|Jul 24, 2012, 05:03 AM|
Joined Dec 2009
HK Hexacopter Build on KK2.0 HELP!
I have started to build my first Multirotor, in this case a Hexacopter in X formation. I have run into an issue with getting it stable an figured I would start a build topic and get some help.
Few things first, build plenty of scratch build aircraft in my time including nitro and electric heli's and fix-wing and I am no stranger to FPV(the reason for the hexacopter in the first place). Have a background in electronics so no stranger to building circuits and soldering, what I'm trying to say is the issues I am having are not related to the pilot or construction of the hexacopter.
1 x HK Talon Carbon Fiber Hexacopter frame
6 x Turnigy AE -30A ESC
6 x Turnigy SK3 2836 1040kv outrunner motor
1 x KK2.0 Board
1 x 200A power distribution board
2 x HK U5 5A/5V UBEC
2 x Turnigy Nano-tech 45/90C 4500mha in parallel
6 x Master Airscrew 9x7 3 blade props
1 x Hitec 9 channel Rx
1 x Aurora 9 TX
1 x Fatshark FPC camera & 600mha TX
1 x HK fiberglass Camera Mount
1 x 360 servo (pan)
1 x 180 servo (tilt)
14AWG wiring for the ESC and motors
10AWG wiring for the battery connection
Bunch of different Nylon threaded stand-offs and electronics "bread board" I had lying around.
So basically I put the top section of the Hexacopter kit together leaving the bottom off, and ten put all the engine mounts on. Simple enough, just make sure you use the right screws in the right holes, there are a lot of screws nearly the same size in lots of different packets often with different sizes together.........tip them all out in a large tray so you can line up the different sizes, easiest thing is to go for is a size that seems to be a little to large.
I then ran all the wiring for the motors and esc to make sure the ESC's would go where I wanted. From there I mounted the ESC's with some small Zip ties, pulled all the wires taught and shrunk some Heatshrink over them to help protect tem and keep them where I wanted (also made them look..Pretty!)
Before I put the heat shrink over I made sure the clear heatshrink over the esc's was as "shrunk as possible" and then cut the majority of the plastic off the metal heatshrink on the ESC (no need for 2 layers of heatshrink over the ESC's heatshrink). I was going to cut the heatshrink off the heatsinks after covering them but after running some test I decided against it as they were not getting hot at all.
OK, then I wired in all the main batter leads and the 2 x UBECs, mounted the UBECs to the power distribution board wit some double sided thermal tape then mounted this board into the belly of the hexacopeter with some short nylon standoffs.
I then soldered all the connections to each of the ESCs to the lines running out to the motors and got them all tucked away neatly, leaving just the ESC's power to be soldered to the power distribution board and walla!
This was all nice and easy and only took an our or so to do. I then mounted all the motors and cut the ESC leads to length and soldered on all the 3mm connectors for the ESC leads (got to be able to replace a motor nice and easy).
Solder on some plugs for batteries and feed all the controlled cords through from the ESCs and I could close the bottom up and see how rigid the frame was,. To my surprise it was rock solid and measured within 1.5mm from each tube/arm end and the same to the motor mounts, happy days.
That just left the control gear, KK2.0 mounted onto the more nylon stand offs and match perfectly with predrilled holes in that came in the Talon Hexacopter kit, fantastic! Couple more nylon standoffs for the a receiver mount made from some of that electrical "bread board", stuck the TX under the bread board and the servo with the camera on it on top of the breadboard. A plastic take away container over the top to protect everything during trials.
All ESC are calibrated and work correctly, all props mounted and tight, all balanced well and had need to adjust. On default settings on the KK unit in hexacopter "X" formation, can't get off the ground without a stack with self level on.
Watch a bunch of tuning videos, read some stuff on the internet, make sure height dampening is off, start with only P gain, the thing will get off the ground then start to oscillate badly and flip.
Play wit gain on self leveling etc, end up here:
- Can take off and move around without any gyro on, only go a a minute or so like this before I basically stack
- Turn gyro on, can get off ground, but went it gets out of ground effect it has same oscillation issue.
- Turn stabilization gain down to 2 and the other stabilization setting upto 100, I can take off, nice and stable just can't move around.
That multirotor calculator says the props have too much pitch and won't hover, yet I can hover manually so the KK board should be able to do a better job than me......Help?
Ps. Pictures to come.
ImagesView all Images in thread
Last edited by Redemptioner; Jul 25, 2012 at 02:06 AM.
|Jul 28, 2012, 01:04 PM|
United States, CA, Santa Monica
Joined Jul 2011
The 2.0 is very strange and inconsistent when not tuned correctly. Try turning the I to 0 and turn self-level off and only fly with the P. When you start out with the P- try 30 or so. See if that helps. Basically with the control board- it's just a matter of trying different combinations that work. I'm sure you will find something sooner or later if my suggestion doesn't work.
|Jul 28, 2012, 11:26 PM|
Joined Dec 2009
Thanks Lpbug, persistence is the key here.
I got one of my buddies brothers (aeronautical Engineer) to do the math's on the prop lift, I'm all good the hexcaopter should hover no problem.
I have spent the last week going over every thing that is available on the internet about tuning these things and after making sense of it all (and some help from my father with how these gyro feedback loop works) I am getting close to a stable platform, this is what I have so far;
It seems the ESC being used is the biggest variance in the settings used for most copters, many people flash their ESC's with custom firmware. Most copters I have seen that fly perfectly have a flashed ESC. This is not to say you can't get a good stable copter without flashing your ESC, its just the firmware loaded in most cases gives a more "heli feeling" to the output from the ESC. I the most part the acceleration rate of the motor is retarded slightly making a softer increase of prop speed which tends to give back more (or less depending how you look at it) range to you PI settings.
Getting the copter perfectly level often won't be the same as the sensors on the board "level". With every thing centered and PI settings low(PG-30, PL-0, IG-0, IL-0) heading hold off, stabilization off, high dampening off, ESCs calibrated, no wind and centre of gravity in the middle of the copter (even with batteries), take off, the copter should be reasonably stable . If it drifts in a particular direction every time, then you need to re-calibrate the sensors with a slight increase in height on the particular side that the copter drifted in (ie, setup on the same flat/level surface and put something under one or more motor mounts to raise the side slightly higher than the rest on the side were the "test flight drift" was seen). When this is pretty good then start editing the PI settings.
Gain settings in PI Editor
PGain - This relates to the amount of movement response the gyro will allow to the axis, so when you increase this value the copter will move in the direction of stick movement faster and allow more travel/lean in that direction per second. Increase it too much and the gyros end up oscillating due to it trying to correct the movement too quickly causing the gyros to over compensate for movement.
Decent Amp ESC that are not flashed tend to have very low Pgain settings in comparison to ESC's flashed with a custom firmware.
PLimit - This relates to how much power is allowed to be use to correct orientation. So if a gyro is detecting movement that is not being given from a stick input, this is how much % of power is allowed to be used to counteract the movement.
The thing to remember about Plimit is you need power for other axis. Many people seem to have Plimit around 100%, meaning a single axis correction can saturate the motor outputs leaving no power to correct the other axis. I have found that on a non-flash ESC you don't want to use more than a total of 100% on all axis combined. There is a connection to ILimit but I will discuss this under the ILimit details.
IGain - This relates to the heading hold function of the board and specifically relates to the output needed to keep the copter facing on a particular heading.
When setting up the copter initially the relationship between PGain/IGain/Level settings can really be a pain as you have to land, adjust, fly, land. adjust.....500 times.
ILimit -This is the Limit in % of power for the Heading hold correction. When this is on, PLimt plus the ILimt will be the total power used to correct the unrequested movement within the axis. So if you have a high Plimt (say 80) and you add in some ILimt (say 30) it can be very easy to saturate the motors if there is a bit of wind or you are trying to recover from a fast change in direction leaving no power for correcting the other axis.
PGain - This is the same as PGain under normal PIeditor, but it is over the top of the PI settings, so affectively taking the signal being generated by the normal PI settings and then reducing this right down and having it feed into another master PI editor. The lower the settings the less control you will have over an axis when you give a stick input. This has a very dramatic reduction on the movement when this is set low and I found the need to turn off EXPO when turning self leveling on.
PLimit - Unlike the PGain in self leveling, it does not seem take the PI settings output and reduce it very much, to be honest it seems to override whatever settings are setup in the PI settings.
IGain/ILimit - Apparently these don't work.
String testing is just not possible on such a large and powerful copter (about 7KG of trust), and is just dangerous to say the least. The calculations did saying I am wasting power as I don't need anywhere near that much lift and to got to a set of 8x4 props to get back a tone of flight time and noise reduction.
PGain - 45
PLimt - 40
IGain - 40
ILimit - 25
PGain - 55
PLimt - 25
IGain - 28
ILimit - 30
PLimit - 60
I hope to make a flight video this afternoon
Last edited by Redemptioner; Jul 31, 2012 at 08:27 PM. Reason: gramma
|Aug 04, 2012, 07:53 AM|
Joined Dec 2009
Ok so the flight video did not go so well, when I thought I was recording...I wasn't, and vice versa.
Did get one shot of what happens when you put some values into the IGain and ILimit settings under self-leveling. Instant flip.
|Aug 12, 2012, 09:18 PM|
Joined Dec 2009
Well, I found what happens if you don't pay enough attention when you are making up conversion plugs. I made a couple of XT60 to HXT 4mm as I am running the HXT's for the battery connections due to the high current and the connectors on my Battery charger is the XT60.
Unfortunately when you take this little converter plug and connect it to a battery with TX60 connector it reverses the polarity on the HXT plug (stupid non-directional plugs). The result was a big spark and smoke, fortunately it only killed the power distribution board and nothing else, $3.50 fix. Blew the copper off the board on one side and melted the paths on the other side.
|Thread||Thread Starter||Forum||Replies||Last Post|
|Help! Tricopter and HK KK2.0 board - Tail servo jitter||rcbif||Multirotor Electronics||156||Nov 30, 2013 07:03 AM|
|Discussion I want to build a Quadrocopter/Hexacopter. Need help!||Linuus||Multirotor Talk||11||Jul 04, 2012 08:35 PM|
|Help! Hexacopter HK V3 FC Frimware Help!?||annonymous||Multirotor Talk||14||Apr 17, 2012 11:54 PM|
|Discussion Wimped out on first time trying FPV on my hexacopter, any help to get over the fear||jasper7821||FPV Talk||13||Mar 28, 2012 08:15 AM|
|Idea HK tricopter,hexacopter on the way?||baconflyer||Multirotor Talk||6||Mar 19, 2012 01:00 AM|