Yesterday, a new flier in the city (also a beginner) was referred to me by LHS personnel for help with setting up his brand new Align 55E 3G heli. It was assembled by the owner, only the radio and FBL setup were pending.  Since I had setup 3G systems before, I though this would take 30 minutes for setup and couple of test hovers outside the house.

The Align 3G system setup is a well documented step-by-step procedure, and we were through it in no time.  After both the tail and the flybarless systems setup, we turned off receiver power to the heli (the main power was not connected at all, a separate RX battery was used).  When we turned on the power again (still on separate RX battery), the 3G systems jumped around, and finally slopped to maximum negative pitch.  We were perplexed, as I was seeing this for the first time. I knew there was a rare possibility of FBL system not initializing if the heli was moved etc, so we tried removing and restoring RX power about 5 times, and each time the swash did the same thing, and the FBL system never initialized properly. 

So we decided to redo the FBL setup once again.  This time, even the FBL setup itself would not finish.  After the aileron direction and movement were properly fed in and the SET button clicked for the last time, the swash did some chicken dance, and never went level.  After a few tries, we were able to finish the FBL setup again, but the system refused to initialize when we wanted to do test flights.

So it was time to do a bit of research... and that lead to threads like these :

http://helifreak.com/showthread.php?p=1811514
http://www.rcheliaddict.co.uk/align-fl760-flybarless-system/51044-need-help-3g-system-will-not-initialized-most-time-when-power-up.html
http://rc.runryder.com/helicopter/t578569p1/

There were many references to 3G systems having issues with some Spektrum receivers (solution seemed to firmware upgrades), to the sensor mod for 3G sensor to replacement of the unit. We tried remounting the sensor, to see if it was a vibration issue.  We also tried with and without even the rubber bushings on the landing skids ! Obviously the owner was disheartened, this was his first heli after flying a small coax for some time, and lots of sim time... all pumped up and ready to fly a proper heli

So we discussed trying to get another 3G controller for testing, but continued to tinker with the system.  It was then noticed that chances for successful initialization increased significantly, if we initialized with the pitch (ie throttle stick) at the middle.  When we tried at zero throttle (swash at maximum negative pitch position), it would never initialize.

That led to the realization for me that the difference between the two positions is that in one case, the initialization process would start at the center pitch position, and stay there if the throttle stick is that middle, while it would start at the middle and drop down to the negative pitch position if the stick was at the bottom. In short, the difference is a significant movement of the swash servos during the initialization servos.  That led us to test the initialization with one servo disconnected at a time, and voila, when one of the pitch servos (brand new) was disconnected, it would initialize every time 

Just to rule out the possibility, we tried multiple receiver power setups, including 5v and 6v lipo+BEC based setups also.

I hope this thread is helpful to people (like in the links above), who are seeing this problem. One more data point to check for them is to try removing servos one by one.

A replacement servo is being sourced, even though this one seems to work well individually. It just seems to drain a lot of amps, and pull down the power system. It feels good that we have a better handle on the problem, even though it took close to 3 hours !

I had a problem with my 550E 3G Setup also. It would go through the entire process and everything would work. Switch off and power on again.. the rudder servo would not work. The rest would be working. I am using a Walkera 2801 Pro. It took two days to figure out that the Rx needs to be paired with a fixed ID number in the Tx. While binding/ pairing with a random code the 3 g system would initialise before receiving a signal from the rudder channel.. It all works well now  .