I would be interested in your views on buying a ready to fly Quadrotor like a Gaui 330S for example and then adding an autopilot at a later stage against starting with a full drone kit (Adrucoptor 3DR + Ardupilot).
I kind of like the idea of getting a 'manual' quad to play with first before getting into the autopilot side, but that means buying something like a Gaui and trying add in an autopilot at a later date.
There seem to be quite a few ready to fly manual kits around and some are not that expensive.
I might be wasting my time trying to do that of course and it could work out cheaper and easier just starting with the Arducopter in the first place.
That's exactly what I did and its a great idea. I started with a Gaui 330X. Its easy to assemble, calibrate and fly and its very crash-resistant. I then bought a FY90Q controller and added it to the Gaui, its a slightly better IMU unit and it teaches you to customise.
Thereafter a few MKs and an ArduCopter. I'm currently building two 'generic' setups - a quad and a hexa, both based on xAircraft frames, but able to accommodate a range of autopilots: Ardupilot, FY91Q, HoverFly or DJI. This enables you to learn a range of platforms and compare functionality and ease of use. Ardupilot remains the most cost-effective relative to range of functionality, by a long margin.
One thing I would not do, though, is to put the Ardupilot into the Gaui 330 frame, unless you use the 500 frame upgrade. You will not only have to tweak the stabilise settings but you will also lack payload capacity. Rather get a generic frame that has payload capacity, preferably in carbon fibre.
If you start out with the Arducopter, you may struggle, because its not 'plug and fly' like the Gaui. It trades off ease of use and configuration for advanced functionality.
Also decide what your ultimate goal is, if you need an autonomous platform for aerial photo and mapping, you need payload capacity and autonomous features (Ardupilot, DJI or MK - although the latter is range-restricted). If you only want it for FPV you can get by with a Gaui 500 frame, which has a nice add-on bay for FPV camera and transmitter.
Thanks for that informative reply!
As I guessed, an ARTF is a good way to start and I can look into autopiloting it at a later stage if I start with a big enough airframe.
I'm only interested as a hobby. I have a couple of electric heli's and a gas heli and I'd like to try a quad rotor out.
I have an embedded software background so dealing with PID tuning etc is no problem and I might look at doing my own autopilot too.
I just want to be sure I don't buy anything that ends up as a dead end, either due to being too small or too proprietory.
I would certainly like to have a play with onboard cameras, still and video but this does not mean it has to be onboard an autopiloted vehicle of course.
I have worked with cctv camera systems, both optical and thermal, and I have a variety of cameras around I could use.
I have a load of Sony CCTV camera modules with zoom lenses - based on older camcorder tech which I'd like to use. They are pretty high quality (but not HD) and have 18x zooms, auto focus, day/night capability, and built in anti-shake. They are not that light at 220gms, but I could reduce that quite a bit by trimming out the chassis - which is sheet steel.
I see that the Arducopter uses 4x850kv motors, but I'm not sure what the payload limit is but I read that around 500gms is a good upper limit. Not sure what the flight time would be in this case.
The Gaui 330 can certainly carry 500gms but with limited flight time. I read that it would carry 350gms for 7mins.
What I meant by 'payload capacity' is not just the power of the quad but also how ready the frame design is for a camera gimbal. For AP you really need to look at isolating the camera from vibration and ensuring that it is protected by the legs of the quad. Some frames are good for this, others not so.
Also, I've no doubt you will be competent in tuning your quad, but starting out with a tuning issue can be discouraging. When I build a new quad frame I put a basic controller - a FY90Q - into it first to ensure it flies OK before adding a more sophisticated autopilot. There are just so many parameters and potential issues, such as solder dry joints to driver issues on your PC. One of the weirdest issues I had was one laptop refusing to reset an EPROM, I assumed the board was fried. Then I tried another laptop and it worked. So much trial and error that you want to take it slowly.
Another thing to be careful of is interference from your various wireless items. I had a crash that I could not explain, till I monitored the feedback in MP and finally worked out that my XBee was causing continual mini brownouts to my receiver. Why? The antenna from the receiver was too close to the XBee's antenna. Be sure to isolate your wireless devices and avoid using high powered FPV transmitters until you know that lower powered ones work. Use low pass filters to avoid harmonic interference.