Measuring Model Rocket Velocity: Is it Possible?

[LT72884]

Question, if i record a video of a model rocket launching, and then falling, is it possible to get the actual (or extremely close) velocity every 1 second?

thanks

 

[russ_watters]

Question, if i record a video of a model rocket launching, and then falling, is it possible to get the actual (or extremely close) velocity every 1 second?

If you know the distance, yes.

 

[LT72884]

total distance or the distance at each second?
thanks

 

[russ_watters]

total distance or the distance at each second?
thanks

Distance between camera and launch site, then you'll need to measure the angle/change in angle.

 

[hutchphd]

You need to know (somehow) both the angle and the distance. For instance knowing the horizontal distance to the pad and the fact that the rocket went straight up is sufficient. In general two fixed cameras whose relative orientation is known will also suffice.

 

[BillTre]

The trajectory of the meteor that blew up over that Russian town was figured out based on lots of videos from different angles.

 

[LT72884]

ok, so if i am 15 feet from launch site and it goes 1100 feet and takes 3 seconds to fall. how do i get the velocity of the rocket at every second or half second?
i know a bunch of rocket equations, but im trying to get the information from the cameras.

thanks

 

[russ_watters]

ok, so if i am 15 feet from launch site and it goes 1100 feet and takes 3 seconds to fall. how do i get the velocity of the rocket at every second or half second?
i know a bunch of rocket equations, but im trying to get the information from the cameras.

thanks

15 feet is way too close. You'll want to be at least 100 feet. And it may not work with a single camera due to the short flight and need to cover a large angular motion - you probably cant move the camera during the flight. Maybe 2 or 3 cameras could work. Measure your camera's field of view, then sketch a diagram of how it would look.

 

[LT72884]

that, and knowing the exact distance it traveled fram by frame in the camera. Maybe i need to be asking a different question.
I am wanting to calculate the Cd from the drag force equation of a falling rocket when the drag force and cd are not known. I just did this with a car coasting over 70 seconds in neutral. I used some numerical methods techniques to find the Cd and coef of friction. I am now trying to do the same with a falling rocket.
but the way i did it with the car, i had a change in velocity and change in time that i knew since every 3 seconds or so, i recorded the speed of the car.

 

[hutchphd]

First you need to understand how big your numbers are (and how well you need to know them) Your "for instance" is wildly off......the missile will take >8s to fall (with no drag). How good is good enough.....?
The rest is an exercise in geometry. Suppose the rocket fills ~one pixel on the image. The line that contains that pixel and the exact optical center of the camera lens constrains the possible locus of the rocket in space. Even if you know the exact orientation and location of the camera, you still need more info. A second camera judiciously located is a possibility. Or you might be able to estimate the distance from the sounds. Incidentally the best way to calibrate the camera(s) orientation is to have known reference objects in all frames @russ_watters is correct that sketches are what you need.

 

[Vanadium 50]

As the great philosopher of the past century, Lawrence P. Berra said, "In theory, there is no difference between theory and practice. In practice there is."

In theory, you have a series of pictures of a rocket over time, so you know the angles of the rocket's position, as well as the angle subtended by the rocket - i.e. it's apparent size. You have r, θ and φ from which you can calculate x, y and z. each as a function of t. Differences between two frames give you your velocity profile, and between three your acceleration profile.

However...

• You want a large field of view to capture the motion, and a small field of view to get a good picture of the rocket. Can't have both with one camera.
• If your camera is moving you need to figure out how exactly it's moving to separate its motion from the rocket's.
• There are lots and lots of calculations to be done - 30 fps x 5 seconds = 150 frames, and you're probably doing several calculations per frame. 500? 1000?
• Not all the measurements will be good. Glare, "lost in the sun", spotting where the plume ends and the nozzle begins, etc.

My colleagues who use sounding rockets film them from several distances and angles. They are not idiots. They do this for a reason. So while in theory, one camera is enough (plus some measurements with rulers) in practice you probably want more. How many more depends on what error bars you are willing to tolerate.

 

[Vanadium 50]

it goes 1100 feet and takes 3 seconds to fall

This is only possible if the rocket were upside-down accelerating earthward. (And maybe not even then)

 

[PeroK]

This is only possible if the rocket were upside-down accelerating earthward. (And maybe not even then)

Or, if it's Felix Baumgartner inside the rocket!

https://www.guinnessworldrecords.co...rst-person-to-break-sound-barrier-in-freefall

 

[Borek]

Broadly speaking this is just a "simple" geometry problems. Try to read about how photogrammetry is done, these are the same principles.

Made more difficult by the fact your setup is moving, so you need to synchronize frames. Still doable, just more work and more sources of errors.

 

[Rive]

Depending on the size (and mass) of that rocket I might rather opt for some speedometer app running on an 'onboard' phone (the rugged kind, which can survive the landing).

 

[Vanadium 50]

which can survive the landing).

If it's accelerating and covering 1100 feet in 3 seconds, it's not going to survive the landing,

You'll be lucky to recover the pieces without a shovel.

 

[Rive]

If it's accelerating and covering 1100 feet in 3 seconds

Oooopsie.

That's definitely not freefall. Let's see, let's see...

i am 15 feet from launch site and it goes 1100 feet

... I would most definitely not try that at home

 

[russ_watters]

I'd been focusing on the ascent. If OP just wants drag coefficient, then they just need terminal velocity in freefall. A stationary video camera that captures the crash is enough, if you measure the distance after the crash and the trajectory is vertical. A few captured frames, distance, frame rate and field of view is all that's needed to compute the speed.

Also, it doesn't need to be dropped from 1,100 feet. A few hundred would be plenty for an ordinary model rocket. You might even consider lifting and dropping it from 50-100 feet with a drone. If this is a normal model rocket without an engine, the terminal velocity will almost certainly be under 100 mph and will achieve terminal velocity in just a few seconds.