Capi's Gyro 101

[rcjames]

If a servo is 960us but the gyro only work with 1520us and 760us, can i run it on 960us?

[DominicD]

I don't believe so.

[walleyevision]

What if C-A-T really spelled dog?

[Tucky]

This is the second time I read this, good stuff, already copy/paste to my personal library, I hope you don't mind but we ( read me) are doing translation fo literature, mostly what i know so far, since there's little to none of this stuff in spanish, we just started a blog and I'm planning on doing the post, so far the first one is helicopter 101, and I plane to go from there. Would you mind if I translate this? I would point to this thread of course.

[DominicD]

No I don't mind at all, just make sure I get credit. Have you read the ESC 101 ?

[Tucky]

I'm on it, but got distracted ( just a little bit) with Finless's gyro 102 video. just learning more on radio programming. yeah, as soon as I do the post I'll let you know and give you credit.

this is our blog so far
hobbylacuesta.wordpress.com

[Jasmine2501]

I really wish more people would read this.

[Laurens]

Almost 50000 views, so thats not bad at all.

[dmssbl4003d]

I'm relatively new to heli's (6 months), and wondered how the Gyro worked.
Now I've got to find how to adjust them and why.

Again, great post!
Thanks for your time!

[KennyC]

Why is the width of the pulse at 1520us the servo center and not 1500us?

[DominicD]

Dunno, doubt the reason would make any sense now adays.

[DannyvG]

I have some questions about the Rate mode way a gyro works.

Lets take another example where the pilot desired a fixed rotational rate but some wind or external force is trying to push the heli to rotate. Again the green area is the actual rotational rate and the lines are the desired rate. The red section is the difference between the desired rotational rate and the actual rotational rate. During the red section the heli actually rotated (area of the red section will tell you how much it turned) because this system ONLY trying to reduce the error between the desired and actual rotational rates NOT the direction the heli points. This is what we refer to as a RATE GYRO SYSTEM

This part from your story says to me that in rate mode the stick position gives the desired rate. And thus the helis rotation rate will try to converge to the actual rotation rate not taking history into account.

But explanations like this

Typical rate mode gyros will try to counter any turning of the helicopter, be it from changes in torque reaction, wind, or changes in tail pitch caused by pilot commands. An old trick was to use a programmable mix from rudder to gain, to reduce the the sensitivity and hence the resistance when rudder inputs were made, but keep the helicopter stable otherwise.
As a simple rate gyro tries to counter even inputs from the pilot, the travel of the servo seen on the ground (and limited by the servo end points set on the transmitter) will likely bear little relation to that seen in the air.

http://www.rcheliwiki.com/Tail_setup

tells me that a rate mode gyro will try to keep the rotation rate zero (edit, it wont try to keep it zero it will try to keep acceleration zero due to the derivative action? see below also) at al times. But that doesnt seem to comply with your explanation of rate mode.

Because for example trim (or a fixed offset from centre stick position) can make the heli to stop rotating. So that doesnt comply with how I interpret your explanation of rate mode.
What am I missing here?

After some thought, what I am perhaps missing is that as soon as the heli is in the air it will slowly start rotating (if midpoint is wrongly set mechanically) and after a short while the gyro will see this as his reference. If you then give steering input it will react but then try to react another equilibrium that in this case is zero rotation rate. But this will only stay when you keep the stick in the same position and external influences also stay the same.
Low frequent (steady state) the stick unput will equal a certain servo position. High frequent (D action) it will amplify the error and react to it. So short disturbances will be damped. While steady state a certain stick position will give a certain pitch.

It would help me a lot if I just had a model of the helicopters tail (or the complete heli ) now I am trying to base things on uncertainties.

Edit2, still dont seem to get it. What does the transfer function or frequency response look like for the Servo input to actual rotation rate?

[DominicD]

Looks like a typo, the gyro shouldn't resist commands by the pilot, only undesired changes such as by wind and torque.

Do NOT use trim on a HH gyro, this is a slow rotational command by the pilot. A good gyro won't need this.

[DannyvG]

Well the reason I started looking at this was after having a discussion with a scale heli pilot who actually also thinks it does that what you consider a typo.

Just to be clear Dominic, you are absolutely sure that in rate mode the stick position also represents a desired rotation rate and the feedback signal is the actual rotation rate?
Because I still find it odd that in rate mode it is possible to get a hover without rotation when applying stick input.

I will look into it further when I have matlab acces again.

[DominicD]

I am pretty sure, but I don't do alot of rate mode flying

[DominicD]

Originally Posted by djamgils

Because I still find it odd that in rate mode it is possible to get a hover without rotation when applying stick input.

I've never seen that, when you give a stick input in rate mode the heli should rotate.

[DannyvG]

Well, for me it also has been a long time ago since I flew in RM.

But suppose you have the slider setup wrongly. What will happen is dat the heli wil rotate in a certain direction. To counter that you simply trim it to the opposite side. But if stickinput/trim would desire a certain rotation rate then that doesn seem logical to me.

Or isnt it possible to trim out that effect? Cant remember. Normally the fix would be to slide the servo over the tailtube or to change the center position pitch.

Presuming trimming in rate mode wont give the desired effect.
Perhaps the pilot thinks he is giving a constant offset stick input but is actually constantly correcting very small amounts.

[DannyvG]

I have done some more simulations and I have come to some possible conclusions.

To get a HH gyro you should integrate the error signal so you go from yaw rate error to yaw position error and then apply a PID controller to get real HH. So this is not the same as how you illustrated it with only adding I to the PD.

And for Rate mode it seems that PD controller on the yaw rate is correct with your story and the practice.

[OlliW]

Hey DominicD,

this is an absolutely great write up.

I actually have a link to it on my web page since quite a while, but have kind of forgotten of it, and it just happened now that I "stumbled" across it again, with however having done kind of extensive work on gyros myself in the meantime (I programed my own gyro (mis)using a GA250, did in-flight measurements of the gyro action, and checked my conclusions by simulations).

There is actually one point I would not completely agree with your statements, namely as regards the role of D for preventing the back-bouncing. What you describe here is according to text book, but I think in a real heli-gyro that's not what is done, or more precisely that's not the main thing which is done. The point is that the servo has only a limited range, and after a sudden rudder stick movement (such as in piros) the controller actually brings the servo very quickly to its limits. Hence a D term in this situation is not very helpfull (and in fact acts like a D2 term). You could make it very damping of course slowing down the reponse to the rudder stick step, but this would also slow down the response to an external disturbance such a wind gust or a pitch pump. So what I believe is done is a "filter" or "rate limiter" only for the rudder signal, which also has the effect to slow down the response upon a rudder stick step, without sacrificing however the heading hold capabilities. I actually believe that this is what all the "acceleration", "deacceleartion" and similarly named settings in gyros such as the quark or fbl units are (for vstabi this is actually explicitely stated in the vstabi wiki).

I have a somewhat lengthy write up of my experimental results and conclusions here.

It's just a little detail, I know.

Cheers,
Olli

[DominicD]

I'll try and find some time to read your write-up this weekend, been super busy with my first job