Thank you Zex, that is a valuable clue!

Failsafe was enabled on rx, and tested functional multiple times prior to and after these incidents.

See this post for details:
http://www.helifreak.com/showpost.php?p ... stcount=22

From the Spirit's perspective, the rx must have restarted, or had a failed connection between FBL and REX.
From my Jeti transmitters perspective at the same exact time, the rx must have been out of range or restarted.

Considering these two events occurred at the same exact time, the one explanation that explains both behaviors is the REX restarting.

Sending this REX back today.

So your failsafe is working properly, when enabled in your receiver?
I mean can Spirit see the RX signal loss?

_________________
Spirit System developer

Yes, failsafe was working correct on bench. Spirit did not lose signal with REX in testing. When powered up on bench and running full headspeed (no blades), I turned the radio. Swash and motor held until the failsafe delay expired, then cyclic went to 0 pitch and motor turned off.

I just got my new REX7 yesterday and installed / wired it up this morning.

Also, my Goblin was grounded during the 2 incidents (confirmed with continuity between tail and motor).
I was using the EX-Bus wire provided with the Spirit between REX and Spirit before and it tested good as well.

Lots of playing with the setup and testing tonight with new REX7. Configured as EX Bus again and at 11ms.

My throttle is connected to the Spirit AUX port like before. With the heli powered and at full headspeed, radio off initiates the Spirit failsafe (motor off) when signal is lost. This occurs with failsafe enabled or disabled on the REX.

I was not able to recreate what I though I remembered with failsafe testing on the previous REX and Spirit not seeing a signal loss. My guess is that I was looking at a previous log instead of the active log when I tested before and saw no signal loss.

Thank you very much for the info.

So does it mean that it is working correctly and you are not able to recreate it with new receiver?

_________________
Spirit System developer

I have not been able to recreate the event outside the two incidents.

That isn't very reassuring for sure. One of my theories is that there was something loose (in the REX7?) that caused problems during high G's.

Everything on my heli was, and is again very secure. Nothing is loose or can get loose. All connections are shoe-gooed (outside connector of course). All wires are wrapped, secured and protected from sharp or rough edges.

I plan to do more field testing today and hopefully fly it without any excitement.
Fingers crossed

Flight tested my Goblin with the new REX7 today.

4 aggressive flights in on the Goblin using the new REX7, EX Bus & 100Hz transmit frequency today with no problems!!!
Flew it out to the extents of my vision too, with great antenna & signal quality recorded in telemetry.
Whoo hoooo!

Looking like I had a bad receiver.

Thanks to the Spirit & Jeti support teams for helping me explore the possibilities and develop a better understanding of how Jeti & Spirit work together in the process!

Perfect, thank you very much for the info.

I hope it will be only good now.

_________________
Spirit System developer

Should have knocked on wood.

Problem re-occurred with a violent crash and burn. Thorough loss of my new Goblin 700 and the 14S Revo pack in it.

Replay of first two events. I activated hold immediately this time but did not regain control.
I reviewed all telemetry data from my DS-16, the Spirit log and the Kosmik log (amazing the SD card wasn’t melted). No errors. Nothing to even hint at an explanation of the event.



I completed tuning Spirit and the heli was flying great. Tail tuned perfect, cyclic exactly where I wanted it, rescue working perfect, done. Ready to fly it like I want to, hard.

Last seconds of flight before event, I was doing smooth piro tic-tocs, transitioning into hard aileron tic-tocs when the heli began un-commanded full + collective and full cyclic. Interesting that it was rotating back elevator to left aileron into the ground. It was not just locked into what I had told it last (left/right aileron with mild collective; not full). I hit hold the instant it was out of my control and did not get control back prior to impact like the previous two incidents. If I had been higher and had a little more time, I may have regained control like the previous 2 incidents. No way to save it without control... pretty scary really.

I have reviewed all telemetry data from my DS-16, the Spirit log and the Kosmik log (amazing the SD card wasn’t melted). Nothing to explain the event at all.
- There was no loss of signal between transmitter and receiver, or receiver and Spirit.
- Jeti: Signal quality was 100% at the beginning of the event.
- Spirit: no signal loss until during fire.
- Jeti telemetry highlites during event:
- PWM 100%
- Rx volts dropped to 7.74v
- A1=8, A2=8, Q=100%
- Spirit vibes maxed at 24% at moment of impact
- UBat dropped to 50.4v (3.6v/cell in 14S)
- IBEC amps was 2.5amps during event
- Temp BEC was 122f
- Temp FET was 111f
- REX failsafe was disabled
- REX port E2 connected to Spirit (EX Bus)
- REX port EXT connected to JLog 2.6GW (red wire removed)

So what is left that can possibly influence the flight in this manner (un-commanded full collective and cyclic)?
- Transmitter? I created a new model from scratch after the first two events. No problems with all my other helis.
- Receiver? This is a different REX7 than in the first 2 events. Unlikely 2 devices have same physical defect. Same firmware and it is v1.0. Some defect in firmware responsible?
- Spirit FBL? This Spirit is new and has only been used on this heli.
No errors in log to indicate any problems, but that isn’t really conclusive either.
I reset the Spirit and created a completely new setup from scratch after the first 2 events.
- Cyclic servos (BK8002HV)? If a servo mis-behaved, the Spirit would have tried to keep the disc stable. No way it would have allowed full cyclic with 1 servo going crazy.
- Power to system? Provided by 30amp Kosmik BEC from Master to Spirit and slave to REX. Complimented by TP250mah 2S lipo by 2 servo leads to REX servo ports functioning as backup and buffer. No evidence of any voltage fluctuations (high or low) from REX telemetry data or Kosmik log.
- Static hit? Continuity between tail and motor confirmed for first 2 events. Added an additional grounding wire between boom and frame after first 2 events. No possibility this was static related.

Out of this list of things that could possibly influence/cause this problem, there are only 2 things not eliminated.
- defect in REX firmware
- physical defect in Spirit FBL

New to me with this heli is the use of EX Bus and the configuration of the Spirit Jeti integration. I have UDI configured on all of my other helis.

Very frustrating to lose my heli so completely out of my control with all this data and no leads at all.

Ideas?

Oh, that is very sad. I have no words.

The best what we can do is to find real source of the problem.

It seems, that the signal was not lost at all during the event which excluding receiver and very likely also the connection between receiver and the unit. So there is virtually no reason from the unit to put full collective or to do any unexpected movement.

I can only say, that there are some customers with BK/RJX/TGY servos (but servos were not the same) that are experienced lot of strange issues because of servos were faulty from the manufacturing process - mostly signal wire was cutted by the aluminium box which caused random noise in the lead. Servo (mostly only one) then jittered or does not held its position as it was without power.

You can find lot of videos on youtube, for example here:
https://www.youtube.com/watch?v=_t9B3HqISCY
https://www.youtube.com/watch?v=fiLte2MST2Q

This issue happened also to our customers and is causing mostly bouncing during hard manouvers as it is short therm. Your servo is very different so it is possible it can't happen there, but could be possibility.

From our side, we can offer you carefull examination of the unit and testing it in the flight to be absolutely sure that it is good.
From my point of view I can't find one reason so the unit could behave like this. It can't absolutely add collective on its own, except when you activate the rescue. If there is any problem with loosing the signal, then it must be visible even during the time you were able to control it (as there must be frame losses all the time).

We are testing with REX6 receivers for long time without any problem even when it was in the development.
I dont think it could be caused by transmitter settings.

_________________
Spirit System developer