Monday, October 04, 2010

GAGA-1: Computer mounting

Not much time to work on GAGA-1 this weekend, so I nailed down one of the simpler items: how the two computers (flight and recovery) are going to be mounted inside the capsule. After considering all sorts of techniques I settled on self-adhesive Velcro "coins" and a cut up fast food container (the plastic containers that much Indian food comes in have nice fitting lids and are made of strong plastic: I use them for storing screws, components and all the miscellaneous stuff in my basement).

Here's a picture of an untouched curry container and once I cut up. The two black dots are Velcro pads on I stuck on the bottom. The container is just the right width to fit in the box and the small vertical portion pushes against the edges of the box keeping it snug. Between that and the Velcro this will provide adequate support for the two computers.

With the two computers sitting on it, it's clear that there's adequate space. The computers will be attached using screws and risers that I have left over from an ancient robot project.

Finally, here's a shot inside the capsule itself showing the Velcro pads that are stuck directly to the polystyrene.

The other thing I spent time on was deciding on interfacing for the Telit GSM862 module. I toyed with the idea of buying a SparkFun USB Evaluation Board to make the module easy to program, but the price put me off. It's very expensive just to get a USB interface.

Since the module will be flown attached to a simple breakout board (see the photo above), I decided that I'll interface directly to that board for programming using a USB TTL converter cable. That way I can program the device in place... cheaply. The breakout board will only have three additional things attached to it: power, an LED on the STAT_LED pin, and a simple switch on the ON/OFF pin.

Note that in the circuit diagram for the SparkFun GSM862 Basic Evaluation Board (the breakout board I'm using), it's not obvious that all four of the VCC connections on the Telit module are connected together, and both the GND connections are connected together. This is super convenient because it means only one of each needs to be connected to the battery, but the SparkFun circuit diagram is incorrect in showing six separate connections. A quick test with a multimeter and a look at the PCB tracks verifies the truth.

No comments: