SK360 on my T-Rex 450

[int2str]

I took the plunge and installed the SK360 on my T-Rex 450S. I'm not completely in love with the head conversion yet, but it'll do for now.

Installation of the SK360 was quite easy and the software works quite well. I tweaked all the servo reversing and gyro orientation settings to suit and off I went.

The first flights went quite nicely. It took off smoothly and flies out of the box (settings based on the "Basic 3D" profile). I turned down the hiller gain a bit using the dial (ended up at 43%). and it flies pretty nicely.

The heli seems to be a lot faster now somehow. The sound has changed a bit and it even seems to handle wind quite a bit better now! :) I'm definitely liking what I can see so far.

Now I have to tune my setup to get rid of a few issues:


1) Tail vibration

The vertical tail fin is visibly vibrating quite a bit. I first thought it was the tail gyro, but it's not. I think it maybe vibration caused by the SK360 itself. I think it's "shaking" the heli. But I'm not sure. Just to make sure the heli is mechanically in 100% shape I'll replace my main shaft bearings and I even bought a new tail shaft just in case (even so I don't think my tail shaft is bent at all).

I initially thought turning down the hiller gain may fix it, but just turning it down a little bit didn't seem to make a difference.


2) Piro's "tumbling"

If I piro the heli now, instead of spinning in-place, it'll "wobble" a bit or "tumble". The center of gravity is right at the main shaft as far as I can tell. Phasing is as close to 0deg as I can get it and the SK360 should be mounted correctly too. Though it does seem to slope a little towards the side with the servo connectors.

What could cause tumbling like this?


3) Elevator axis roll in FFF on pitch changes

When I fly forward (nose down) pretty fast and then change pitch, the heli will roll around the elevator axis. As in, if I fly nose down and give more positive pitch, the nose comes up and the heli slows down. Conversely, if I move towards less pitch, the nose points down and the heli speeds up.


4) Piro stop causes heli to tilt

I have my 450 setup for insane piro speed. Now when I do a fast piro (or a few :D) and then stop, the heli "falls over sideways". Would this be expected on this kind of system or is this something that can be tuned out?

Thanks!

[ImNix]

That heli was lookin pretty baaaadd today andre. im very jealous. cant wait to see it fly!

[Skookum]

> 1) Tail vibration
"The vertical tail fin is visibly vibrating quite a bit..initially thought turning down the hiller gain may fix it, but just turning it down a little bit didn't seem to make a difference."

Up and down oscillation, or actual vibrating?

>2) Piro's "tumbling"
>If I piro the heli now, instead of spinning in-place, it'll "wobble" a bit or "tumble".
> What could cause tumbling like this?

It's because a flybar is in the spinning reference frame of the blades, and the gyro + heli body + swash is not. So this is the hardest thing for a flybarless conversion to handle.

While in normal flight, the gyro is compensating for wind, tail drag, maybe the swash wasn't totally level or there is some aileron/elevator/pitch interaction. A bunch of things it hides from you.

But when you piro, some of those factors spin with the body of the heli, and some do not. Also, the servos aren't infinitely fast and the linkages are not perfectly stiff. So even if the gyro knew everything about pyro rate and wind and the swash was told to "swirl" perfectly, its going to have a bigger problem: the swash will lag and so be out of phase. 15 degrees of phase error seems to be where problems are noticed, and at 2 revs per second that's only about 20ms of servo+mechanical lag.

What can be done? Give it less to compensate for, faster swash response, and better natural stability. Good CG, fast digital servos, solid mechanics, and sometimes longer blades help. Remember that most of our blades and heads are not yet designed for flybarless - although that's in the works.


>3) Elevator axis roll in FFF on pitch changes

That's a new one - maybe it's interaction? You might want to check that while in setup mode, the swash is level at max, zero, and negative collective pitch.

[int2str]

> 1) Tail vibration
Up and down oscillation, or actual vibrating?

The vertical tail fin is vibrating side to side. The bottom tip to be specific. Very high frequency (it's pretty much a blur).

>3) Elevator axis roll in FFF on pitch changes

That's a new one - maybe it's interaction? You might want to check that while in setup mode, the swash is level at max, zero, and negative collective pitch.

I'll double check, but I didn't have any interactions last time I checked (Trueblood leveler).

[SpeedVision]

The vertical tail fin is vibrating side to side. The bottom tip to be specific. Very high frequency (it's pretty much a blur).


That's typically a balance issue in the head, from my experience. Bent main shaft, feathering shaft or blades out of balance could be the culprit...

[int2str]

Thanks.

Main shaft and feathering shaft are brand new.
I'll try balancing the blades.

[Skookum]

>The vertical tail fin is vibrating side to side. The bottom tip to be specific. Very high frequency (it's pretty much a blur).

For sure not the SK360 in that case, because it can only cause oscillation as fast as the servos can oscillate. In fact that high frequency vibration could be causing gyro issues.

>3) Elevator axis roll in FFF on pitch changes

After thinking about it this makes sense in terms of aerodynamics. The advancing blade on the left side is seeing more airspeed than the retreating blade, so increased pitch will cause nose-up. The gyro should be compensating, but isn't doing it enough or in time.

So it sounds like you need either higher hiller gain (which may be possible with less vibration), and/or a larger cyclic pitch range, maybe +--10 instead of +-8 degrees. Stiffer head damping also allows for higher gains (until the resulting vibration is too much).

[int2str]

I'm running an all metal head at this point with Trueblood dampeners. Not sure how I could get the head dampening stiffer. Additional shims, I think?

Asides from that, here are the changes I am making:

Vibration related
- New tail drive shaft
- New tail belt
- New main shaft bearings

SK360 related
- New mounting position

Other changes
- Went back to a 12T pinion (3550KV motor) for lower head-speed

Planned setup changes
- Fix gyro settings for new orientation
- Up the hiller gain until I see oscillation (not vibration) and then back off a bit
- Possibly increase swash mix settings some for better flip/roll rates

I'm not sure if I'll be able to fly before the weekend, but I can't wait to play with this some more.

[SpeedVision]

SK360 related
- New mounting position

Planned setup changes
- Fix gyro settings for new orientation


Hahahahaha.:lol:

That one cost me a set of Align Woodies!!

I went from bell/hillier pots facing the tail to facing heli left (looking from tail to nose).

Had to uncheck rotate 90 and had to check "flip aileron and flip elevator"... Just take your time and don't get too eager try and fly

[Skookum]

> Went back to a 12T pinion (3550KV motor) for lower head-speed

If you go to a low head speed, then you *for sure* want +-10 degrees of cyclic pitch, or you're in for some surprises with hard collective, especially in FF.

Also, Re wandering during fast piros, what helps are longer blades (335s?), higher headspeed, or just a larger heli. The smaller a heli the less natural stability; down at the 300 size for example a heli with standard blades is totally dependent on the cyclic gyro, while a 600 size heli can be hovered ok with no gyro assistance (though FF might be a mess).

[int2str]

New problem...

I've finished all the modifications I listed above. I also moved the SK360 from the back of the heli to directly under the main shaft. With the "front" label actually pointing towards the front now.

After re-orientating the SK360 and wiring everything back up, I made 100% sure that all the software settings are correct. If I tilt the helicopter right, the swash tilts left, if I tilt the helicopter right, the swash tilts left. All the servos move in the right directions as well - all seemed good.

Now when I try to take off, after a few seconds the heli wants to tip over hard to the right. I'm not touching the left stick at all during spool up, but as soon as it reaches a certain percentage, the heli just forces itself over hard right.

Is this noise from the main gear or something?

[SpeedVision]

Here's my end of the conversations I've had with Skookum regarding my nasty tipping issue, but to the LEFT. My SK is on top of my Mini Titan in the gyro spot... Since the front of my unit is facing the tail, I have "flip aileron" and "flip elevator" checked, rotate 90 is unchecked.

Also, please note that my unit FAILED the cold test and I choose to use it at my own risk... Whether the "cold problem" is the culprit or not, the SK was at 70 or so degrees when initialized... A replacement is on the way...

Here are some excerpts of the information I provided:

I wrote:

I’ve developed a nasty tipping issue just before takeoff. This, for obvious reasons doesn’t inspire confidence. I lived through the 25% hillier decay takeoff’s early on (without incident), so I don’t think habits or skills are much at fault. I do keep my thumb on cyclic because, sometimes, I can catch the tilt, but it really snaps over (always to the left) when it wants to. I had an earlier hypothesis that it could have been contributed to the slope of my driveway, but I have since eliminated that from consideration.I moved the Sk from inside the frame of my Mini Titan to gyro mount on top. I had a couple tenths of a millimeter gap to each side of the frame and I thought it was just too close. (Edit - no tipping problem with this setup, bell/hillier adjustment pots facing the tail, rotate 90 checked)I do get some high frequency vibes from the tail on the bench. In the air the convenient little auxiliary horizontal fin on top of the vertical tail fin isn’t showing any signs of vibrating. Since it’s plastic, it shows everything, when vibes are present. I did lower my control rates to 170 on the advanced tab because my Heli really seems to have a mind of its own in the air. I have aileron/elevator expo rates at 30% on my DX7 to also try to calm things down.I spool up with a linear pitch curve and a 0-35-65-85-100 throttle curve. Tip happens seems to happen just as positive pitch starts to kick in.I am trying to rush my takeoffs, which upsets the tail and makes things more tricky.Taking heli from inside (70 F) to outside (45 F). Had the heli sit outside for 5-10 minutes with no change.Don’t think I can go too much higher than 50 on hillier pot without getting oscillation.Phase is set per manual specifications.Once airborne (Edit: if I'm lucky) everything responds to stick inputs correctly. As noted above, really feels like a ball bearing on glass – all over the place.Swash is level with Trueblood tool. Blade tracking is spot on.Running HS65 MG Servos at 6.0 volts, .11.Out of fairness to Skookum I will summarize some of his ideas relative to the problem, but I'm not going to cut and paste his answers. He can expound and correct if necessary. Also, Skookum has been very helpful and quick to respond, so service has been outstanding!!! I am confident that these bugs will be corrected and I'll be able to spend more time flying and less time "tuning"...

Summary of Skookums comments:

Could be vibrations during spool up, make sure SK is stationary to establish a good "zero" initialization value. (My comment: Vibes are under control)Less likely possiblility is TX "center stick" is off causing the tilt. Increase dead band to 3% to test. (My comment: Didn't help)Tilting could be a problem linked to units that failed cold test. (My comment: This could be the problem and a flight tested unit is in route to me, so I should have feedback as early as this weekend if it arrives in time. Until then, we have to continue to consider other causes of the problems...)They are working on a firmware upgrade for spool-up and improved self-diagnostics...Anyway, if you've made it to the end of this novel, that's all I know right now. As soon as I have the new one installed and tested, I will report back ASAP!!!

Peace,
Michael

[hamslice]

i was also getting the kicks as you described but only to the back like pulling back on the elev. i found that when i reduced the hiller dial it would go away. the only thing i haven't got right yet is in fff when i give higher collective the heli will nose up and continue to do so so much that i can do a loop without touching the cyclic stick. i have been increasing the pitch gain a little each flight so i still might be low on that. when i do a hard climbout upright or inverted the heli climbs very straight. i think that i need a higher hiller setting, but if i do the heli will kick back on its own. i also have the heli vib free. i have spent more time concentrating on getting all the vibs to disappear than ever before.

[SpeedVision]

Interesting... What orientation is your SK? So far we have, gyro front facing forward - tips to the right. Gyro front facing tale - tips to the left...

My hillier is at 50% or less and bell at 30% or less. Early on, I was able to take off just fine with hillier too high, it woud just oscillate once airborne...

Only other notable change I made... I lowered my swash percentages to E=51, A=50 and C=32 and increased my servo travel percentage on the servo tab to 115% for each. This gave me the suggested pitches without having the swash percentages in the 60's...

[hamslice]

all of my swash %'s are at 65. the gyro is facing sideways so i had to check the rotate 90 button. i didnt know that the swash % should be less than 60. my hiller settings are way down. like 15 and 11 just to get it not to kick. i can go higher without the oscillations but it will kick randomly.

[int2str]

Interesting... What orientation is your SK? So far we have, gyro front facing forward - tips to the right. Gyro front facing tale - tips to the left...

Interesting!

And just like you, the SK360 worked fine for me rotated 90 deg...
I think I'll try to rotate it back.

[int2str]

in fff when i give higher collective the heli will nose up and continue to do so so much that i can do a loop without touching the cyclic stick.

Mine does the same thing. It's point "3" in the first post..

Or shall I say mine DID the same thing. Since I'm grounded until I can resolve the tipping issue... :(

[SpeedVision]

all of my swash %'s are at 65. the gyro is facing sideways so i had to check the rotate 90 button. i didnt know that the swash % should be less than 60. my hiller settings are way down. like 15 and 11 just to get it not to kick. i can go higher without the oscillations but it will kick randomly.

My goal was to get the Hillier high enough to get a stable hover... Right now my heli feels like a steel ball on glass and won't hold reasonably still in a hover, unlike when it was flybar'd

This is where I tread on "thin ice", so here's Skookum's comments. I will quote his reply since this is more technique and less "problem solving"... As always I reserve the right to remove this content if he wishes...

Skookum wrote:


"Yes, I'd increase the servo travel and then lower the swash values.

Swash values should ideally all be below 60 or so to prevent interaction at max collective. So if you set the servo travel to say 110%, you could then set the swash mixing to -60, 60, -46 and you would avoid interaction. In reality I've used 65% and never run into an obvious problem, but I mostly just sport-fly.


For the gains, I don't remember if you said what your Hiller gain is up to, or what your cyclic pitch range is. With 8 degrees of cyclic pitch each way, I'd expect you could get the hiller dial up to 60% or more, at which point it shouldn't wander. At +-10 deg cyclic pitch, 50% hiller would be the same effective gain.


I did notice that on the Advanced tab the Damping gains are higher, probably you pressed Copy From Dials at some point. What copy from dials does is show you the actual gains the gyro is using. Otherwise its mostly useful when you want to lock down the gains, so you'd go "copy from dials" then enable "disable gain dials". If you press it otherwise to check what's going on, I'd press "get setup" again after to undo what it changed.


Basically the gyro takes the ratio between the hiller gain dial, and the Base hiller gain for elevator, which is say 55/50 = 1.1. It would then use the hiller gain dial value for hiller elevator gain, but for hiller-aileron and both damping numbers, it would multiply the value in the table by 1.1. See kind of wordy to explain, maybe the manual needs a diagram or table there. The idea is that ratios don't need changing that often, so it makes for one-dial field tuning.


Anyways, back to the best gains. Probably the Damping gains are too high, because they are scaled by the hiller gain dial too, that is limiting the Hiller gain. So on the advanced tab, try setting Damping gains to 16 for elevator and 12 for aileron."
The biggest issue we are facing, in my opinion, is getting the best mechanical head setup and proper head geometry to allow full servo travel (without swash binding) to allow higher hillier gains... Until the aftermarket responds, we'll have to "wing it" and do the best we can. At least we'll learn A LOT in the meantime.

[fireup]

Interesting... What orientation is your SK? So far we have, gyro front facing forward - tips to the right. Gyro front facing tale - tips to the left...


For me it seems to tip away from the PC ^ label on the gyro (Gyro mounting with Pot. in the front makes heli tip toward the front left). I think the problem is from vibration and resunance. It seem it's very sensitive on how you mount the gyro. For me, it was better if you use spongier tape like to ones provided or two or three layer of white 3M foam tape. The gray 3M tape with red backing tape don't work very good.

[Finless]

Yes when testing my unit on Kevins heli this last weekend it would go full forward elevator at a certain RPM. He had it stuck down with the thin gray double sticky tape. We remounted it with 2 pieces of the 3M white foam stick tape and moved it back a bit away from the tail belt drive bearing and that solved it. I really don;t know if ti was vibration or noise from the bearing that was causing it but it was for sure one of the 2.

Bob

[Skookum]

Easy one first...

Hamslice>"in fff when i give higher collective the heli will nose up..."

That's almost certainly the gyro running out of cyclic pitch to apply. The fixes would be to increase your headspeed, or increase cyclic pitch range to +-10 degrees. You should also increase servo travel and get the swash mix #'s 60% or below to prevent interaction at full collective pitch. The manual's going to be changed to advise +-10 for cyclic pitch :-P

>i have spent more time concentrating on getting all the vibs to disappear than ever before.

Re vibs, I'm not sure it's that vibration sensitive. It does make a difference in how well it holds like for any gyro, but to get random kicks it usually takes quite bad high frequency vibration. Like out of track blades etc won't do it. Static build up from the tail boom can happen though, the cure is just to ground the boom to the frame or motor case.

I'm thinking now that for many the problems are mechanical/aerodynamic. I spent 12 hours (!!!) failing to convert a blade 400 yesterday, which tried to flip at take off, wandered and kicked, and couldn't stand the hiller dial being above 35%, with well-proven SK360 on it. I also had access to directly logged data to graph here. So I could confirm the gyro wasn't sensing or doing anything weird, but the rotor disk and swash was doing weird stuff. My theory now is that the blades for the 400 have a bad chordwise CG and/or the grips are too sloppy, the combo is aerodynamically unstable, and cannot be properly controlled by the servos. So the servos just "lose it".

That's not to say it's everybody's problem though. Int2str's issue started when he rotated the mounting 90 degrees, so it's possible there is a resonance effect happening there. Hard to explain otherwise.

The problem coming out at a certain rpm is another good clue. *All* of the sensors in the gyros on the market work using a tiny part is vibrating at above 10 khz, so it's easy to see how just the right high frequency could mess with them. Rotor vibration though tends to be at most a few hundred Hz; the khz range is more like nitros or gear mesh territory. Maybe one day they'll get a laser ring gyro etched into a chip...

>as soon as it reaches a certain percentage [during spool up], the heli just forces itself over hard right...

FYI the timing is probably due to the "spool up mode" timing out, which applies very high hiller decay to calm spool up. So the gyro either thinks its tilting left, or thinks you are applying right cyclic.

Re the "cold bug", the problem *can* occur at room temperature. The "cold test" was to force it out into the open. What happens with the flawed gyros is the sensor fails to start; if you wanted to pre-flight check to be totally sure its ok, just pick up the heli and tilt it left/right nose up/down, and see if the swash responds. If it responds then, it'll be fine later.

[SpeedVision]

Yes when testing my unit on Kevins heli this last weekend it would go full forward elevator at a certain RPM. He had it stuck down with the thin gray double sticky tape. We remounted it with 2 pieces of the 3M white foam stick tape and moved it back a bit away from the tail belt drive bearing and that solved it. I really don;t know if ti was vibration or noise from the bearing that was causing it but it was for sure one of the 2.

Bob

The fact that this is happening on such a wide variety of Heli's is what has me puzzled...:dontknow

Mine tips towards the adjustment pots.

When I first tested the SK my heli was setup like CRAPOLA and it didn't tip. Bad main shaft bearing, tracking was out by an inch, main blades were not properly balanced, tail blades weren't properly balanced, case of the SK was rubbing on the Titan frame (SK provided tape), blades were so loose that they would fold during spool up, it would do this :banana for a few seconds, then smooth out when they found their centers...

When I get my replacement (failed cold test), I'm going to stick it right where the other unit was, reload the same program and let her rip...

Good part about all this "beta testing", I won't have to buy mulch for the flowerbeds this year...I have a nice pile of red/white painted balsa to use...:smokin:

[int2str]

So what should I do?

Turn it back 90 degrees?
Move it further back on the heli?
Different mounting tape?

I tried holding the heli with one arm last night (I know, bad idea :P). I figured, maybe it's the aileron axis being reversed.

If I spool it up holding it in hand, beginning at a certain RPM, the swash plate visibly tilts to the right and then settles in. The gyro still reacts every motion from there and compensates correctly. But it always goes back to this slightly titled position (that pulls the heli over, if it was on the ground). It seems like the "resting position" is skewed.

I tried re-initializing the gyro a few times and it always consistently settles in about the same spot.

[int2str]

Mine tips towards the adjustment pots.

Same as what Fireup describes (away from the PC -> label is towards the pots) and same direction mine does.

When I first tested the SK my heli was setup like CRAPOLA and it didn't tip. Bad main shaft bearing,

Same with mine. I rushed to test it and had worn out main bearings in there and didn't spend much time optimizing the head and layout. But it flew fine.

Now I did it "right" and I can't take off anymore.

[hamslice]

skookum,

i am getting a ton of cyclic in this setup (<+-11deg) so i just dont see how i could not have enough. i am making a new mount for the gyro to provide a sturdier base to mount it on. as far as slop in the head there is virtually none. i am running radix blades maybe they are not right for a flybarless setup. like everyone else i am also having the kick go towards the gain dials on the unit. another thing is that when i have the hiller gains down low it will not kick, but when i have them up to right before i get oscillations the heli will kick. one thing that might have something to do with it is the long servo arms that the hfx frame requires. i am using mixer arms in the head to add a better ratio though so i dont know. i am running out of ideas.