A little while ago I drop a vid in the Tricopter thread show 3 basic yaw mechanisms and explained what I thought were their pros & cons. It was a very simplistic view on my part without due test & trials, what I thought would improve yaw speed was nothing of the sort.
Not sure why I did this but I started to test different props on my CFT400 tricopter running a KK2.1.5 with Steveis v1.19s starting with stock (factory) settings and adjusting gains to improve flight control.
Now in the process of this experiment I came to the conclusion that having a yaw mech that can exceed 45deg each side of the vertical was all put useless as the FC did not require that extreme. Not until I slowed the videos down was I aware, nor in the same process was I aware how much input the FC does to the yaw servo independent of your stick inputs. The FC really does all the hard work.
3 types of props were selected (pretty much stock on hand) HQ brand Multirotor 8 x 4.5 polycarbonate, HobbyKing knockoff DJI 9443 in soft plastic and a no name plastic 8 x 4.5 slow fly. The yaw mech, David Windestal's steering knuckle.
I had been running the HQs for some time and the tri was tuned using them so I reset all to factory and started again. I settled on only adjusting SL gain, basically for the new guys, hopefully give them an insight in what to expect and how much adjustment would be required for safe flight.
The differences in flight were very apparent on factory sets and gain...Continue Reading
Posted by cynr100 |
Oct 09, 2013 @ 05:04 PM | 3,535 Views
I'll start a separate blog for the QBrain itself as the first setup test failed.
The Turnigy program card will work BUT you have to program each ESC individually the "hub" connector worked on two only.
Program throttle ends points individually again not with the "hub".
No luck with the KK2.0 board, incessant beeping and no throttle, KK board shows no throttle input.
With direct connection to a Rx throttles work!! more investigation to be done on the KK2.0
Rudder servo not working.
Calibrate the ESC using buttons 1 & 4 on the KK2.0, hold down before connecting battery wait for a single peep then move throttle to off, acknowledgement peeps followed by a single peep should be heard.
When setting up as a Tri Copter power needs to be provided to M2 or M8 to provide power to the servo at M4, power to M1 is still required.
M1 to M8 are not ALL connected as stated in the original KK2.0 setup.
Okay...................so I just destroyed 2 not 1 but 2, KK2.0 boards following this power setup.
I was just a plain dumbass, DO NOT provide power to the M (right) side of the KK2.0. Power to M1 & M2 or M8 only provides magic smoke.
Power to the board can be provided at port 5 of the RX.
The bedding of a new project mating an MD500 scale body to a Tri-Copter.
The build is to be around Bird Dog 900kv motors, BD 10x5 props on a KK2.0 board.
Will start with this pic where the idea came from.
CF tubes and plate has arrived along with the HK QBrain 4x20ESC. Here's a pic along side a Turnigy 18A ESC, the QBrain is larger than I expected and it is not going to be a weight saver. At 112g it doesn't save weight on a Turnigy 18A @ 19g nor even a 30A @ 25g, oh well a few dollars saved only.\
A very quick scratch built tri was setup to test the QBrain which has turned into a "No QBrainer" so I have started a separate blog to sort the issues before I start cutting and installing any CF to the MD500, I may yet have to go with 3 ESCs
Disaster, see above, toasted my 2 KK2.0 boards. Will be ordering Turnigy 30A ESCs for this project. In the meantime will start working on the CF rods and plate for fixing to the fuselage.
Coming along now with CF rods and plate cute to size and dry fitted, will probably go with a silicon adhesive to fix to the shell.
Sikaflexed the CF frame to the FG body, something looks crooked. Can't quite get that square look to the skids, shouldn't affect flight performance. If it starts to annoy me I'll remount the frame.
KK2.0 board, power dissy, ESC's and 2 JST outlets installed. The dizzy is velcroed to the CF frame and each ESC will be velcroed to the fuselage.
Just a few changes from the first build. She now has a TBS Core because I want to run a 4S but only need 12v for the 600TVL and Vtx.
Have also gone for the all black arms, circa Mongrel Gear.
Have ordered a AEE Magicam so whilst waiting I'm using the trusty HK Wingcam which suffers from jello. After experimenting with different types of anti vib pads I have reverted back to the trusty pool noodle foam, its looks bogan but is very effective.
I was told to go with 10x5 props but noticed that things were a little warm and she was a little hesitant in flight. I have now gone back to 9x5s, running gear is cool and flight is a little more sure footed. I have lost about 1 minute of flight time but I have also change the low battery settings so that now I finish with 35% rather than <10%. I'm not going backwards to do comparisons, if I get 6-8 minutes flight from a 3S 3300 pack I should be more than happy with 4S 4500, it's not a fing competition.
Not getting much flight time, and then only in my yard, due to a knee op. Another few weeks and she'll be right .
Desperate to get some leg extensions as wet season is coming on and where I fly does not get mowed that often. Searched around and the only ones I could come up with were from HK International W/h too expensive with freight
I had some 10x10 box CF left over from the NewsCopter and came up with this as a temp measure, I think it might be permanent.
Posted by cynr100 |
Aug 20, 2013 @ 04:20 AM | 3,599 Views
There's plenty of examples out there of TBS Disco builds , just so I don't get left out here's mine.
All the gear was sourced from www.mongrelgear.com.au I have gone without the "Core" unit as funds did not stretch that far
Bird Dog 900kv motors
Bird Dog 30A Optos with SK f/ware
Naza-M Lite with GPS
Bird Dog 9x5 CF props
I'll get it up and running before adding my FPV, Sony 600TVL,Hornet OSD and Fatshark Vtx
In reality a very simple build once you sort out wire orders and locations.