High altitude balloon circuitry

Thread Starter

wasutton3

Joined Nov 3, 2009
4
Hi,
I am working on a high altitude balloon project for my school and my part of the mission is communications and recovery. For the recovery aspect i am using heated nichrome wire to melt through the nylon balloon cord. To do that i need a seperate circuit (5 9v batteries). But i want to be able to control it with our sanguino microcontroller. I have set up a transistor that just uses a switch to control it, but i think there is a better way. The sanguino can only output 5v so i dont know which tranisitor to get. Does anyone have any suggestions?
TIA
 

ELECTRONERD

Joined May 26, 2009
1,147
Check out the microcontrollers datasheet and see the minimum and maximum temperature's it can withstand, and compare it to the altitude temperature it will be operating at.

Austin
 

Thread Starter

wasutton3

Joined Nov 3, 2009
4
Hey Austin,

We have already looked at that, we will be within the temperature bounds as we are including internal heaters and a healthy dose of insulation. My only real concern is the actual triggering of the mechanism.
 

spacewrench

Joined Oct 5, 2009
58
I have set up a transistor that just uses a switch to control it, but i think there is a better way. The sanguino can only output 5v so i dont know which tranisitor to get.
I'm not sure what you mean by controlling the transistor with "a switch," but you should be able to use a microcontroller GPIO line to turn on a transistor, which would then sink current from the 9V batteries to heat the nichrome wire. The microcontroller would need to put out a small amount of current to turn on the transistor, but the transistor would do the heavy lifting for heating the wire. Either an NPN transistor or a FET would probably work. Do you know how much current you need to get the wire hot? (Don't forget that you'll have to heat the wire up more at altitude.) If you provide the required current, it should be easy to pick a transistor that'll work.

[[Added:]]

Actually, you could also use an SCR to do this job, since you just need to turn the wire on once. An SCR would be easy to trigger, and would stay on without further help from the microcontroller, until the wire burned out or the batteries went dead.
 

Attachments

Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Might want to go with a logic level power MOSFET instead of an SCR. If a program "glitch" caused the I/O pin to go high momentarily and turn on the SCR, that would be the end of the flight.
 

Thread Starter

wasutton3

Joined Nov 3, 2009
4
An SCR would be a poor choice as well as the nichrome wire would continue to be hot and probably destroy the parachute. The idea of using the microcontroller would be so that when the vertical acceleration suddenly spikes, it cuts the power to the circuit saving the parachute from melting as well. The circuit operates at 9V and 4.5A so its quite a heavy load for a sangunio. I am sorry im so new at this.

SgtWookie: It takes about 5-6 seconds to melt through the nylon cord, i dont know the exact characteristics of our flight computer, but it seems to be relatively reliable
 

russ_hensel

Joined Jan 11, 2009
825
9 v batteries are not good for high current. NiCads are. Lower voltage with a nicad might work better ( assuming you stay with the hot wire ).
 

Thread Starter

wasutton3

Joined Nov 3, 2009
4
I have thought of the gillotine type but the payload gets rocked and twisted enough that that wouldnt work too well and would probably lead to premature decent. This system has been used before by other teams, but they have not been able to give me the switching method. We havnt decided completely on a battery (i know we have some rather large lithium cells around here) The average amperage is about 3A but voltage varies from 6-16V. Previous missions have been using 9v batteries to great success.
 

SgtWookie

Joined Jul 17, 2007
22,230
OK.
An IRFZ14 logic-level power MOSFET should do the trick for you:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=IRLZ14PBF-ND
Rated 10A, 60V. $1.27 each.

You could also use these: NTD4815NH
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=NTD4815NH-1G-ND
Not much stock left; obsolete design, but cheaper and smaller than the IRLZ14's.

Use a 10k resistor from the gate to the source terminal, and a 270 Ohm resistor to the uC I/O pin. Source terminal gets connected to ground. The 10k resistor keeps the gate turned off in case the I/O pin fails, or is set to input by mistake. The drain terminal will sink up to 10A current.
 

THE_RB

Joined Feb 11, 2008
5,438
You would have a more reliable system using a little RC servo to release a hook, the micro ones are only about 1" and require maybe 100mA to operate.
 

SgtWookie

Joined Jul 17, 2007
22,230
Interesting idea, RB.
Those little servos are mighty light, too.
I'd thought of something vaguely similar; a small solenoid retracting a clevis pin from a shackle, but something like that would probably have to be fabricated.
 
Top