I know that people have discussed this before, but I have yet to see anyone declare that they built a worthwhile system or create any documentation. I did see some youtube videos using pyro charges, but that makes me nervous. I want to use CO2. If anyone has any information or observations I would love to hear it as I start this project. I am throwing all my ideas out there to see what people would do differently, or to find out if it is a pipe dream (pun intended). So here we go!
Goal:
Develop a lightweight and independent failsafe system for multicopters.
Idea Summary:
Use an Arduino micro with an accelerometer to trigger the ejection of a parachute. The system will have to be independent of all other power and mechanical systems on board so it will have it's own power source.
Detail:
This would be primarily designed for multicopters carrying a large and expensive payload like a DSLR. The goal is to slow the copter down enough so that the payload is less likely to be damaged if an emergency occurs. The independent power source must be small and light weight.
I researched different methods to see if anyone had tried to worked with Arduino and pressurized CO2. Unfortunately, there were only unanswered questions or some hypothesis with no follow up. While brainstorming I thought about paintball and thought that a solution might exist there since they use electric triggers in many of the markers.
I picked the CO2 system from the Tippman Tipx pistol because it was built around CO2 cartridges, instead of a refillable bottle system. I think I can use the puncture valve assembly from this marker to interface with the CO2 cartridge. I need to figure out if the puncture valve assembly will actually act as a valve, or if it only provides the breach. I also need to find out if the an e-trigger solenoid provides enough force to puncture the cartridge, or if it has to be manually punctured.
The Solenoid will be triggered by an e-trigger assembly, this assembly will be hooked up to an arduino micro with an accelerometer. The accelerometer will trigger the solenoid when either the multicopter tips past a certain point or the rate of descent reaches a specified velocity.
The gas from the CO2 cartridge will travel through a pipe into the adjacent tube that is packed with wadding and a parachute. I will need to run some tests to see how big the chute needs to be, but I am sure it will be significant.
Reference:
Tipx Exploded Diagram: http://www.tippmann.com/pdfs/manuals/TiPX%20Schematic%208-11.pdf
Valve Puncture Assembly: https://www.pbsports.com/Tippmann-TiPX-Puncture-Valve-Complete.html
Solenoid & e-trigger: https://www.pbsports.com/Virtue-Redefined-Tippmann-A5-Upgrade-Board.html
Arduino Micro: https://www.adafruit.com/products/1086
Accelerometer: https://www.adafruit.com/products/1018
Here is a really rough draft of what I am thinking:
So lets hear what you think!
Replies
Parachutists, as of 20 years ago, when I jumped a few times, used two different systems for releasing the chute manually (free-fall jumps, not static-lin jumps).
One is a smallish chute, which is kept in a piece of tubing, usually on the right hip/outer thigh
The other is a steel-wire with a handle at the left shoulder - this is the one you always see in movies.
The steel wire pulls out a pin, releasing a spring, which ejects the smallish chute, which, in turn, pulls the real parachute out.
I wonder if a spring-loaded system couldn't do the trick - and at least weigh a lot less? It would definitely cost less.
Maybe some sort of mouse-trap trigger could release the spring, which ejects the parachute; so a small, not so powerful servo could "pull the handle" ?
A question: are you making some sort of free-fall detection in the "Arduino & E-trigger" blue box?
EDIT: in any case, a parachute has a minimum height, from where it will work, as it takes time to unfold. Depending on the job, of course, I would much more worry about "failures" that occur under the minimum height of the parachute, but still high enough to cause destruction to sensitive payloads. This makes me wonder, if some sort of leightweight (foam?) protection around the payload wouldn't do the job better, overall?