scruffy days

The shows get better and better every year. This one featured a slew of eco-minded projects: BioBronc, Solar Jewelry, Solar Time (pictured here with my parents ), BikeJus, Device Power Monitoring, Blue Phoenix, and the debut of Sustainable ITP - a showcase of student and faculty work in the program’s sustainable practices initiative.

Solar Time, a project with Gilad Lotan, is a meter displaying the total amount of energy in watts available from ITP’s 80-watt solar panel located on the Tisch building’s roof. Readings are retrieved via a wireless radio network and logged in a database.

Also in the show was Under The Level, with Catherine Colman, a mapping project of New Orleans’ post-Katrina destruction on the streets of New York. More details at underthelevel.org.

A simple metering interface to increase awareness of the solar energy available in relation to daily electricity consumption. [some documentation to date]

hooray! my geranium bloomed!

Concept: Solar powered Bluetooth modem for recording environmental data (temperature, air quality, etc.) for broadcast to your mobile phone. Microcontroller would still require separate power source, but blips of power could supply enough & trigger the Bluetooth function in this type of application.

———————————————————————————–

I attended the Eyebeam workshop last summer on DIY energy and built a basic solar engine. For this project, I wanted to learn more about how these BEAM bot components store/release energy and apply that to something beyond the bot design.
SOLAR ENGINE
I tried another take on the solar engine, with an LED as load. I also tested this with a mini vibrating motor I had around, but required too much current so it would never power (1.5-3VDC, 62mA). I thought I could try out some small fan motors, but these required even higher current - one at 0.06A and the other 0.22A (both at 5VDC).

Measurements:
Voltage - max of ~2.6V
Current - max of 5.8mA
Capacity in joules - 5.8mA * 1 hour * 2.6V = 54.28800 Joules
Power consumption in watts - 5.8mA * 2.6V = 0.01508W

Energy storage method:
1 capacitor, 4700uf 6.3V
8212 ENGINE / PM3 POWER MODULE
My next attempt was the 8212 solar engine. This uses hysteresis for power discharge, which is triggered at certain voltage levels, dependent upon the combination of resistors used with the Maxim 8212 chip.

Highlighted below are values I tried for R2 (trigger) & R3 (hysteresis). Didn’t work out though.

Currently, for another class, I’m working with a BlueSMiRF to receive and broadcast data. This can be powered with 3-6VDC, 25mA average. You can reduce stand-by consumption to 2mA with an AT command, but so far the modem’s only responding with error messages. I wanted to try and see if it was possible to periodically provide the modem with bursts of power so it could possibly broadcast a bit of data to a local computer or mobile phone. Measurements of the solar cell under a desk lamp (26W CFL): 6.4V, up to 10.4mA. Datasheet also rates it capable of 20mA in a short circuit setup.

Below is my first attempt at the circuit, using 100kOhm resistors for R1, R2, R3 & a wire jumper for R4. This didn’t work out as the configuration of resistors was incorrect, at these values according the chart above would trigger at values too low.

Next attempts below, changing the resistor values. R1 was still wrong, and Jeff F. recommended setting trigger value higher. Second image below uses these values:

R1 - 100k, R2 - 400k (trigger at 6V), R3 - 244k (hysteresis at ~3V), delta-V of 3V

Still didn’t work though, as the voltage remained stagnant at around 5 volts.

Referred to Jeff LeBlanc’s layout of the same circuit, and made these changes to the resistors, now seeing a change in voltage. Toggles between 3.6 - 4.5V. Problem is that the TR3904 isn’t discharging that voltage though, and unsure as to why.

Measurements - need to calculate when load issue is resolved:
Voltage - triggers around 4.5, down to 3.6V and back up again
Current - load not releasing
Capacity in joules - (Amps*Time*Volts = Joules)
Power consumption in watts - (Watts=Volts*Amps)
Energy storage method:
2x 4700uf 6.3V capacitors
(Also tested with a 1F 5V cap, but this significantly changes the circuit’s behavior. The climb in voltage is much, much slower…)
Where to go from here
Figure out problem of load issue in PM3. Research other BEAM circuits that maybe better suited to supply power for Bluetooth broadcast application. As I continue with the Bluetooth project, I’ll record consumption rates for specific functions.

(Response to Urban Computing entry: 05. Crossroads)

“What happens when two or more city streets intersect?”

I often think of the rules embedded within street intersections. Painted stripes dictating lanes and crosswalks, traffic lights and signage all orchestrate this flow of movement between people and autos. Sometimes I think it’s crazy that these simple colors and symbols dictate go, slow down/hurry, stop - that just our understanding of these codes and agreement to comply keep all hell from breaking loose. Kate Ascher breaks down the digital network of NYC’s traffic signals in The Works, showing how much more complex the system really is. Her diagram (2nd image below) shows the computers, video cameras, timing set for local conditions, and sensors at work.

It’s that richness and density of people, information, cars, things and events which “stand as synecdoches for an entire city…” It’s these intersections that are so full of goings-on, that they have to be understood, regulated, and calculated to some degree so planning can attempt to keep things running smoothly. You wouldn’t need as complex a system for a residential, suburban intersection.

:D  ha ha ha

Finally, I’m gonna take a moment and blog something besides Xbee code and alternative energy. Ahhhh… Almost forgot Suzanne had gotten tix for This American Life’s live tour - the show was excellent. Actually, I didn’t really realize when we got these tickets that they were for the launch of the t.v. show. I was skeptical like most of their decision to test the waters outside of radio, but tonight’s show was evidence of how cautious they were in approaching the whole thing. They opened with an animation telling the story of how some elementary school kids got caught up in a trend of playing with fake cameras made from cardboard/toilet paper tubes. This was executed so well - a compelling use of visuals for storytelling, and such a good intro for the show’s transition to moving images.

It was amazing to see Ira Glass in the flesh, working his radio magic on stage, and being able to associate his voice with physical mannerisms. There was discussion with the show’s director about delivering the show in a completely different medium - how terrified they were of a “sucky show”, avoiding the reality show angle, and using carefully crafted cinematography. The clips they showed still carried that distinct TAL essence - I’m looking forward to it. All this plus awesome performances from Sarah Vowell, Jonathan Goldstein, Dan Savage and Mates of State.

1 9V battery (Duracell) lasts approximately 1 week. By increasing sleep time, battery life for 8th floor circuit should improve, hopefully significantly. Reset pin 5, take low by dipping wire to ground.

XCTU Settings:
SP setting: 214 (hex = 532/60 = 8 sec)
SP is for setting/reading sleep period for cyclic sleeping remotes. Max is 268 seconds. Since we want to sleep every min instead of 8 sec, 6000ms = 1770 hex as new value.

Hmmm. This change alone didn’t work…
Seems ST (Time before sleep) needs to be increased also. When testing with Processing script, no readings appeared. Was 30 (48ms), then tried 150 (336ms). But now, no chance to sleep.

Next, tested these new settings together:
ST 54
SP 1770
GT 54
Results: It works! Yay! Every minute, reading is sent through…

(Response to Urban Computing entry: 03. Door) 

Doors got me thinking of Donald Norman’s The Design of Everyday Things where he discusses the affordances of doors, and what they signal users to do with them. He cites an example of when that fails with a picture of a little boy who has to open a cabinet door without any handles by tying a string to it.

“We expect to find some visible signal for the correct operation. This tells us where to act. The next step is to figure out how: we must determine what operations are permitted, in part, using the affordances, in part guided by constraints.”

Doors are necessary; we need entrances and we need exits. Sometimes we need doors layered on top of doors, to feel safer. This is how I grew up, and though we know that locks can be picked, having layers provides some mental level of added security. And then sometimes we even need doors within doors. I think also of the inadequate doors that we put up with on a daily basis. For example, how certain locks become slightly dysfunctional and we learn the tricks needed to manipulate them.

So where have we failed to accomplish the proper affordances in digital spaces? Or where do we have “doors” that we just put up with because their design is not within our reach to change? I’m sure we’ve all run into sites or use applications that frustrate us because it’s not obvious how to find some information we need or how to accomplish a simple task. Visual or sensual cues are so important - more important than the aesthetics of an object, tool or space. Doors are an introduction to what lies ahead.

To respond to the question posed in the reading, “What might be communicated by the difference between an open door and a closed one?” Catherine, Corrine and Alex point out that first and foremost, the door must be seen and that doors are socially constructed. In 1960, J.C.R. Licklider wrote in The Computer as a Communications Device:

“For the society, the impact will be good or bad, depending mainly on the question: Will “to be on line” be a privilege or a right? If only a favored segment of the population gets a chance to enjoy the advantage of “intelligence amplification,” the network may exaggerate the discontinuity in the spectrum of intellectual opportunity.”

I think that open source and ecologically minded design are major components to progressing accessibility to these opportunities.

Megan & I decided to use the playfulness and motion of the salad spinner as a children’s toy. One Kid Band is a portable, modular toy that encourages individual expression and musical exploration. Kids can customize their own LED light displays, attach speakers and other interfaces and power them through their own physical activity.

Features:
- Creative play
- Learning by making
- Enables imaginative solutions

Before OKB, we attempted battery charging:
100 tugs ranging from 10-20V charged up AA battery 0.02V
We tested this twice, and voltage on battery rose 0.02V each time. Scenarios this might be implemented: gym, possibly physical therapy. But the salad spinner as a charger alone wouldn’t be very practical - it would be best to integrate with systems that are already utilizing a similar action and allow the charging to take place over time.