Author Topic: my latest experiment with gyro sensor (Read 2782 times)

Igor Burger · « **on:** February 26, 2008, 03:15:41 AM »

first flight with too strong regulation:

it was sometimes too difficult to controll shapes and conrers especially when regulator tried to slow down

after few flights and softer regulation:

much better, but I still thing the 4-2-4 simulation worked better (for feeling)

Kim Doherty · « **Reply #1 on:** February 26, 2008, 09:31:02 AM »

Hi Igor !

That is very good!

Congratulations!

I like the second setup much better than the first.

Kim.

Dean Pappas · « **Reply #2 on:** February 27, 2008, 09:40:41 AM »

Igor!
That's dynamite.

You are using the gyro to measure yaw rate?
Dean

Dennis Adamisin · « **Reply #3 on:** February 27, 2008, 05:28:20 PM »

Igor:
Congratulations that looks great! That looks like a big step forward.

Are you finding that the gyro and the governor are fighting each other? Is that part of the reason the second flight was smoother?

I have not even completely fihnished my first ECL airplane - and now it is OBSOLETE!

Igor Burger · « **Reply #4 on:** February 27, 2008, 11:41:13 PM »

Thanx Dean, but I must repare you, it is not Dynamite, it is Danubit

(just for the info - danubit is plastic explosive which got the name from the river Danube which is under my home windows Danube=Donau

)

Igor Burger · « **Reply #5 on:** February 27, 2008, 11:52:49 PM »

Dennins, I do not think they fight each other, the Danubit unit

USES the governor ... the governor makes its action the same every time ... the gyro can be sure that if it says 5000, thne governor just makes it for sure.

That first video just demonstrates that we cannot get really constant speed, the setting of the control unit was so strong, that the motor was almost off and that made the model almost uncontrollable in some moments (very well visible on wingover where I almost crashed and in bottoms of triangels).

The second flight I did it softer = allowing more speed variation and you see that the flight is closer to what we expect.

The same fact I found when I measured speed of large model, I foud that speed overhead falls down to approximately 2/3 or even less. (measured on electric model with governor only)

Dean Pappas · « **Reply #6 on:** February 28, 2008, 08:39:15 AM »

Hi Igor,
Danubit ... I like it. I have a work buddy who is an amateur munitions expert: this morning he told me more than I want to know about it!

Please confirm for me that you are inferring the flying speed, relative to the ground, by measuring the yaw rate and using that information to vary the governor set-point.
If that is true, then flying in outdoor wind will be an interesting test.
In normal non-turbulent wind, the airspeed will be low approaching almost all maneuvers,
and the plane will be accellerating just as dead-downwind is reached. That should be okay $:-\$
It might make it easier to fly good shapes on the overheads, because the extra airspeed when pointed into the wind will
help prevent the classic flat-spot.

So now we have two fundamentally different approaches you have tried.
One is to keep the airspeed constsnt, the other is to keep the ground-referenced speed constant.
Interesting. Thanks again for sharing.

Hey Dennis, the governor can't fight the gyro, because it is slaved to it. Right Igor?

later, Friends,
Dean

Igor Burger · « **Reply #7 on:** February 28, 2008, 11:15:21 PM »

Short answer is yes, yes ,yes :-)

However I affarid of some problems in wind. I had seriouse problems to make proper shapes in that first flight on the video, and it can happed that low throttle downwind just before entering the figure will make the same. But as I wrote already, the system with gyro maker errors on some place while the system with 4-2-4 makes it on another places, I want to make them in serie in hope that they will cancel their mistakes each other. For example the 4-2-4 system will make low throttle upwinnd, but gyro will recognize error and it will add little bit. The same downwind. But they will work together in maneuvers, so I will make very conservative (soft) setting for both of them, it should be enough for maneuvering, but will make only small (or I hope none) errors in wind ... we will see. I must do it quicly, before someone comes with idea to disallow motor controlling

Dean Pappas · « **Reply #8 on:** February 29, 2008, 11:30:32 AM »

We are on the same wavelength, Igor!

I just had this long conversation with Bobby Hunt, after I asked him a deceptively simple question: "would you prefer the system to try to maintain airspeed or groundspeed (maneuver timing) as you turn to upwind in the overheads?". He started to give me an answer three different times, then he saw the inherent compromise. I need flying weather! Everything else is an educated guess. Do you have a preferred gyro vendor? Did you roll your own with a commercially available part?

I have simulated an accellerometer that is immune to attitude (in microcap). Now I have to build it. It is a spring-centered see-saw that is balanced at first-order moments, but intentionally unbalanced at second-order moments. This would provide similar, but time-leading, information as the yaw rate gyro. maybe the gyro plus a low-gain differentiator would be better.

later,
Dean

Howard Rush · « **Reply #9 on:** February 29, 2008, 12:35:25 PM »

The maneuver I'd worry about would be the wingover entry on a windy day. If the lines go slack, the airplane might get into an unstable left yaw. I reckon that one should start by looking at yaw rate data from an instrumented, but unaugmented airplane.

Dean Pappas · « **Reply #10 on:** February 29, 2008, 01:11:36 PM »

Hi Howard,
Yeah that's another one, but even well trimmed ships flyingh this maneuver normally tend to do a few quick degrees of yaw "in" at the very top of the circle. Ouch. Hint ... tailess airplanes like the Genesis do it less.
Dean

jim gilmore · « **Reply #11 on:** May 15, 2008, 03:26:53 PM »

Very cool video. I wish everybody would upload videos this way.
How long are your lines, and how high is that ceiling ?

Igor Burger · « **Reply #12 on:** May 15, 2008, 10:20:38 PM »

The lines are ~15ft and ceiling is in touch of right wing tip :-)

jim gilmore · « **Reply #13 on:** May 16, 2008, 12:18:57 AM »

Again thanks for answering my questions. Could you upload, or point me to where I can see a static picture of you're model ? My computer is a t-mobile mda, so my screen is only about 2.5x1.5.
Thanks in advance
Jim Gilmore

Igor Burger · « **Reply #14 on:** May 16, 2008, 12:33:57 AM »

http://www.rcmodely.sk/zobrazclanok.php?id=300
it is in slovak language, but pictures say all :-)

Jim Moffatt · « **Reply #15 on:** May 22, 2008, 06:09:54 AM »

Igor:
What is the wingspan, wing area and length of your plane?
Also how thick is the foam and what is the all-up weight?
Thanks
Jim

Igor Burger · « **Reply #16 on:** May 22, 2008, 06:28:37 AM »

The weight is 180 grams, foam thickness 3mm and span & length 900mm

The wing has carbon LE & TE

Jim Moffatt · « **Reply #17 on:** May 22, 2008, 08:58:38 AM »

Igor
I was totally astounded at your video! I have been flying electric CL for several years but I have never seen anything like your plane. I fly a Brodak Baby Clown with Suppo 2208-8 motor on 2S 1350 mah cells and 30' line length. I get too dizzy for the handle so I do everything by RC. I hope to evolve to a plane like yours with about 20' line length for indoor use. Thanks so much for posting your work. I am truly in awe of what you have accomplished.

Igor Burger · « **Reply #18 on:** May 22, 2008, 09:01:14 AM »

Thanx Jim, BTW the lines are only 15ft long :-)

Jim Moffatt · « **Reply #19 on:** June 05, 2008, 06:28:11 PM »

Igor

I have two questions:
1. Did you consider using flaps on your plane?
2. If I implement one gyro how should I do it?

Thanks
Jim Moffatt

Igor Burger · « **Reply #20 on:** June 06, 2008, 12:36:28 AM »

I think flaps are not good idea. Flat wing does not work well with flaps (airflow can separate also at 0 angle of attack) and flying in calm gymn not need any flaps, because such a model can have any wing area ... flaps are necessary outside to limit wind area for turbulent air. And last reason, higher lift coefficiet with flaps will make more wing tip worticles, and it will make troubles in calm air. This wing works well, has enough of lift and does not make feeling of turbulence in repetitive figures.

Regarding gyro - I have my own microchip translating timer signal and gyro signal together. I affraid we do not have any well programmable gyro to use without such a device. May be I will make it available later, but it is not finished yet and for sure I will not play with it until WC in France. No time for such work ... may be october, november before winter indoor seasone. I think you can easily use constan rpm mode until that time. The device uses functions available in Jeti Spin ESC in heli mode.

Jim Moffatt · « **Reply #21 on:** June 14, 2008, 06:46:46 AM »

Igor

I have tried flying my RC/CL plane with and without governor and the stunts all work better with governor mode. Without your post I would not have tried this.

Once again thanks very much for your posts.

Best wishes,
Jim Moffatt

John Cralley · « **Reply #22 on:** June 14, 2008, 02:08:29 PM »

Igor,

Is the gyro you are using something like the one in this link???

http://www.cy-model.com/product_info.php?cPath=33&products_id=509&osCsid=1a153c50a571780e8ab537a88f4e748a

John

Igor Burger · « **Reply #23 on:** June 14, 2008, 05:24:40 PM »

Not exactly that one, but it LOOKS it can be same system.

John Cralley · « **Reply #24 on:** June 15, 2008, 01:55:33 PM »

Quote from: Igor Burger on June 14, 2008, 05:24:40 PM

Not exactly that one, but it LOOKS it can be same system.

Igor, I have ordered one and will see if I can make it work.

John

John Cralley · « **Reply #25 on:** June 25, 2008, 12:02:29 PM »

Igor,

The E-Sky EK2-0704 gyro arrived today. Naturally, there are no directions (it came from China) but it appears that it is to be inserted between the timer (throttle for an RC) and the ESC. I presume it is powered by the BEC and passes through the power to the timer. It has a normal and reverse switch and a - to + pot on the top for, I presume, setting the sensitivity. I have not hooked anything up yet so the above is all assumptions. The weight is 10 grams.

How do you orientate the gyro on our airplane. I suppose maybe the orientation may be different for different brands of gyroscopes. Is the idea to increase the RPM when the plane starts to climb and decrease it when it dives?? These may be stupid questions but I have no idea how the RC helicopter people use these gyros on their helicopters.

Thanks for any help here.

John Cralley

Igor Burger · « **Reply #26 on:** June 25, 2008, 02:42:06 PM »

turning left will decrease the throttle

John Cralley · « **Reply #27 on:** June 29, 2008, 02:06:18 PM »

Igor thanks for the reply,

I've just got back to feeling good following hospitalization to treat a bleeding problem after a colonoscopy and have been off of the Internet for a while.

If I have this correct, you are using the gyroscope to control yaw and the ESC governor mode to respond to loading during maneuvers. So you have the gyro set to decrease RPM whenever the plane yaws to the left (in a counterclockwise flying model). Humm, I'll have to think about that a bit because it seems to me that you might want to increase the RPM to counteract the yaw rather than decrease it.

Anyway, thanks for the information.

John

Author Topic: my latest experiment with gyro sensor (Read 2782 times)

Kim Doherty