I'm building a model rocket and payload system to collect some data and I am trying to figure out how to use an accelerometer to determine when the engine burns out in order to fire a parachute ejection charge. The accelerometer I am using is the Memsic 2125.
I plan on mounting it flat on the upper region of the electronics bay. Other than streaming the data and logging it to an SD card, I am not quite sure how to program it to fire the charge.
I'm thinking that it will have to be an If/Then command but I really have no idea what the numerical values would be.
You could watch for a drop in acceleration indicating the engine has stopped and start a timer to deploy the parachute.
You don't to deploy the parachute right when the engine stops. The rocket will likely coast upward for a bit.
You really want to detect apogee which you can't do with a single accelerometer by itself. You might be able to detect the rocket's rotation at apogee if you used a gyro.
A sensor like the MPU6050 only costs about $3 and should be able to detect apogee (I think). Another option would be to use a magnetometer and watch for the change in tilt of the magnetic field.
I'm building a model rocket and payload system to collect some data and I am trying to figure out how to use an accelerometer to determine when the engine burns out in order to fire a parachute ejection charge.
In the old days a small section of fuse would be used (lots of weight saving for other payloads). If your rocket is in the amateur size range, a tilt sensor or gyro might detect when the rocket starts to roll over.
zoomkat:
a tilt sensor or gyro might detect when the rocket starts to roll over.
I think a gyro could detect rollover but I don't see how a tilt sensor could. The rocket will be close to "free fall" conditions from the time the rocket engine burns out until the parachute is deployed.
If you were to throw an accelerometer data logger into the air, it would register free fall from the time it left your hand until it hit the ground. I don't think there are any "normal" tilt sensor which work in free fall.
A barometer could work. The aerodynamics on the nose cone could affect the reading. A small hole in the leading edge would create a high pressure. When the pressure drops, start a timer sequence to deploy the chute. The lowest pressure could also be recorded for calculations of altitude. You could also record temperature at the same time. I think a combination of all of the above posts would be cool.
I have a friend that had a rocket company and went to a lot of LDR meets (Large and Dangerous Rockets) and he built acquisition modules for his rockets. I think they had one that went about 13 or 15 miles and lost it. Law enforcement brought it back for them.
These are all very interesting approaches. However, I'm not familiar with magnetometers or tilt switches. The barometric pressure sensor on the nose cone idea is very unique. The payload section of the rocket is actually equipped with a barometric pressure/temp sensor in order to determine altitude throughout the flight. I'll have to look into incorporating a gyro as well.
Would it feasible and reliable to use only an accelerometer to detect engine burnout? As of right now I have the electronics bay assembled and was banking on using only the accelerometer for this task. However, if you don't this this is a very reliable method, would a countdown timer increase the certainty? The engine burns for about 8-10 seconds during static tests. Coast phase is projected to last an additional 6-8 seconds. Do you think an 18 second timer would do the job?
The accelerometer will detect engine burnout.
Apogee is another problem.
As was mentioned before, after burn out, it is basically in free fall, until
it hits the ground.
It will reach some terminal velocity while falling and 1G will be seen
once it is at a constant speed.
This is controlled only by wind drag. Of course, you are designing the
rocket to have minimum drag so by the time you see 1 G, you might be traveling
too fast to deploy a parachute.A sensitive enough accelerometer can see
the change to minus but it is hard to get it to also work at 10+ G.
Dwight
A pitot tube would give you both an altitude measuring point and an
air velocity, with two pressure sensors.
One for static and one for differential.
Dwight
I'm not familiar with magnetometers or tilt switches.
You might research amateur rocketry from simpler times. A tilt switch I've seen described was a small metal ball chain hanging thru a metal ring, with a camera flash bulb and battery connected. Back then the the rockets had tail fins which had aerodynamic drag and resulted in the rocket nosing over and down at apogee, causing the ball chain to complete the flash circuit. These types of flights were made at night with long exposure film to track the flight.
...and unless you have at least a minor in math you might not like the answer
Wow...that was very extensive. Nonetheless, it answers my question. I might have to go back and get my master's in math to wrap my brain around that one :o
The ball chain and flashbulb idea is actually a really good one as well. You all have been very helpful with your responses. I am very much appreciative!
You guys put a lot of faith in the results of a Kalman filter.
I don't think they are much use for such a purpose.
I didn't read the article but do have some understanding
of a Kalman filter.
Dwight
