Jump to content
manuelit

Need help for gyro stablized camera mount

Recommended Posts

...Please forgive my terrible English...

I have been reading about the PAN CAM Chip, and I find it a great start for what I would make.

But I need help from some more expert people than me.

I would like to make a gyro stabilized camera mount, it should be stabilized on the 3 axis, but I believe that the most important is that it keeps the horizon on the roll axis.

The movements of the mount should be made by 3 servos helped by the POWER GEARBOX SERVOS of Servo City.

I made some experiences with very bad results installing the piezo gyro directly on the moving part of the mount and simply connected to the servo to keep that moving part in balance.

It seams that or the servo throw or the gyro action range are not enough to do that work.

How about if I connect betweene the gyro and the servo a PAN CAM Chip, would this help the servo in it's travel to balance? and is it possible to use the PAN CAM with a modified continuous rotation servo, that would help the the system to make longer travels (even in the need of larger pan and tilt needs?)

I do not want to make a very long and boring Post, but I would have many other questions.

I will post some more if some of you guys find the argument interesting.

Please HELP ME, what I am try to make is some thing that would help me much in my work, I am a young photographer and I am looking for some new services (aerial photography and filming) to offer my customers here in Italy.

Thanks

Manuel

Share this post


Link to post
Share on other sites

I do not know how the control response would behave if PanCam was used with a gyro (I never tried it). It will certainly work differently than if the PanCam was not involved.

These sort of gyro control applications are difficult to make work correctly. That is because the loop response of the R/C gyro is optimized for stabilizing the full airframe. A gyro that has programmable parameters (like CSM's) will help.

However, these folks have had some successes with adapting R/C heading hold gyros with a post processor board: http://runryder.com/helicopter/t115762p1/

As easy as it sounds, using off-the-shelf R/C gyros to stabilize a pan/tilt is not a simple plug-n-go sort of exercise. It will usually require a lot of experimenting.

Share this post


Link to post
Share on other sites

Okay, I've had to read the PanCam description again :)

In fact I think I could help. I've designed a similar device, which converts the position control into a speed control. Apart from camera pan/tilt and landing gear retract I haven't been able to find another use for it before around 2 days ago when I came across a gyro.

If you put the gyro between the module and the servo you get a correction like on a heli with a standard gyro. That's unstable so useless.

What really gets interesting is when you put the gyro between the RX and the module. As the module behaves as an integrator, you simply get a cheap heading-lock! (after the gyro's gain has been properly adjusted of course!)

I had absolutely not thought about this before trying, but in fact it's logical...

So for example if you attach the gyro to a servo and connect everything, when you turn the servo the horn will always point in the same direction, which will be changed by an action on the control stick.

The movement is very smooth, always thinking that we are using servos which work at 50Hz.

That's like an AVCS gyro. So it could be used as an "add-on" for a cheap gyro!

Will have to thes that on a heli once to confirm, but I'm hopeful ;)

What could be useful to you is that I have included a programmable servo travel limit on my module. With a push switch you can set both end points as close as possible to the servos mechanical limits, getting around 180° out of most servos without any hardware modification.

Wil have to test further the heading-lock thing for a pan/tilt, but it really seemed to work nicely. Now resonance problems could pop in when a 1kg camera is attached to the thing...

If you're interested I could give you the design, or send you ready modules if you don't have the equipment to make them. It uses 4 SMD components, one of them being a PIC12F675.

Let me know if you're interested... :)

Oh I just saw your line about continuous rotation... In fact a servo modified in this way (detached potentiometer) then behaves like a standard one with the PanCam. You just have to add the gyro between the RX and the servo and you get what I've described before. Also shortly tested, but in my case that particular pan/tilt assembly with modified servos had a mechanical reduction between the servo and the mount, so the gyro's gain could not be set high enough to get sufficient compensation.

Regards,

Kilrah

Share this post


Link to post
Share on other sites

Thanks a lot KILRAH, what you are describing is very interesting for my problem.

I would very appreciate to know some things more about your device

because if it makes the gyro to work as a AVCS I think it could be a grate help for my mount, and I find absolutely fantastic the idea that I could get the device “ready to fly” from you and I would like to know how much it would cost.

Do you think that it could keep the position for long time in flight condition?

I wonder about the problem of gain and reduction between the mount and servo..I would also use a 5.1 gear, could it be a problem?

And I was wondering if you have any idea of what would happen if I would get out of the servo it’s potentiometer and connect it mechanically to the center of the rotation axis, do you think that the servo would turn for all the possible movement of the mount bracket where the potentiometer is attached? and would it do it in proportional way? and would it work with your device?

I am sorry for all these questions but I find too much interesting to know from a clever and experienced guy like you...

And an other question is do you think that could be “electronically” possible to give the system an HOME position where it could gently go back in case that it would loose the horizon balancing (horizon is most important but even other 2 axis), I mean to give it a kind of position reset during flight, where it could start back to keep the pointing correct?

Thanks a lot!

Manuel

Thanks Mr. RC - CAM, the link you gave me is very much interesting.

Share this post


Link to post
Share on other sites

I plugged the PanCam into a rate gyro and it indeed behaves like a HH/AVCS gyro. The "centering" feature on PanCam even provides a remote/automated way to recal the servo back to its center.

Thanks for the tip!

Share this post


Link to post
Share on other sites

Hehe, stupid nobody had thought about this before.... that's why the good old method of trying still is useful :D

BTW as he's done the same I add my answer to his mail for the others...

Hello!

Firstly as I said I've just been playing with it for a few minutes for the

moment. So nothing precise yet. I've tried with a very cheap gyro, and the

AVCS worked correctly. I've compared with a "real" AVCS gyro on a friend's

heli today, and the behaviour was the same, although I thought there would

be a slow "return to center" to avoid overcompensation before take-off.

Nothing like this. Of course it's interesting!

When I tried on the pan/tilt assembly I found that it was precise even at

low speeds, which was quite amazing according to the conditions. What I've

been able to notice is that a small drift was coming after a few left/right

movements. This must be the gyro's fault, as I've used the same ones for a

head tracker and had noticed that their sensitivity was not symmetrical

(responding more in ont direction than in the other). Decent gyros should

not have this problem as it's not seen on good heli gyros, and the one I

used was the cheapest I could get here. Anyway this can be compensated a bit

by trimming on the R/C or by moving manually with the stick.

The problem with the gear is that my gyro's gain was not high enough. If you

use a reduction, the servo has to turn a long way to compensate an external

rotation. As the gyro is not basically meant to be used like this the

maximum gain is limited to a range that is decent for a helicopter, with

direct drive. In fact I could compensate this lack of gain by amplifying it

on the module. I've set it to a fixed value that gives good min/max panning

speeds with direct drive, but it would be no problem to change it.

The solution of putting the potentiometer after the reduction would be the

best. This would allow you to have the full travel (in this case the 270° of

the pot) easily, eliminating also the gain problem. Just don't forget to cut

the plastic travel limits in the servo :-)

That makes me think about using modified servos (with a fixed pot). I

haven't tried enough yet but it works at least without reduction. I think

that when adding it we would be stuck with the gain problem again. Another

issue is that it's difficult to trim the R/C to completely stop the servo,

which does not appear with my module as I've set a fair margin around the

neutral position that is stored at power-up.

I'm thinking out loud, but if it can give ideas....

I should have access to a pan/tilt assembly on monday and I'll test a bit

more as the owner is also interested. I'll tell you more after that.

I'm not sure if I understand your center position thing. Do you mean if the

AVCS has drifted, to be able to return to a certain position automatically?

You could do it yourself with the R/C stick... The gyro should not give a

signal that is offset from the info you give. I mean that if the gyro

doesn't move, it doesn't change the info it receives from the R/C. Having

the stick centered should still mean "no movement" according to the neutral

position stored at power-up by my module. At least if the gyro's circuit is

well-designed.

The problem with a thing like this is that what we are trying to do is like

professionnal and expensive systems, but with common and accessible devices.

As for now I haven't found anything very serious we have to experiment

ourselves... But there must be a way to it.

Oh just for info, do you intend to use this system just to shoot

pictures/video, or also to fly the aircraft?

Regards,

Kilrah

Share this post


Link to post
Share on other sites
Hehe, stupid nobody had thought about this before.... that's why the good old method of trying still is useful

That is so true. Even silliness can turn into a stroke of genius. Not that this idea was silly, but you know what I mean. ;)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×