Long overdue update:
Well, I finally turned the corner on all this and have things working quite well. Three things caused most of my issues:
1. There was no way this bird was going to fly the intended payload with 14" props and 5S batteries. I'm really surprised that the staff at Droidworx were ok with this selection. Although they did point out that there are a lot of variables and every build was different so they only make recommendations. But I would think that after selling thousands of copters they would pretty much be better informed than most people. Nonetheless e-Calc confirmed my choice wasn't that good and later on someone from Droidworx did recommend that I move up to larger props and 6S batteries. Easy to say but about another $1,000 out of my pocket. I also had to change the ESCs as there was conflicting info regarding if they could handle 6S voltage that I only discovered after one of them melted and shorted out in flight. The copter wasn't damaged but of course changing ESCs means virtually rebuilding the copter.
2. One of my big goals was to see if I could get an APM 2.5 flight controller to work with this copter. I'm quite involved in open source projects and know one of the Ardupilot programmers who I work with in a group that develops UAVs for search and rescue. I also want to be able to talk to those who develop the software so I know whats's going on. I had quite a few issues with the APM that I falsely put down to tuning issues. So that wasted a lot of time. The main issue was that in my comparison of how Bluerex's DJI based Droidworx copter flew I failed to realise that many people flying DJI flight controllers are flying in GPS modes whereas I was using the APM totally in manual modes given that the GPS modes in that FC are used primarily for autonomous flight which I didn't need for filming purposes. Indeed, at that point in time the APM software didn't even support take off and landing using GPS modes.
After long discussions with Ardupilot developers about using copters for video work the light globe came on and I started talking to them about how DJI works and why. They knew all that of course but I just don't think these things were the priority at the time. Anyway, it's a brilliant team of professional coders and they were already working on some things that turned out to be a game changer for me. The first thing that changed was that they added the functionality to take off and land in GPS modes. So this gave me a valid point to work from in comparing DJI and APM software from. I don't really advocate taking off in GPS modes but that seems to be what most DJI users are doing. Anyway, after adding this functionality to the APM I found that my SkyJib would perform damn close to what I was seeing with BlueRex's copter. Indeed, even with default settings and no PID tuning at all, the copter was flying beautifully even with my full frame Nikon loaded up. Needless to say I'm very happy about this.
3. The weather has been awful and slowed me down. For three weeks we have had high winds and I've only had a few one hour patches here and there to do any testing. So most of the time I just can't test as I tune the copter, add video TX and bring it all together.
So, in summary:
SkyJib very happy and flying just fine. After I get my gimbal balanced and practice a bit I'm hoping to have some actual video to finally share here. Be kind as I cut my teeth in this exciting area please. I've actually registered a business here in Australia and I'm currently studying to obtain certification from the Civil Aviation Safety Authority so I can do commercial aerial photography. That process will take another 6 to 9 months.
As for using the APM and Arducopter software for flight control? Well, I have an unopened DJI Wookong sitting here that will be up for sale shortly. I have no need for it. The APM and software has so many more features and I get to actually talk with the developers and contribute to the community. Having said this, I wouldn't recommend that people who might want to use it or switch to do so for another six months or so. Why? Well, the APM is based on an 8 bit board and the code quite simply doesn't fit any longer. The CPU is at max as well. The next generation hardware from 3DR is the PixHawk which will be shipping soon. It runs the same exact software but is 32 bit which will give the developers a lot more memory and power to implement all the cool new things coming soon, like auto tuning. So if you are flying a complex system like an octo set up for filming, at this point in time my recommendation is wait for the Pixhawk. If you are interested just have a look at the 3DR web site. It's quite impressive and at around $250 a lot cheaper than most flight controllers.
So there you have it. If the weather here ever calms down hopefully I will play much less with the copter side of things and concentrate much more on getting some quality video of this part of the world which is really stunning.
A huge Thank You to everyone who contributes here. All the info and feedback has been extremely valuable. I really appreciate the time people take to share and help out.
