Hoverfly Scary descents..

[RotorFilms]

Hi All,

I'm sorry if this question has been asked before, but I need to ask it again, is anybody having some serious uncontrollable descent problems with the Hoverfly Pro FC? I'm onnestly not sure where the problem is, if it is with my setup or something else. I have a quick couple of questions, if anybody can give me some clarity then I would really appreciate it.

1. Would the fact that it is a X quad have anything to do with the problem? Would more motors be better?
2. If my gain is low would this add to the problem?
3. Descending in Auto-Level mode? I found the quad is more stable descending in manual mode than AL mode.
4. Did version 4 of the firmware make a big difference to the speed of descent?

Any help regarding this would be greatly appreciated? Almost lost my quad today due to very unstable descent...

Andre

 

[Efliernz]

Hi Andre

I can help a little. I have HFP quad and octos and I have had trouble getting my octo down without the wobbles. This is what I found through testing last night (cut n pasted from the RCG HF forum...):

--------------------------------
My diy octo has just done it's first commercial stills shoot (yes I wore my HF t-shirt... under my fluro vest) and weather was perfect. No wind, little traffic. I have been really careful on the descents in case it became a falling unstoppable rock. I only have 10 batteries through it so far.

At the end of the last hop, I was descending (in AL) a bit quicker than before (but still slow compared to all my other flying) and at 30' it started oscillating forward/backward and sideways to about 30 deg bank - at approx 0.5 second oscillations. I added power to arrest the descent (which it slowed) but it kept oscillating. I dis-engaged AL but it still didn't stop the rocking. I reduced power and landed it (on the road) with only a tip-backwards without damage ('cause I'm good!).

------------------------
and after some testing last night...
Well... I got home, moved the battery off the u/c frame to just behind the dome on top of the rear arms. Checked cog - perfect. The removal of 530g off the u/c meant that the rubber isolators were not struggling and everything felt firm. Off for a series of climbs and descents in zero-wind conditions.

After two batteries and a heap of hops until it was dark, I now have nearly doubled my original "end points" on the gain channel (haven't looked at the HF figures on the client yet) and I suspect I may go higher still.
It flies so much better than it did. Locked in the way you expect it to be I can now descend twice as fast without the shakes.

--------------

Your other questions.
My quad and octo descend about the same in manual and AL. They both rock - but they both still have a small wobble when descending too fast. Remember the gains for manual and AL are different and need to be tested and set up individually (without worring about it being different to each other). That was my fault before last night.

My quad is a X-quad and it flies fantastic. I am flying Firmware 3.2.560 and it is great - a friend has changed to V4 and is having some yaw and stability issues. I suspect that V4 had quite a few changes that HF haven't really explained that well. They have changed the throttle response, restricting how much throttle you can use (ensuring you always have enough power to yaw with good control / motor corrections. I am sticking to V3.2.560 until I see some favourable results from lots of users on V4.

Pete

 

[RotorFilms]

Hi Pete,

Thank you very much for the reply.. I just quickly want to get some clarity on a couple of things in your reply..

You said during testing you pretty much doubled your end points on your gain, does this mean you pretty much doubled the gain values you started with? (sorry still trying to get my feet with all the lingo and jargon)

What does u/c stand for? is it under carriage?

I'm thinking of reverting back to a previous firmware as well as I can never remember the descents being quite as scary as they were this afternoon, very similar to what you explained with the rocking fore/aft and sideways at about 30deg. At one stage I thought it was going to flip.. also stopped the oscillation by increasing throttle.

I'm going to also go out tomorrow to do a number of tests and to do some adjustments to see what works for me.. I think I must increase my gain though as currently it might be a tad on the low side.

Andre

 

[Efliernz]

My stats of the octo:

860mm motor-motor
RCTimer 2836/11 750Kv
APC 10x4.7
4S4000 battery
3550 grams all-up.
Motorcalc has it at 2.05:1 at the reduced / used battery voltage.
Hover is at 36A, climbs well at 42-44A. It is a stills ship, not a rocket. Climbs really well at 55A
HFP on 3.2.560
Using 15.0.42 client, I had my Main Gains set to 63 Normal, 61 AL.


My gains mean nothing compared to another model! I haven't looked at what my gain is on the client but it is probably over 100... which is ok for THIS model.
Keep adding gain (in small increments) until the model in hover is starting to have a small shake... that is where it is over-correcting. Then back the gain off a few numbers. Repeat in AL. That is how I did my quad and it is so locked it!

U/c is undercarriage. I'd show you my machine but my crappy work pc won't let me attach photos. You can see my models at my youtube site "efliernz"

Just a warning. Hoverfly's IT host have had a meltdown and you can't get firmware off their site at the moment... but they are working on it.

Pete

 

[RotorFilms]

Thank you for all the information, really appreciate it. I agree that gains can not be compared between any two MR's. I'm just waiting for another suitable weather window then I'm going to go out and do some testing.. I honestly think I might have gone a bit low on my gains and that is where my problem is.

Thanks for the heads up on the Hoverfly IT host, I was not aware of that..

 

[Efliernz]

The IT issue is sorted now.

When I flew my machine with the gain too low, it flew "ok" but felt lose and sloppy (hard to explain). It is a credit to the HFP board that even a poorly set up octo still went to 350' safely! It was locked in better in AL but it wasn't as nice to fly as my quad (which I had set the gain for properly).
Seriously... upping the gain makes my light-weigh octo handle like my quad Tight and locked in. It doesn't wobble on takeoff, it just climbs level without any wobble.

Pete

 

[Wingspinner]

Although there may be tuning issues making it worse helicopters of any type have a problem descending quickly unless there is sufficient forward motion to ensure the blades aren't descending into their own prop wash. If you descend too fast without enough forward motion the copter enters "ring vortex state" which is just a fancy name for descending into the prop wash. It's also known as "descending with power". This can result in a crash because in some severe cases all control of the copter is lost and the only way to recover - if you have sufficient altitude - is to cut the throttle and let the copter fall into "clean" air then re-engage the throttle and stop the descent.

Scary stuff in a real helicopter....