Cause of Right Tilt and Left Turn Problem

[Golden Eagle]

I've noticed a lot of complex explanations for the right tilt that the MSRX exhibits, as well as the left turn problem. But its actually pretty simple:

Has anybody observed that the reason it tilts to the right is simply that it is heavier on the right side. For whatever reason, Blade decided to put the motor on the right, but there is nothing to counterbalance it on the left side of the control board. Turn it around, sight down the tail and look at the placement of the motor; it's huge, and its positioned to one side.

Turn around an MCPX and you'll see that the motor is dead center, with one servo on each side of the bord, perfectly symmetrical.

Turn around the old MSR. Same thing: motor dead center and servos on either side.

The MSRX is ridiculously assymetric. It's no wonder it flies like it does. I'm willing to bet that the AS3X has built in programming that attempts to electronically compensate for this. However, the result is that it behaves differently on right and left turns because the heli is mechanically assymetric. Try to balance it out by weighting the opposite side only partially helps, but I suspect that this might confuse the AS3X which is programmed to expect the heli to be lopsided.

So when you see your heli hanging in the air with one skid higher than the other - its not complex interactions of torque or the counterforce of the tail - its just a heli that was built with a somewhat questionable assymetric weight distribution.

[Learnig2Fly]

No kidding... I've only had a Coax before. So do most of the other helis fly level. I was wondering 'bout that too ....

[Golden Eagle]

Unfortunately yes. I have two blade MCPX, an MSR and a 120SR. All fly level and are very easy to turn either right or left. I don't think flybarless in a micro with lopsided weight distribution works well together for some reason.

[itsmillertime0]

My understanding is coaxial's are level in flight because of the counter rotational blades. I doubt weight distribution has anything to do with it. Hold the msrx by the blade grips and see if it hangs to the right. It doesn't.

[Golden Eagle]

Yes I tried that, but I am sorry, I can't refute the evidence of my eyes or common sense. The heli is obviously constructed with a heavy motor on one side. And that just happens to be the side to which it leans. When you hear hoof-beats, look for horses not zebras. Usually the most obvious explanation is the right one.

[asodestrom]

The board and both servos are to the left of the main shaft. If you tighten up the blades so they don't swing and align them along the center and hold it up by the blade tips, it hangs pretty straight. Maybe not so out of balance as your eyes might suggest.

[Golden Eagle]

I have tried suspending it from a string down the middle. It does appear reasonably balanced. However I still have suspicions about the assymetric design. A spinning motor must also induce gyroscopic affects. What happens when you put a tiny gyroscope to one side of a heli and not the other? Are there any other helis that have off-center motors? I can't believe that the degree of tilt and the absurd difficulty of left turns is normal, or the odd assymetric design is coincidental.

[wulfrun]

The tail rotor pushes the heli to the left, so it's forced to lean into a hover - just normal for the type. My Belt CP has a noticeable lean into a hover and that's almost perfectly symetrical. My friend's Blade 400 leans even more noticeably and even full-size ones do it.

Since it's leaning, any piro will tend to make the nose rise or dip because of the canted attitude; again no surprise there - gyro-effect of the head. That accounts for the turn difference and I think it's just more evident because it's such a small and lightweight machine. Compare the mSR, which has a canted shaft so that it hovers reasonably level. There's still a difference between left and right turns, just less so because of the cant.

As for balance, it does balance fairly well along the axis. The motor is offset by the board. Maybe not perfection but I'd say it's close enough.

[toys2cars2toys]

This effect mentioned above is called translating tendency
http://www.copters.com/aero/torque.html
- Note that this text talks about tilt to the left, that's because most real helis have counter clockwise main rotors and tail's pushing to the right

One of my helis has a CCW rotor, it is very disconcerting to see a left lean, once you've been conditioned to accept a right lean

The effect is more pronounced if the tail rotor is a lot lower than the main rotor disc - which it is on the msrx

I was surprised at the amount of lean, it is the most that I've seen on a heli, to the extent that I checked the longitudinal balance, as Wulfrun said the msr has a canted shaft to mask the effect - this was widely criticized

Interesting comment about the gyroscopic effect of the motor itself, I don't think that this has an effect but do not know. Every model heli that I can think of bar 1 exception has the motor offset from the main shaft, it's just normally in front or behind, I've not heard of that being a problem

[wulfrun]

Originally Posted by Golden Eagle

A spinning motor must also induce gyroscopic affects. What happens when you put a tiny gyroscope to one side of a heli and not the other?

First off, it's a coreless motor so the rotating mass is very low and is compact. Second, whatever gyroscopic effect it has is actually opposite to the main rotating mass' effect because it spins the opposite way. The fact that it's offset to the right, rather than in front or behind the main shaft might make a tiny difference.

If you're worried about the main motor, what about the tail? Every time you pitch up or down the tail's rotating mass will want to cause roll but no-one notices that. Again, largely because the mass is small.

[Ric66]

All good discussion, but I agree the right lean is normal for a single rotor heli (with a perpendicular main shaft) and is caused by the side thrust of the TR. Dasfriek has actually balanced his mSRX on the head of a ball point pen from underneath the main shaft. I don't recall him saying CG was off in the roll axis. Das where are you? Please weigh in! On the subject of left turn theories, I have personally observed the "return to starting point" behavior of AS3X on the swash when introducing a rudder command while holding the heli in my hand at half throttle. No idea whether this happens when the bird is unconstrained in free flight, but I believe that is why an un-level swash starting point helps with rudder turns (the MM mod).

[koko76]

Originally Posted by Golden Eagle

Yes I tried that, but I am sorry, I can't refute the evidence of my eyes or common sense. The heli is obviously constructed with a heavy motor on one side. And that just happens to be the side to which it leans. When you hear hoof-beats, look for horses not zebras. Usually the most obvious explanation is the right one.

But your eyes are not measurement instruments, nor is "common sense". You can determine through experimentation (pointed out earlier in the thread) that the helicopter does balance about it's center. This does refute "looking at it". The motor may be on one side, and it may be "heavy" but so are other things on the board that balance this weight. You also don't stop looking once you think you know the answer. Just seeing one heavy thing in the helicopter doesn't indicate that it is the source of an issue.
The right lean, and the flying characteristics are due to the type of helicopter this is, and the limitations that come with fixed pitched blades (main and tail). People go so crazy trying to get rid of what they perceive to be a design issue, when in reality they need to learn to fly the heli that is.

[Golden Eagle]

Everyone make good points. But the MSRX still doesn't turn like other fixed pitched helis. So here is a new suggestion:

The other big difference with the MSRX is of course that it is flybarless.

If you read the excellent review here: http://www.rchelicopterfun.com/blade-msr-x.html

He makes some important points:

"You are not in control of the cyclic servos, the AS3X gyros are. When you give a cyclic command on the radio, you are telling the gyros to allow the heli to move in a certain direction at a certain speed; they then move the swashplate servos to the required position to fulfill that command and are constantly monitoring both your instructions and what the bird is doing to keep both in sync."

Furthermore:

"These gyros operate in a feedback loop".... and..... "As with all single rotor helis with tail rotors, there will be some left TRTT (tail rotor translating tendency) that you have to anticipate for and may have to hold in a little right cyclic to compensate for the tail rotor thrust just as you lift off, but other than that, the Blade mSR X will lift off in perfect cyclic trim most of the time and once in the air, the AS3X system keeps it in perfect trim throughout the entire flight"

An the revelant point for us:

"it will even lean the heli to the right to vector in some cyclic TRTT compensation"

My conlusion:

When you turn left you are breaking out of a feed back loop and sending new intructions. The ASX3 has to decide how much left turn input to process, and how much compensation for TRTT to maintain. It's a constant battle between the input of the pilot and the feedback loops that maintain the heli "in perfect trim", particularly when these instructions contradict each other. When you break out of these loops, the heli lurches one way or another requiring you to compensate.

[koko76]

If you want to understand what is going on in the A3X system you should try reading up on some of the open source systems of similar design. Aeroquad has a "gyro only" mode that operates very similar to the systems used here, down to the use of the same sensor (HH is using the oem version of the ITG3200 3 axis gyro). There are many other projects of similar nature and varying complexities.
The concept of "feedback loop" isn't really used correctly. It is not the case that the heli battles to stay upright and counters your input. In fact there is no way for a system with 3 gyros to know which way is up, sensing only rotation about each axis. To know what way is down you need an accelerometer. Similarly when you move in a linear fashion the gyro cannot sense this. These (and other) issues are well documented in the referenced projects, which is why auto stabilizing systems use combinations of sensors.
I do think that the A3X is contributing to the symptoms, but I think that article you reference confuses things more than it helps. All the A3X is doing is countering uncommanded motion about an axis. There is only so much it can correct for. You don't "break out of a feedback loop" with a different command. Feedback is simply the sensing of the thing you are changing, like the pot in a servo.

[toys2cars2toys]

Originally Posted by Golden Eagle

When you turn left you are breaking out of a feed back loop and sending new intructions. The ASX3 has to decide how much left turn input to process, and how much compensation for TRTT to maintain. It's a constant battle between the input of the pilot and the feedback loops that maintain the heli "in perfect trim", particularly when these instructions contradict each other. When you break out of these loops, the heli lurches one way or another requiring you to compensate.

I think that you are right but with the wrong conclusion

Any proper closed loop control system should not "break out of it's feedback loop" when a new instruction is made, it should be added into the existing loop

The mcpx does not do anything odd like this during turns it has essentially the same control system, FP or CP is not relevant to the discussion

However the msrx DOES do something odd to it's control system when rudder is applied, it's as though it goes out of a hold into a rate mode with much reduced gains
This makes no sense (coming from an industrial controls background that I have studied for years to a high level), my best assumption is that the msrx's control system could not keep up with the heli in piro and hence a work around had to be made - less than ideal but fine in flight once the swash is trimmed properly

Have a look here for a further explanation and a solution
https://www.helifreak.com/showthread.php?t=372983

[toys2cars2toys]

If it helps any this is a typical control system



The sticks would be the input on the left and the servo the "Controlled process", the movement of the heli the output on the right
The rectangular arrowed circuit is the closed loop, any commands are simply added in, an error is generated and acted on accordingly

This is how the cyclic of an mcpx works all of the time, so does the msrx when the rudder is not on

Practical test
With the blades removed (main and tail), try holding the msrx flat in your hand, throttle up, don't move the cyclic, gently tip the heli and watch the swash move as the AS3X tries to correct against this unwanted movement (error). Keep bringing the heli back to level so the AS3X is happy again and not tilting the swash, keep playing with it like that and you'll get a feel for how the gyro is working
Now repeat, but this time whilst holding in a little rudder, you'll notice now the swash reacts in a very different way, it first "centres" the servos to their midpoint and then sluggishly tries to correct against the roll

This does not make sense a heli's cyclic control should not go into a reduced mode just because rudder is applied. The fact is that it does on the msrx (not on the mcpx or bigger FBL helis) which is why I believe that this is a compromise due to a basic control system

I should stress that my msrx still flys well and now properly trimmed, the effects are minimal in flight

So a heli with a 3 axis gyro may drift during piro, but it should not roll - the msrx does roll and hence why it can shoot off if not trimmed

[koko76]

Originally Posted by toys2cars2toys

The mcpx does not do anything odd like this during turns it has essentially the same control system, FP or CP is not relevant to the discussion

But it doesn't really have the same control system if you think about it. Cyclic authority varies with rpm in a fixed. You don't have the same response from the cyclic as in a CP as the head speed is constant. It's definitely going to be less linear. I would imagine this makes tuning of the PID's a little stickier.

[toys2cars2toys]

Originally Posted by koko76

But it doesn't really have the same control system if you think about it. Cyclic authority varies with rpm in a fixed. You don't have the same response from the cyclic as in a CP as the head speed is constant. It's definitely going to be less linear. I would imagine this makes tuning of the PID's a little stickier.

I accept that there will be differences, but for the purposes of this, I was trying to explain that they both have 3 axis gyros that in principle work in the same way, In a hover you can make a reasonably fair parallel between them

Really the point is that the mcpx and others don't revert to a "dumbed" down mode when rudder is applied


When I was in the process of selecting servos for my FBL Trex's I did a lot of reading, there were numerous comments along the lines of "that one won't be quick enough to keep up" even though it was fine for it's equivalent FB'd heli, the problem is more apparent as the heli gets smaller, lower mass and faster reacting

I wondered why the msrx had a higher head speed than I was expecting and also heavier blades, I'm sure that this is to give some gyroscopic stability and help with the control. I tried a set of cut down msr blades on mine last night, but the heli was very poor and wobbly
I think that HH would have ideally liked to have had lower headspeed and more pitch to make a friendlier indoor model, I'm sure that they reached the point where that gyro/processor/servo combination could not keep up

[CopterNator]

I just turned on my new msrx and did notice this lean toward the right. I thought my floor may be crooked but thanks for sharing what I always though. And yes, seemd like there is a larger weight dist. on the right side of the heli causing this tilt.

[LawnDartHeli]

Look at the old msr, the main shaft is tilted to the right (slightly, but noticeable). The msrX, like most single rotor helis with a straight main shaft, will tilt slightly to the right in flight. It is TT, there is no way to stop that (and why would you want to?). The left turn issue in my opinion is just a nasty side effect of a very lightweight FBL FP, youll just have to learn to fly around it.