Accelerometer acting as a switch in space

[fnumechHV]

Hello,
I am seeking advice on how to use an accelerometer as a switch on a CubeSat form factor. Incase you are unfamiliar, a cubesat is 1cm cube device designed to go into space. The problem is that the CubeSat must not have any electronics active during launch. I am seeking to use an accelerometer to act as a switch as soon as the device reaches 0 g at apogee. Any ideas?

 

[simplex]

As far as I know a CubeSat must have a Kill Switch, otherwise it will not be accepted for launch. If you really want an accelerometer aboard the picosatellite you can put one as a secondary safety device but a device that can not do anything till the primary Kill Switch is depressed when the picosat quits its launching device.

There are also other strong reasons against using an accelerometer as a kill switch. Read the explanations at this address: http://cubesat.ifastnet.com/forum/viewtopic.php?f=9&t=58

 

[gutemine]

I was thinking about such an approach for an N-sat but this would not even have the space for an accelerometer, so I dropped it

On of the possible way to be sure that the SAT is outside of the launch device when it is powered on would be using a small solar cell for switching the rest of the SAT into operational modus (you can use a capacitor loading from the solar cell to have a delaytime to the first beam of light it is facing)

But then if the SAT is ejected on the night side of the orbit this could mean waiting up to 45 minutes on a 90minute orbit until the light goes really on (and make suret hat it is a triggered circuit - meaning it stays on when darkness comes back).

Besides this only mechanical switches would work really stable - if holding/mounting is released also power is witched on.

gutemine

 

[mugaliens]

Hook anyone of a dozen very lightweight and widely available accelerometers to a timer circuit such that it would require approx 10 seconds of near zero-g (less than 0.01 g) to activate. For redundancy and reliability, I'd use three, along with a logic gate that would side with whichever two were in agreement.

Both the iPhone and the http://en.wikipedia.org/wiki/IPod_Touch" have such accelerometers. If your project is actually going into space, I'd contact Apple's PR department. They'd might be interested in giving you the accelerometers, provided you rename your project the iCube. :)

 

[simplex]

Hook anyone of a dozen very lightweight and widely available accelerometers to a timer circuit such that it would require approx 10 seconds of near zero-g (less than 0.01 g) to activate.

Most likely the cubesat will face this scenario:

Supposing the launch provider accepts you without the standard Kill Switch and allows you to use the accelerometer instead, most probable you and him would run into big troubles because the moment the accelerometer feels its first zero g, after the lift off, does not coincide with the time your cubesat is scheduled to be released in space.
For instance, if the cubesat you have built is attached to a mother spacecraft , all the accelerometers inside that big spaceship including the one belonging to your cubesat, will start to indicate zero g as soon as the mother satellite goes into orbit and begin moving around the Earth without being powered by any motor (the free fall zero g condition).
It is quite clear that your cubesat will be powered by the Accelerometer Switch while it is still attached to the mother ship. By definition such an event is not allowed to happen.
http://cubesat.ifastnet.com/forum/viewtopic.php?f=9&t=58#p106

What are you doing with that 10 seconds delay timer?

 

[mugaliens]

What are you doing with that 10 seconds delay timer?

You said it yourself: "the moment the accelerometer feels its first zero g, after the lift off, does not coincide with the time your cubesat is scheduled to be released in space."

It's to allow for one stage to stop firing while the next stage kicks in. I suppose you could simply use a timer, instead. right?

 

[simplex]

And what are you doing with the timer if the main payload of the carrier rocket, having a few cubesats attached to it (this is a common situation), is put in orbit and its engine is stopped but for some reasons the command center decide to release the cubesats two days after the insertion in orbit (the start of the free fall, zero g, state)?

Before leaving the cubesats alone in space, the orbit of the mother ship has to be fine tuned in order the picosats be placed on proper trajectories (for instance sun synchronous orbits). This takes time and a few burns at hours or days apart.

What are you doing if the upper stages of the rocket, including the mother satellite itself are spin stabilized. It is quite clear that the accelerometer inside the cubesat will never feel zero g, not even after the spinning mother ship is released from the upper stage of the carrier rocket and all the engines stops.

 

[D H]

It's to allow for one stage to stop firing while the next stage kicks in. I suppose you could simply use a timer, instead. right?

Maneuvers in space are typically of the burn-coast-burn type, with a half orbit or longer coast period. That timer had better wait for a long time.

However, how are the timer and accelerometer going to get power? A CubeSat is supposed to be not only merely dead, but really most sincerely dead, and that is from the time the "remove before flight" pin is pulled to the time the CubeSat is activated by its ejection from the vehicle. Anything other than a most sincerely dead CubeSat will require a waiver.

While the waiver form does have an electrical waiver checkbox, I strongly suspect that obtaining such waivers are very, very hard to come by. CubeSats are second class citizens. (That is, after all, part of what makes them so cheap to launch.) The launch provider is primarily worried about the multi-million dollar primary payload(s). Those CubeSats in the cheap seats had better shut up and get out of the way. Ensuring that those CubeSats are dead, dead, dead and will remain that way throughout the flight makes acceptance a lot easier.

A waiver for active electronics would probably have to accompanied with lots and lots of shielding in the CubeSat to ensure that no signals leak out. That shielding will cut down the mass and volume available in the CubeSat for the experiment.

 

[mugaliens]

This takes time and a few burns at hours or days apart.

Hence the second paragraph of my post: I suppose you could simply use a timer, instead.

 

[mugaliens]

However, how are the timer and accelerometer going to get power? A CubeSat is supposed to be not only merely dead, but really most sincerely dead, and that is from the time the "remove before flight" pin is pulled to the time the CubeSat is activated by its ejection from the vehicle. Anything other than a most sincerely dead CubeSat will require a waiver.

Solar power. So long as they're in cargo, it's dark, and they're dead. Immediately upon release, they're in sunlight, and no longer dead.

 

[simplex]

It looks like this "mugaliens" does not have a clear understanding of picosatellites and rocket carriers.
Any discussion with him about cubesats seems useless.

 

[simplex]

Solar power. So long as they're in cargo, it's dark, and they're dead. Immediately upon release, they're in sunlight, and no longer dead.

In this case you no longer need any accelerometer! The electricity generated by the solar panels (suddenly exposed to sun light) will begin to power the cubesat! Once it has power its micro-controller starts and the cubesat can do whatever it wants, including connecting the batteries to the power bus.
(As a note: a cubesat looks like http://cubesat.ifastnet.com/forum/viewtopic.php?f=11&t=54 ).