Swash 'sticks' over and tilts on it's own

[RC_Nut]

Can anyone help me or has anyone seen anything similar? Last week i had a tip-over on spoolup on my Kraken, cost me main and tail blades and a new boom

I checked the model over and the cause of the tip-over is the swash is not 'levelling' after an aileron or pitch input, plus if left with no input the pitch (front left) servo gradually 'creeps' down until the arm is at full downward travel.

Another strange thing is that I've had this before. My goblin Black Thunder had the same issue a year ago. There are absolutely no parts in common between the two helis, not even the transmitter is the same. Only thing in common is they are both using Spirit's one a 'Pro' and the other a Spirit2. It may be significant that both problems showed up in cold weather at about this time of year. I didnt get any re-occurrence during the warmer months.

Last year I put the Black Thunder issue down to a faulty gimbal on my old Spektrum Tx, but I've changed to a Jeti now, so that is 100% ruled out.

Here's the video of the Kraken problem:




Here is the Black Thunder thread from a year ago: viewtopic.php?f=24&t=2945

And the video i shot of the Black Thunder a year ago:




For info the Spec of the Kraken is:
Spirit2 with latest firmware
Kontronik Kosmik 200 (BEC at 8.0V)
Kontronik Pyro 750
BK servos all round

[ZeXx86]

Hello,

is there any fan mounted at the helicopter?

Also it might be enough when there is a wind and the model is moving in some way.

[RC_Nut]

There is no fan on the Kraken. The ESC is a Kontronik Kosmik 200 Kool, it has no fan fitted. This was something i eliminated when I had the problem the last time on the Black Thunder, that ESC did have a fan but I disconnected it and that made no difference at all. It was flat calm on the day I took the video so wind was certainly not the problem. I also took great care to not disturb the heli during FBL initialisation as i thought that may be part of the problem (it wasn't).

The two helis are totally different setups, nothing in common between them other than the FBL. Even the flying site and the transmitter were different. The only common external factor was that it was quite cold, 4 DegC or less.

My next step unless I hear other suggestions is to do a 'return to factory defaults' on the FBL and make a clean setup from scratch. It's grasping at straws but i cant think of anything else to try.

[ZeXx86]

Temperature has impact on this for sure. I am afraid but factory defaults will not help.
We could possibly send you a firmware with a small modification that could help or at least we could find where the problem is/isn't.

[RC_Nut]

Yes, I would be very happy to test a modified firmware.

[RC_Nut]

So a bit of an update, I havent received a trial firmware yet, so I've been trying to narrow down the cause with a few experiments.

I tried running the FBL off a 2S LiPo rather than the Kosmic BEC to eliminate the possibility that electrical noise from the BEC was the cause. This made no difference. I did notice that I could not replicate the problem 'on the bench' at home, even when the temperature in my workshop was pretty cold. I also noticed that even when the swash is responding normally I can make it 'stick' in position if a gently 'fidget' the heli side to side with my hand. This got me thinking, and I've formed a theory.

Tomas if you could maybe confirm if I'm on the right lines here:

I think that the way the FBL programming works is -

Normally when all is operating correctly and the heli is stationary, the gyro senses that there is no movement so it 'knows' that the heli is static and not flying. In this condition the programming automatically returns the swash to centre following any no stick input. The purpose of this feature I think is to prevent tip-over on spoolup as would happen if the swash stayed tilted when you spooled (which is what happened to me). However, as soon as movement is detected this 'self centring' feature is turned off and the swash then 'stays put' in commanded position (according to stick and gyro inputs) which is what's required for normal flying.

So the problem in my case appears to be that in cold conditions this 'swash centring' feature is not being activated. I think that the cold might effect the gyros so that they detect movement when there isn't any. This 'false movement detection' makes the FBL go into 'flying mode' so the swash doesnt return to centre but instead stays in the last commanded position.

So Tomas, am I close?

[ZeXx86]

You are perfectly right in all aspects.
I am sorry for delayed reply, tomorrow we will send the firmware finally. I hope it will help you.

[ZeXx86]

Hello,

here is the firmware and software:
http://spirit-system.com/dl/fw/beta/spi ... -beta2.4df
http://spirit-system.com/dl/settings/be ... -win32.exe

No need to change anything after update. Temperature comepnsation has been improved for Spirit 2.

[RC_Nut]

Brilliant! Thank you. I will test it this weekend

[RC_Nut]

I got three flights in today using the new firmware. The Spirit2 behaved itself perfectly, no sign of the previous problems. To be 100% sure that it's fixed I need to try it on a colder day but so far so good

Does this software also improve temperature compensation on the Spirit Pro?.. I ask because I've previously seen the same problem on one of my Pros (see second video above)