oh and here are the pics of the CF frame:




Chris

[Marschy]

Looking good Chris, are you still planning on swapping over the FPV250 hardware?

Here is what I was talking about with regard the tethered jig for setting the roll and pitch gains.

I have cut out a length of 3mm plywood and laminated triangles onto the points, where eventually a hole will be drilled to pass a light rope through.



The base board should give me enough room and options to secure any of my quads running with the APM to the board.



Then I can put a rope through the hole that will be drilled through the centre of the triangles and tie it off nice and tight between the poles on my pergola. Once the quad is secured to the base board, I can tune the roll and pitch gains as per this video.

http://youtu.be/aq1jXHMiJgg
	Small | Large

[CBRK]

Looking good Chris, are you still planning on swapping over the FPV250 hardware?

Here is what I was talking about with regard the tethered jig for setting the roll and pitch gains.

I have cut out a length of 3mm plywood and laminated triangles onto the points, where eventually a hole will be drilled to pass a light rope through.



The base board should give me enough room and options to secure any of my quads running with the APM to the board.



Then I can put a rope through the hole that will be drilled through the centre of the triangles and tie it off nice and tight between the poles on my pergola. Once the quad is secured to the base board, I can tune the roll and pitch gains as per this video.

http://youtu.be/aq1jXHMiJgg

Hi Mark,

I'm not sure yet, I have the HK frame plus a spare as well as two of the FPV250 kits.  I think the biggest challenge with switching over will be getting everything to fit.  It's a tight fit already but I think I can do it, if I swtich to using the mobius as the FPV camera too (I have the cable harness for it) I think I can make it all fit.  The only concern I have is that I sling the battery underneath the miniQuad at present, on the FPV250 frame it will have about 3-5mm clearance so it will be resting on it on some terrain.  It currently has around 1 - 1.5 cm clearance.

Ummm, how is the quad secured to the frame?  Unless it's by cord I would have thought it will impact the behavior of the frame?  I like the idea of it though make it easier to test different frames.

Chris

[Marschy]

Ummm, how is the quad secured to the frame?  Unless it's by cord I would have thought it will impact the behavior of the frame?  I like the idea of it though make it easier to test different frames.

I'm thinking on my feet as I make this, but it's coming along. I've CA'd some pool noodle foam onto the board and it is being held in place while the glue sets with velcro tape. The velcro tape will also hold the frame onto the board.

I'm trying to keep it as light as possible as I think the weight of the board will influence the gains settings. It weighs 88 grams, so if it does affect the gains, once this weight is taken away the gains could possibly be too high. Time will tell, I'll test it this weekend and get some more photos.

[Marschy]

Crikey, this might work.

This picture shows my X-mode Alien attached to the board with velcro and the pitch attachment point which I will line up with the CoG.



This one shows both the pitch and roll attachment points. Only pitch or roll is adjusted at one time. I just have to centre the quad on the board and tie it off between the posts of the pergola and I should be right to go. The pool noodles have worked a treat, I can tighten the quad down onto the board very securely without damaging anything on the underside of the quad.



The mini quad fits as well.

[CBRK]

Crikey, this might work.

This picture shows my X-mode Alien attached to the board with velcro and the pitch attachment point which I will line up with the CoG.



This one shows both the pitch and roll attachment points. Only pitch or roll is adjusted at one time. I just have to centre the quad on the board and tie it off between the posts of the pergola and I should be right to go. The pool noodles have worked a treat, I can tighten the quad down onto the board very securely without damaging anything on the underside of the quad.



The mini quad fits as well.

Looks good, I'll be very interested in the results as it is definitely different to other designs.

Well good news on my gimbal front, just plugged in signal and gnd from A11 into the pitch connector.  Configured my transmitter to output on CH8 as the pitch control, set it to -90 and +90 in APM.  Had to manually set the PWM min and max values, I can't have calibrated that channel before.  In simpleBCG configured the same + and - values as APM and removed the trim setting that was there.  I also set the pitch control to listen to the PWM signal.

It then just worked, it pitches up and down on request, it's great.  I played with the speed element and went for 4 as a good compromise, 5 was ok but felt 4 was better and 2 was way too slow.  1 was so slow it wasn't funny, took seconds after I had selected a value for it to tilt up to it.

I wanted a 90 degree down I decided to have +90, I had it originally set to 90 down and 45 up but the center was then about 20 degrees down.  I think I could have fixed it with a trim value but I don't mind have a +/- 90 degree control.

I'll be testing it out tomorrow morning, so hopefully it all goes well.  I'll also be testing a much reduced Stab Yaw P - down to 1.2 and a slightly increased Rate Yaw P and I.  I also have left the 0 to 0.003 on the CH 6 tuning, so I can work out if I can remove it.  I think the reduced Stab Yaw P will reduce my Yaw authority by a reasonable amount so I might have to use some expo on the channel to make it a bit stronger.

This will be my last flight before I take it out to Abercrombie River NP for the Aus Day long weekend.  Feeling pretty happy about it all.  Not sure how I'll go with flying in the NP.  I felt confident taking it up to 30m, so I'll see if the conditions are alright for a bit higher in the morning.

Chris

[Marschy]

Hello Chris,

Would you mind sharing your PID's for the gimbal. Ever since I reflashed my gimbal controller with 2.1 null firmware the motors emit a constant tone. If I can sort that out I'd be more than happy. Please share the motor gains as well.

Cheers, Mark

[CBRK]

Hello Chris,

Would you mind sharing your PID's for the gimbal. Ever since I reflashed my gimbal controller with 2.1 null firmware the motors emit a constant tone. If I can sort that out I'd be more than happy. Please share the motor gains as well.

Cheers, Mark

Hi Mark,

Below are the screen shots.

I took it up to 40m today, got some great shots.  I thought I would test the descent rate on it as I've had troubles in the past with the frame.  So up i went, tilted the camera around as I was going up, anyway I get to 40m in loiter and then I start my descent.  I brought it down to 30% and it was still staying up there, so I brought it down some more.  Seems my expo was so high (combined with how loiter works) that I didnt get any descent happening until it was barely 15% on my throttle.  And it dropped like a brick (12m in 4 secs), luckily I put it into stab mode which has much better control and I guided it down the remaining 20m nice and smoothly.  I've since changed it from 60% expo down to 10%, handles good in Stab mode still but now I have some control in ALTHOLD and LOITER.....

When it was coming down the frame tilted nearly 90 deg to the left (had my going for a split second) the camera gimbal footage barely notices it.  On review of the footage I thought, maybe I was seeing stuff.  I'm shocked now I've reviewed the FPV cam's footage.  LOL that gimbal is good! I was being buffeted around up there I'd say a good 20 to 30 deg correction was being applied by the APM to hold me in position, damn it was windy up there.

Since it does such a good job once I'm back to work I'll pick up the YAW module and a brushless motor to go with it :-)

The Yaw PID changes barely affected my Yaw authority, it's slowed by maybe 10% at most.  It is alot more stable, less sway but there is still some when I remove Rate Yaw D, so I'll leave it on for the time being.  I worked out that if I keep Rate Yaw D on and I actually slide the power up close to take off speed my rapid oscillations disappear!  It just takes a few seconds.  I think the oscillations are caused by the new battery tray, the rear is unsupported so if it gets moving it can pull the frame a bit so I will put some twine around it to support the rear of it.

Chris

[Marschy]

This is not going to help the FPV cause in the US at all.

http://www.smh.com.au/technology/technology-news/drugladen-drone-crashes-near-usmexico-border-20150122-12wekr.html

[Marschy]

Thanks for those screen dumps Chris, I'll give them a go this weekend.

I sorted the Bluetooth adapter out on the mini quad. I had to change the baud rate using the FTDI programmer which meant I had to make a cable connector for it and I downloaded a terminal emulator so I could send AT commands directly to the com port. Droidplanner is now working happily on my Nexus tablet. The good thing is no external antenna which is a win in my opinion on the mini quad.

I use the minimOSD for telemetry, not the Bluetooth adapter. I only plan on using Bluetooth for changing settings so range is not an issue. The other thing that I think is a win is now I don't have to worry about the OTG adapter or ground module for my tablet.

[Mandrake]

I am on the Phantom Australia Facebook forum and some guy in Townsville just got fined by CASA $850 for flying within 30 m of people .. His YouTube video was used as evidence..
They are watching us ,
http://eftm.com.au/2015/01/850-drone-fine-casa-using-youtube-to-identify-illegal-drone-use-19067

I think that's it .

[Marschy]

A guy on the ZMR250 thread on rcgroups posted a video of his mini quad chasing a freight train in the US. He was flying within a metre of the top of the train, and buzzed in front of the driver.

People were patting him on the back saying 'awesome dude' and all sorts of accolades.

I pointed out that what he did would be illegal in Australia because we cannot fly within 30 metres of a moving vehicle, someone else piped up about 'what was a mini quad going to do that could hurt a freight train, it would be like an insect splatting the windscreen'. I responded by saying if the driver was distracted and he missed a signal, how cool would it be then. Another member piped up and said he was pretty sure it was illegal in the US as well. The video was up for a total of about 10 minutes before he pulled it off youtube.

[Mandrake]

OK guys I need some pointers / ideas ...

The Alien has plenty of ZOOOOM on Aileron , Elev and Roll BUT Thrust has very limited range -- I've done the ESC config thing a few times and all dings and pings say that I've done it correctly .. Settings in the TX are all 100% to 100% for A E T and R all same ... On the bench the motors throttle up nicely from idle to max but as soon as props are on and it starts up I have limited thrust range

So I'm thinking that I don't have enough power to run all 4 motors at the same time ....  A E and R only run 2 motors ?? 

So bigger battery ?
Extra battery ?
or change the ESC 4in1 to 4 seperate ones Might this help ?

Confusion here I can tell ya ...

Steve

[Marschy]

Is your rig 3S or 4S Steve?

[Mandrake]

Still 3S - 4500 mah 35C battery

[Marschy]

Direct from the product list on Hobbyking:

Required:
Motor: 2820-3110 (recommendation)
ESC: 30A (recommendation)
Prop: 11~12in (recommendation)
Battery: 3000~5000mah 4s 14.4V (recommendation)

[Mandrake]

I'll get a 4S and see what it can do .... Budget freeze at the moment ...

[Marschy]

I don't know if it's the ESC's, but I would have thought that at the very least the motors should spin up fine, maybe not lift off if underpowered, but definitely spin up to full power. What motors did you end up getting after the Multistar 935kv's?

[Marschy]

Sounds to me like wrong size propellers.

[Mandrake]

1045 props , Multistar 2212 980 kv , 4 x 25A ESC 4in1 ...