Last night we worked on the wiring for the bicycle wheel – with a focus on robustness. We want to be able to throw it in the rack on a train and not worry about any of the wiring breaking – and we don’t want anything shorting out on all those metal spokes! And we have a bit of a chicken-and-egg problem – we don’t want to machine sew around the wires, but we don’t want to wire into some complex object, either. Our solution – undocumented by photographs as usual – involves quilt batting and beading with our old standby, heat-shrink plastic.
I’m just slow with the reminder, and you’ll have to treat me a bit gingerly – I’m still recovering from an attack of the something-or-others. Entrance through the basement, 7:30, City of Edinburgh Methodist, and no doubt a prompt finish at 9:30.
Progress on the bicycle wheel – to the level of a final design and fabric choices for the fascia, and pseudocode for the Arduino firmware – plus homework that will probably lead to final assembly at the next session and our first field tests of a mid-size sensor network.
It’s still the soft fruit season (just!), and the Festival, so we’re still in laidback mode for tonight. Forward planning will resume soon, honest! And here’s some advance notice: anyone who fancies it, please join me on Doors Open Day (26 September), at Christ Church Morningside in Holy Corner.
I think I’ve finally figured out why our equipment performs flawlessly in test situations, and then goes down and stays down in our real target buildings. It’s something about the firmware on the wireless dongle (the EDIMAX EW-7811UN). In a nutshell: when the wireless signal strength is low enough, then the networking releases the DHCP address and disassociates the wireless connection from the dongle. I don’t know yet whether it checks the signal strength periodically to try to reconnect, because even if the dongle is in an area with good signal, it still reads the strength as zero. Continue reading I think we’ve cracked the wireless…