September 14, 1998 
Mentors "Transponder" Meeting Summary
Attendees:  Lars, John, Jamie, Mars, Corina, and Del

        At the beginning of the meeting, we went over all the different schemes we can communicate from picosat to earth.  The mentors thought we should just use the big dish in the hills and not worry about putting anything on OPAL.  We then talked about the various choices we had in taking advantage of sharing different parts of the OPAL comm system.  Tapping into OPAL between the OPAL onboard CPU and TNC would involve too much coding on OPAL's part, especially with Greg out of town till January.  Having a complete separate system onboard OPAL for Artemis might be too much to do with the resources we have right now.  So it was agreed that tapping into OPAL between the transceiver and the TNC would be the best bet.
        From OPAL's radio-to-TNC line, we take a wire into Artemis' black box.  In this black box, we have our own TNC and CPU.  The received signal goes from the OPAL radio to our TNC, to our CPU, then when prompted by ground, we transmit the signal through our TNC, then OPAL's radio, then their antenna.  This is good because OPAL doesn't have to worry about changing any code in their TNC to accommodate for the Artemis addition.  This scheme was deemed to have the least amount of OPAL involvement.
        We also went over how the ops would go at a pretty high level. The pico almost acts like a beacon and have two antennas (the 1m VLF antenna, and the 70cm transmitter antenna).  It either continuously transmits data collected until we have a new piece of data (if we have enough power), or transmits each data set once (this is the case of limited power), so that if OPAL doesn't get it, we lose it, and hope that there's enough out there for the science.  The data is received through OPAL's radio onboard, the Artemis black box TNC filters it and puts it into a clean text form, and the Artemis black box CPU stores the data until a command 
is sent from the ground to transmit the data down.  Lars actually offered to supply the design and the program for our black box TNC, which takes a huge chunk of the job off our hands.
        Jamie's main concern with sharing their comm system was the scenario that we might take over OPAL.  The communication channels might be saturated with picosat signal, so if we're transmitting all the time, then communications with OPAL might be lost.  The mentors thought there would be lots of fail-safe techniques to prevent this from happening.  John suggested putting an RF diode to blow a fuse when needed (??); Lars mentioned putting a capacitor in the picosat's transmitter circuit so that it discharges to prevent the picosat from transmitting all the time (??); Jamie brought out the idea of cycling power to the black box so that it can't take over OPAL, the problem with this option is that the black box might lose data in RAM; there was also the idea of putting a logic gate at the junction of shared lines so that OPAL can always disable the black box to ensure OPAL's priority to use their radio.
        The mentors also said that the Artemis black box TNC and CPU could be one unit, possibly the 68HC11 with about 64K (advantage: we've already used this), or a simple PIC with serial RAM.  We would need to use full AX.25, and the next meeting should address modulation schemes.