Spirit System

Hi all;

Hoping you can help. I have a Spirit Pro (FW 2.5.1) mounted in my scale Mi-24 Hind, 600 size. Helicopter specs:

• Thunder Tiger Mi-24 Hind Fuselage, approx 14.0 lbs AUW
• Audacity Pantera P6E mechanics (with Stage II clutch)
• Scorpion HKIII 4035 450 Kv motor on 12S (2x 6S) through a Jeti Mezon 90 Opto ESC
• Century Diamond 5 Blade head, 600mm Funkey blades (symetric profile)
• 3 Bladed Miniature Aircraft tail, belt driven through custom intermediate gear box.
• Control is Jeti DS16 through REX12 reciever. Spirit connected via EX Bus.

The helicopter last flew in July with the below settings (in a test rig, not the fuselage), and was pretty stable (changes today in bold). Today was the maiden flight in the fuselage... Almost ended in disaster. I'm finding that the Spirit is barely stabilizing the helicopter at all. Roll & pitch are very sensitive. Tail is rock solid. Hoping that you all can help.

Video of today's flight is here: https://youtu.be/v8SZaC90Kh8
I really apologize for the shakiness, it was a GoPro on my head - not the most stable of positions! (Camera wife was at work!)

Spirit Settings:
General:

• Horizontal orientation
• CCPM 120° (reversed)
• Jeti EX Bus
• Flight Style: 4 (was 5 during first part of flight, reduced to 4 after middle landing)

Diagnostic:

• Tail Gyro Gain: Switchable between 56 & 65%
• No Bank switching enabled

Servos

• Cyclic: ProModeler HV BL 1520/333
• Tail: Outrage Torq tail servo (model escapes me) 760/560

Limits:

• Pitch Range: 255 (pitch range limited in radio to -5° / +10°)
• Cyclic Ring: Unchecked
• Aileron/Elevator Range: 180(~14°)
• Rudder end points: L 121 / R 55

Sensor:

• Cyclic Gain: 70 (was 65 during first part of flight, increased to 70 after middle landing)
• Rudder Gain: In Tx, switchable 56 / 65%
• Rudder Common Gain: 1.00x
• Cyclic Rotation Speed: 5
• Rudder Rotation Speed: 5

Stabi:

• Stabilization Mode: Normal, no rescue (Stabilization was only on at the very beginning, after the first attempt to roll over I shut it off from the TX)
• Sticks Priority: 6
• Direction Control Rate: 3 (I may have changed this midway, I don't honestly remember. It might have been 5 prior.)

Advanced:

• Geometry: 104
• Elevator Filter: 2
• Rudder delay: 5
• Revomix: 0
• Cyclic Feed forward: 5
• Rudder dynamic: 4
• Piro Consistency: 160

Advanced -> Expert:

• Stick Deadband: 4
• Elevator Pitchup Compensation: 0
• Cyclic Phase: -56 (there is no mechanical phasing to ensure links are vertical / not binding)
• Pitch Pump booster: Unchecked
• Signal Processing Extended: Unchecked

Here's a picture of the setup in the helicopter:


I'm really hoping that someone can help here. These fuselages are next to impossible to find, and I would be heartbroken if I stuffed it in.
This is a matter of national Czech pride... Because this is what he'll look like when done!


Yup... That's 7353, who is part of the Czech Air Force, in Namest as part of 221.lbvr. He's won best livery at the international Tiger Meet.

Thanks for the help!

cheers,

ed

Does no-one have any thoughts on this?

As an update - on the bench, the swash is responding appropriately to compensate for body movements (I know - a good bench check is not a guarantee of everything working well in the air). The log shows only "Good Health" reports. Vibrations from the flight in question were mostly in the 2-4 range, with rotor start-up / shutdown spikes to 13.

As an FYI, this is the same unit that was involved in a static hit, as documented here: http://spirit-system.com/phpBB3/viewtop ... =20&t=2829

Hello,

from listed parameters I can't see any reason why it would not work properly. Changes you have made will not make basically no difference in stability.

Usually when adding fuselages a mechanical issues appear which are causing stress on the servos. Please make sure that all servos are moving completely freely.
Alternatively the problem could be in a wrong swashplate phasing - maybe something changed mechanically by mistake?
Lastly, please measure vibrations by Spectrum Analysis - for example by In-Flight Vibration measurement - at least at Y axis is enough. This will tell us much more.

There shouldn't be anything else wrong.

You can always attach your settings in a 4DS file. No need to rewrite all parameters as all are contained in the file and can be viewed online.

I can confirm that:

All servos are moving freely, no binding, full range available.
Helicopter balances neutrally Fore/Aft & lateral (tested by rotor head hang & side tilt methods) - no excessive demands on servos from out of balance condition.
Swashplate phasing is correct - no mechanical or electronic alterations.

I'll try to do a live Spectrum Analysis run - though as I mentioned above I do log the Spirit vibration level in the Jeti flight log, and it was average of 2-4 during flight, with a couple of 13 spikes during rotor startup.

Have you had a chance to look at the video? The rotor disk is not stabilized, and seems to be osculating. My recollection is that it was more pronounced on roll than pitch. That could be a) just my memory of it, b) related to the fact it's a long, narrow helicopter - more stabilizing mass in fore /aft plane.

Remember, this unit took the shut down level static hit a year or 2 ago (though it was flying fine in the test rig up until July). I'm concerned that my roll axis MEMS may have failed.
Alternatively - would a cyclic gain of 70 produce this slow oscillation on the rotor disk?

I did realize I could attach the 4DS file - after I wrote it all out and posted it! (Smacks head)

I'll try and get a Y Axis spectrum measurement in flight and post it back.

Thank you

ed

Hi Tomas;

I've run the Spectrum Analyzer on the bench, with no main blades on.
Main Rotor RPM is 1240 rpm (avg, range 1234-1249, 5 blade main)(20.66 Hz)
Main rotor to Motor ratio: 1:8.7 (Motor runs at 10,788 rpm)(179.8 Hz)
Main Rotor to Tail Rotor Ratio: 1:5.25 (Tail runs at 6,510 rpm, 3 blade tail)(108.5 Hz)

Vibration X No blade 1240 rpm


Vibration Y No blade 1240 rpm


Vibration Z No blade 1240 rpm


Vibration In Flight No blades 1240 rpm


When Mother Nature works with me, I'll try to get a in flight log.

cheers & thanks,

ed

Just to note - the above spectrum graphs were WITH tail blades on (not what the manual procedure calls for); I'll redo tonight without them on.

Aircraft was also "hard mounted", not on soft surface as manual calls for.

ed

I am rather convinced the gyro is not damaged in any way. It would be the first time it happened after a static hit and also the behavior would be different.

From your plot I can see very high spike from the motor at one axis. This could be possible reason, because it should be not higher than half of the plot to say motor is OK.

If there was a crash motor shaft and a bearings could be always damaged in some way. Gyros are usually the last thing that could damage since there is no movable part. But whatever is moving at the model has very high probability to cause issues at any point. So I always recommend to focus just on the things that changed, not that are same.

Any settings in the unit except Geometry 6° value will not cause such instability. Since your Geometry value is same I think there is no reason to play with the settings.
From our experience the problems are unfortunately nearly always in the mechanics if something started to work incorrectly "on its own".

Hi Tomas;

Yeah, some thinks showed up on those analyzer runs. I've torn the head apart, and am in the process of "blueprinting" it (matching all components up to ensure a normalized mass across the rotor head). There were some mass imbalances present which I've corrected.

The swash driver is the most obvious culprit on Main Rotor vibrations. I'm looking at ways of correcting that (as much as is possible).

Main, tail rotor shafts, clutch and motor can have all been dial indicated and shown to be running true. I'll be looking at the tail system next - wondering if there is a bad bearing in there.

As a question, if I move to a slightly softer mounting tape (gel type or foam type that would have better vibration absorbing properties) am I looking at a significant degradation in the Spirit Pro's ability to operate?

Thanks,
ed