MEMS Gyroscope inside rockets

Discussions about anything electronics, from altimeters to cameras.

Moderator: Moderators

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 2085
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: MEMS Gyroscope inside rockets

Postby SpaceManMat » Thu Mar 31, 2016 7:29 pm

Looks like a great project you got there.

Just a work of caution, you cannot build a guidance system as that would be well on the way to creating a weapon. You can however create a stabilisation system, which it sounds like what you are actually doing.
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 13,647 feet
Fastest Flight: Mach 1.55
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

joeman
Rocket Crew
Rocket Crew
Posts: 333
Joined: Fri Mar 06, 2015 9:05 am

Re: MEMS Gyroscope inside rockets

Postby joeman » Thu Mar 31, 2016 7:38 pm

Yes, more a stabilisation system.... to correct for inaccuracies in build or external influences that cause rotation.

Its sole purpose at present is to reduce rotational velocity. I hope to get it to compensate for the rotation as well...and keep it pointed up.
L1 - Callisto (H133) - 11-Jan-2016
L2 - March Fly (1633K940) - 18-Mar-2018 - RSO

joeman
Rocket Crew
Rocket Crew
Posts: 333
Joined: Fri Mar 06, 2015 9:05 am

Re: MEMS Gyroscope inside rockets

Postby joeman » Fri Jul 01, 2016 6:34 pm

Hi,
Been a bit slow to show results.

Here is an interest graph of recent flight - rotation rates about 3 axis.

http://leederville.net/rocket/wp-conten ... _image.png

The actual ejection charge didn't ignite for about 1.5 seconds after the data points. So no 'crazy' movement.
(Was able to deduce this based on camera footage of camera inside rocket)

You can see that the motor burned for 1.5 seconds and you can see the dampening of the motion about x, z axis during the burn, as aerodynamic forces act on the rocket.

You can also see at the end how the rocket starts to tip over....ever so slightly..as it's velocity decreases and the stabilizing forces reduce some what.

Interesting stuff.

Cheers

Joe
L1 - Callisto (H133) - 11-Jan-2016
L2 - March Fly (1633K940) - 18-Mar-2018 - RSO

User avatar
Happy Heyoka
Rocket Flyer
Rocket Flyer
Posts: 147
Joined: Wed Jul 22, 2015 7:01 pm

Re: MEMS Gyroscope inside rockets

Postby Happy Heyoka » Fri Jul 01, 2016 10:58 pm

[quote="joeman"]Hi,
You can see that the motor burned for 1.5 seconds and you can see the dampening of the motion about x, z axis during the burn, as aerodynamic forces act on the rocket.

You can also see at the end how the rocket starts to tip over....ever so slightly..as it's velocity decreases and the stabilizing forces reduce some what./quote]

Trying to understand your graph - how are the sensor axes aligned with the rocket body?
Also, the values look absolute (not rate?)
(humour me, I'm thick at the best of times, Friday night after a beer doubly so)
Ha ha ha ha ha! You can't fool me! There ain't no Sanity Clause!
Chico Marx - A Night at the Opera (1935)

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 2085
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: MEMS Gyroscope inside rockets

Postby SpaceManMat » Sat Jul 02, 2016 7:56 am

Can you please remove the markers from the graph? You'll get a much finer line. Also I'm not terribly fond of radians.
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 13,647 feet
Fastest Flight: Mach 1.55
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

joeman
Rocket Crew
Rocket Crew
Posts: 333
Joined: Fri Mar 06, 2015 9:05 am

Re: MEMS Gyroscope inside rockets

Postby joeman » Sat Jul 02, 2016 4:19 pm

The x axis goes to left, Z axis out of the page and y-axis goes up to the stars.

They are rates. They are output from MEMS gyroscope.

Sometimes these things are better shown as a movie/animation.

https://youtu.be/Buk305yh3do

I do have a movie somewhere of the launch taken from under the rocket and the motion is similar. Though admittedly it does look like it is gyrating a lot at launch. I did try to get the animation to show the motion in real time...but it is hard to get it perfect and even harder to get it into a video as it happened in real time.

Yes, I really probably should have used gnuplot, instead of excel. But was impressed with what it showed and then didn't go on to tidy it up (which I really should have).

Hope that inspires a few people.

Cheer!

Joe
L1 - Callisto (H133) - 11-Jan-2016
L2 - March Fly (1633K940) - 18-Mar-2018 - RSO

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 2085
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: MEMS Gyroscope inside rockets

Postby SpaceManMat » Sun Jul 03, 2016 1:40 pm

Nice movie. Did you rite the code or just adapt it?
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 13,647 feet
Fastest Flight: Mach 1.55
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

joeman
Rocket Crew
Rocket Crew
Posts: 333
Joined: Fri Mar 06, 2015 9:05 am

Re: MEMS Gyroscope inside rockets

Postby joeman » Sun Jul 03, 2016 2:52 pm

L1 - Callisto (H133) - 11-Jan-2016
L2 - March Fly (1633K940) - 18-Mar-2018 - RSO

User avatar
Happy Heyoka
Rocket Flyer
Rocket Flyer
Posts: 147
Joined: Wed Jul 22, 2015 7:01 pm

Re: MEMS Gyroscope inside rockets

Postby Happy Heyoka » Mon Jul 04, 2016 12:37 am

joeman wrote:The x axis goes to left, Z axis out of the page and y-axis goes up to the stars.
They are rates.


cool, I was a little bit confused by that point of inflexion around 1500ms on the Y axis but it makes sense now.

just to clarify, this is just data capture - you didn't actually have your stabilisation hardware running?

Sometimes these things are better shown as a movie/animation.

yes, helps a lot. I like your software too - I'm not sure if there are Java bindings for OpenSceneGraph but it's got a lot of nice widgets you might be interested in (it's pretty high level - which I like for this kind of stuff, not unlike Java3D). But you've got something working which is half the battle.


I do have a movie somewhere of the launch taken from under the rocket and the motion is similar. Though admittedly it does look like it is gyrating a lot at launch. I did try to get the animation to show the motion in real time...but it is hard to get it perfect and even harder to get it into a video as it happened in real time.


jase has a thread somewhere about a package that generates a configurable overlay from telemetry data... I think you're right in that syncing the two can be difficult. I'm messing around with high speed cameras and eventually when I get my act together I'm going to try and come up with a "foolproof" scheme for doing that. One possibility is that most of the action cams today set timestamp metadata from the GPS clock; if you have an onboard GPS you can also timestamp from the GPS and Bob's your mother's boxing instructor.


Yes, I really probably should have used gnuplot, instead of excel. But was impressed with what it showed and then didn't go on to tidy it up (which I really should have).


Good, cheap data visualisation software is not easy to come by. I do use gnuplot quite a bit and there are several decent graphical front ends for it. Maybe something like LabView? Something interactive would be really interesting to play with (ie: click one a data point and see the correct video frame and sample values).

Back in the day I did a lot of work integrating data visualisation with video using QuickTime - but Apple knifed the QTJava baby quite a while back and I'm 99% Linux based these days anyway. I guess I should have a look to see if any of the Linux MPEG4 libraries implement a vector graphics codec to use for an overlay video track (the ISO/MPEG4 container - aka QuickTime move file format - has a nice mechanism for correlating data with different sample rates).
It's one of my back (back) burner projects to learn enough about Blender to try and use it's timeline and video integration stuff to do this (animation is pretty easy from external code/data sources if Python doesn't drive you insane).

But off course, you should feel free to continue using whatever tools you're comfy with!
Ha ha ha ha ha! You can't fool me! There ain't no Sanity Clause!
Chico Marx - A Night at the Opera (1935)

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 2085
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: MEMS Gyroscope inside rockets

Postby SpaceManMat » Mon Jul 04, 2016 9:25 am

Dashware is the program you can use to combine video with telemetry data. Had a lot of fun trying out a few of the gauges with Raven and Eggtimer logs.

viewtopic.php?f=32&t=5443
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 13,647 feet
Fastest Flight: Mach 1.55
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

joeman
Rocket Crew
Rocket Crew
Posts: 333
Joined: Fri Mar 06, 2015 9:05 am

Re: MEMS Gyroscope inside rockets

Postby joeman » Tue Jul 05, 2016 9:47 pm

Thanks for all the interesting info on software. I've found it helpful to combine what I see in camera footage and gyroscope, so I expect this merging of data to be even more important in future work.

There was no stabilization mechanism in that rocket flight. I'm doing this project incrementally....trying to set achievable goals for each step. For that launch I wanted (among many other goals) to test the build of the payload section and characterize the motion of the rocket without any stabilization system.

I'm working on a Servo-base system now....which I hope to use in next flight. It will however just move the CG to the 'left' by about 1.3mm (max) at some pre-determined time after flight (probably about 0.5 seconds....after the rocket has stabilised). What I'm expecting of course is for it to rotate ever so slightly in the direction the CG has moved and stabilise due to aerodynamic forces. I want to try and see the effect it has...get some data so I can better understand how it will affect the flight.

What I'm ultimately interested in is creating a PID controller (or PI) that can help stabilise the system. I'm hoping I can get a bit of a feel for 'gain' factors.

Lots to do!
L1 - Callisto (H133) - 11-Jan-2016
L2 - March Fly (1633K940) - 18-Mar-2018 - RSO


Return to “Electronics”

Who is online

Users browsing this forum: No registered users and 4 guests