EEGuy

Not a networking guy here, and this is nowhere near all you're asking for, but here's a classic GUI-controlled router setup which allowed me to totally block my son's internet access when needed. Sounds like your TV needs that too:

* Router was a Linksys WRT54GL [1] re-flashed with Tomato [2] firmware

IIRC it took two steps to stabilize things:

* One of the menu options [sorry, don't recall which one] allows watching the MAC and IP addresses of connected devices in real time as they come and go. You can somehow assign names to devices in this list to help sort them out. Unhook or power off off everything else or use the OUI lookup or somehow otherwise identify the TV. Click on 'static' so as to force Tomato's DHCP service to make the current IP address fixed for the TV's MAC.

* Now that the TV has a static IP on your LAN, you can use Tomato's 'Access Restriction' on that IP to disallow all outside access. Works as well for restricting one's kids' access to reasonable time ranges -- and cutting them off when necessary ;).

--------

[1] http://www.newegg.com/Product/Product.aspx?Item=N82E16833124...

[2] http://www.polarcloud.com/tomato

I am a net-metering solar PV owner who's not averse to a reasonable, independently audited "connection charge".

But I would be adverse to having to pay via a "connection charge" for what looks to me like a failed uprating of a nuclear plant, if I were I a Southern California Edison customer.

The IBP (tiered rating) isn't a birthright of utilities, geez. It's a conservation measure!

Whether I shave off my higher-tiered consumption by changing my thermostat, changing my lighting or by putting in solar, that lost high-profit consumption isn't the company's property, it's mine to do something else with.

Hmmm.... and all that waste heat from the Peltier junctions go into /heating/ the room -- making it a little warmer for everyone else.

mapt, this. Maybe I should do the math, but I can't imagine magnetic levitation (not propulsion) running off on-board batteries.

This, svantana.

My dentist's ultrasonic dental plaque removing tool produces audible squeals reminiscent of analog tuning of an AM radio. That's one possible example of probable nonlinearities (in my gums an teeth) producing detectible audible power from ultrasonic energy transmission. No reference -- it's just what I hear every time I have my teeth cleaned.

Granted an ultrasonic water stream (dental application) has different physical characteristics than an ultrasonic air pressure wave (ultrasound).

Saw this just today. Interesting how the company is spreading the risk of this hardware project by:

* Outsourcing-and-(partly?)-open-sourcing the manufacturing (and probably all of the customer service) to anyone who'll sign up via their "100k Garages" website

* Open sourcing the API

* Soliciting app developers to build job-specific smartphone apps to send a CNC data stream to the hand-held CNC router table.

Seems like the target end user market would be carpenters and contractors who do custom high-end work on-site.

There isn't much useful "work" one can do with a microwatt at "human" scale in a day. Except, as this paper shows, build a networkable wireless system camping on those microwatts.

Not meaning to be critical, rather just want to clarify that our modern-day gigawatt-scale energy grid yields an enormous 15 orders of magnitude more electrical power than the microwatt-scale "harvesting" of TV signals can. That's an enormous power spectrum, shown to have a newly useful application (batteryless mesh networking) at microwatt scale.

Agreed, StandardFuture. This is impressive, but TV channels and signals are owned by TV broadcasters. So:

* Not only the FCC, but the NAB [National Association of Broadcasters] will be on this pronto.

* So, consider either:

___o Allowing the FCC to retain a channel for a few years of ambient backscatter experimentation and evaluation (funding a set of transmitters for a while, experimenting with what transmitter signal structure might allow wider ambient backscatter bandwidth and node-to-node distances), then offer it as a tax-supported service ( as with GPS ), or sell it, else

___o Having the FCC allow an interested private owner of TV channels privately experiment with ambient backscatter engineering as proposed (locally, multipoint-piggybacked over 8VSB TV), develop the technology, then license it.

* In no way am I intending to cast aspersions on the authors' achievements here; instead I am very, very impressed! It shows indefinitely durable (battery-less) RF networking having practical applications drawing only half a microwatt for 'transmission,' carrying a general purpose programmable microcontroller on board, using a passive method to modulate an ambient RF signal with a man-made signal below the by the intended (TV) signal. And using analog components where they give significant power advantages, while retaining programmability sufficient to support a networking stack. Wow.

Disclosure: I work in support of the Broadcast TV Industry

Edits: (1) Tax-supported is not 'free' (2) Clarify that ambient backscatter is multipoint, 12 orders-of-magnitude lower power, and distributed, in contrast to the single point of origin of the 8VSB TV transmission

In terms of energy delivery:

  ( 85 KWhr / 90 sec) * ( 3600 sec / hr ) = 3400 KW = 3.4 MW

3.4 megawatts, not bad.

I'm intrigued by Dr. Post's statement regarding electrostatic rather than electromagnetic force.

When I think of the "motor-generator" component of a flywheel-based energy storage device, coils and magnetic poles and all that electromagnetic to/from mechanical energy conversion stuff comes to my mind.

Electrostatic to/from mechanical energy conversion is known [1] [2], but not to my knowledge known for sustained high power and energy density appropriate for practical flywheel-based energy storage.

Intriguing to consider that Dr. Post's research might be going in such a non-traditional direction as electrostatic motor-generators. Perhaps there are some incremental efficiency gains or design simplifications to be had in using the flat plates, good insulation and high voltages characteristic of electrostatics instead of the big coils, cores and high currents of electromagnetics.

That said, there's no free lunch: No matter which electrical-to-mechanical-to-electrical energy conversion mode used may be, the energy stored remains rotational kinetic energy, thus limited by the same critical physical limitations (failure modes, maximal energy density, maximal power density) of traditional (electromagnetic, to date) flywheels.

--------

[1] https://en.wikipedia.org/wiki/Electrostatic_motor

[2] https://en.wikipedia.org/wiki/Electrostatic_generator

>Why is it 7-bit and not 8-bit?

To all the other history cited, there was also an early digital telephony multiplexing ("T1") practice of the day called "robbed bit signalling"[1] [2]. You got 7 bits clean, 8 bits - not so much.

----------

[1] https://en.wikipedia.org/wiki/Robbed-bit_signaling

[2] https://en.wikipedia.org/wiki/8-bit_clean

Rooftoop solar is good for a mere 8 Watts / sqft [1] [2] when the sun's shining, and it's not a constant 8 Watts output over the day even if the weather is clear all day.

So repurposing a Sears building's rooftop area for solar power would not cover even the minimum requirements you cite (300-400 W/sqft -- x 24hrs/day) as necessary for a datacenter re-purpose.

But then there is that big, unused parking lot. If it's got, say, 30x the rooftop's area and can be fenced in, maybe then a ground-mounted solar system might supply some significant fraction of daytime energy.

------

[1] http://www.ehvacdesign.com/solar_system_design.htm

[2] http://www.solar-estimate.org/?page=solar-calculations

Thank you kken for your clarifications.

I don't know the physics modelling switching devices or arcs, but I do recall reading long ago that once an arc starts, its resistance drops dramatically, continuing to drop as more current flows through it (within limits).

So an arc becomes a runaway situation if current flow isn't ballasted or elsewhere interrupted. But if current flow is controlled, an arc has practical application in fluorescent lighting and arc welding.

Until recent news such as this, I didn't know it was the "conceptually simple" but practically difficult ability to switch (interrupt) a GW or so of DC energy flow preventing the implementation of better than point-to-point HVDC (i.e. multiple access point) power grids.

Evidently interruption's easier with AC because of the zero-crossings of current every half-cycle.

Three alternate solutions to HVDC circuit interruption are described in this article. Two are electronically integrated with the output circuitry of the bidirectional AC-DC converters (bidirectional energy flow AC-DC converters being an interesting, but off-topic subject in their own right).

I find it an ingenious use of an IGBT bridge surrounding a capacitor to throw the cap's charge into reversed polarity so as to stop the current flow -- effectively opening the circuit. This tells me the cap would seem to be in series with the DC side somehow. Were it in parallel, well, it wouldn't make much sense to reverse the cap's polarity as you'd only get to do it once.

I also find it fascinating that the mathematics it takes to describe AC circuits is far more complex than for DC -- one has to accept that there genuinely exists a square root of negative one, "imaginary" though we may call it.

As in programming, good nomenclature is often late to the table. I wouldn't try (and I'm certainly not qualified) to "refactor" centuries, maybe millennia [1] of mathematical terminology.

-----

[1] https://en.wikipedia.org/wiki/Imaginary_number

LA's MTA has a bike locker rental program, find out here [1]. It's worked for me; stash a beater bike at the "work end" of the train commute and ride the "last mile".

If the "work end" is in Burbank, here's the info on their bike locker rental program [2].

There may be a wait list.

-------------

[1] http://www.metro.net/riding_metro/bikes/images/locker_rental...

[2] http://www.burbankca.org/transportation/resources_lockers/lo...

* The 1897 Pasadena-to-LA Cycleway looks appealing to this native Angelino [perhaps in part because it's empty in all the photos?], but had it been completed and survived, I would it to be heavily subcribed now.

* Some related thoughts:

___ o My favorite LA commute replaced a 80-minute auto commute with a 90-minute total { auto + Metrolink commuter rail + bike locker + 1.6 mile bike ride } commute. By driving to the "home end" and renting a bike locker (where I kept a bike in storage overnight) at the "work end" of the commute, I solved the "last mile problem". And I got 20 minutes of reading time each way on commuter rail.

____ o Folding bikes [1] solve the "last mile" problem given that (within certain restrictions) they can be taken on-board folded on LA busses [2]. Having put over 8,000 miles on three different folding bikes, I have found them to be practical but at-times finicky.

____ o LA is getting a bike share program under beta rollout in "LA-adjacent"-Anaheim [3]. These bikes use driveshafts (instead of chains) for cleanliness.

____ o Adapting a mountain bike or hybrid bike using an "Xtracycle" attachment [4] creates a long wheelbase cargo bike cable of carrying a week's groceries for a family of four. I have over 2500 miles on one of these. It won't go on public transportation nor on the bike rack of a car, but it can carry up to 300 pounds (if the hills aren't too steep and the load is well-distributed).

____ o My present LA commute is 5 miles, suburban all the way, and takes 30 to 35 min by bike, independent of auto traffic. Taking my auto instead saves 10 to 15 minutes, depending on traffic.

-------

[1] https://en.wikipedia.org/wiki/Folding_bicycle

[2] http://www.metro.net/bikes/bikes-metro/

[3] http://la.streetsblog.org/category/issues/bike-sharing/

[4] http://www.xtracycle.com/

180,000 cyclists and peds did get out of their cars to walk or bike on the most recent "CicLAvia" event [1] [2].

------

[1] https://billrosendahl.wordpress.com/2013/04/23/ciclavia-to-t...

[2] http://www.ciclavia.org/about/

Agreed mturmon; an open-air LA-freeway-adjacent bike path poses many technical, aesthetic, scale-of-access, cost and possible health issues.

I do not think a freeway-adjacent bikeway would be popular with cyclists if open-air. LA has a freeway-adjacent bus rapid transit line (the "Sliver Line") which isn't very popular at 12K avg boardings in March 2013, compared to the far more successful rail-route-replacing "Orange Line" at 30K+ avg weekday boardings for the same month.

The Orange Line has a parallel bike lane which is popular with cyclists and pedestrians, even into the evening as it's well-lit and well-landscaped.

Edit: Reference for LA Metro system boardings: http://www.metro.net/news/ridership-statistics/

Some quick research on generator excitation:

* Each of Hoover Dam's generators appear to me, based on this photo [1], to have three stages of excitation, vertically stacked on the rotor shaft.

* A year 2010 upgrade to digital controls at Hoover Dam [2] shows:

___ o A total plant capacity of 2.074 GW from the 17 generators

___ o 20 seconds to go from reactive power control ("condense mode") to generator mode.

___ o Pictures of the original and replacement governor equipment; the red cabinetry shows a beautiful 1930s design style

* In "condense mode" [3], the turbine's penstock valve ("wicket gate") is closed off, the turbine volume filled with compressed air to displace the remaining water, then the generator operated as a massive synchronous motor. In this mode, the excitation controls can provide marketable power factor correction, aka "reactive power" for the grid.

------

[1] http://bayimages.net/view-photos/electric-power-generators-h...

[2] http://www.hydroworld.com/articles/hr/print/volume-29/issue-...

[3] http://www.bpa.gov/Finance/RateCases/BP-14RateAdjustmentProc...

Something not qualified numerically about LED or CFL lamps, but a moderate consideration for me is analog FM radio interference ("RFI").

In one of the smallest rooms of my house, there are two of the Philips LED bulbs of the "yellow when off" variety. In an adjoining area there is another bulb of the same type.

There is audible RFI with the room's lights on when the battery-powered FM radio (with a rod antenna, about 3 feet from one of the bulbs) is accurately tuned in to a favorite but moderately distant (perhaps 30 miles), moderately powered (6 KW, not a lot) FM station. Moving the rod antenna away can stop the RFI, but then the antenna happens to block a doorway.

Turning off the lighting stops the RFI. The lamp in the adjoining room is apparently distant enough not to interfere with the FM radio.

I wonder if the CREE bulbs are any better in this regard, but not seeing anything in the teardown pictures that looks like shielding or ferrite chokes makes means I'll probably keep the existing bulbs; the RFI isn't bad enough.

Chandeliers with prismatic glass are another problem area for LED lighting. My home has two such fixtures, previously populated with nine 40-Watt incandescent lamps each.

That's a lot of power dissipation (360 Watts), particularly since one of the fixtures was on up to 12 hours per day (4KW-h almost daily for that).

So I replaced all nine bulbs with TCP LDCT3WH30KCC 3-Watt (15 W incandescent luminance equivalent) 3000-K 80-CRI bulbs. 27 Watts, down from 360.

Excellent energy savings, excellent dimmability (better than the incandescents), but all those beautiful violet, green, blue, gold and other various facets of color from the prismatic glass were just gone -- collapsed down to a cold white; as if the glass weren't prismatic anymore.

I do not understand why I don't see at least some narrow-band stripes of some few individual colors.

So I put back just two 40-watt incandescent bulbs. That was enough to bring back a nice refractive color spread from the prismatic glass. The dissipation is up to 91 Watts from 27 Watts, but it's beautiful again.

Energy problem resolved satisfaction (75% reduction), while keeping the aesthetics alive too.

* Here [1] is a position-paper proposing an HVDC pan-European "supergrid" of up to 4KM distance, including undersea HVDC, overlaying and working with the existing, less capable AC grids. No magic, no HT superconductors, just recent evolution of existing HVDC technology which will enable a "meshed" HVDC grid (not just isolated HVDC point-to-point links). It elementally observes that while high cable capacitance limits undersea cable AC transmission distance, it doesn't for HVDC; DC doesn't get through capacitance once charged, whilst "reactive" (capacitive and inductive) effects limit how much power can be extracted at the other end of a "long" transmission link (sorry, no reference right now on that). Avoiding reactive transmission losses is one of the reasons AC power distribution is optimal for moderate grid and distribution scale at its low frequencies of 60 Hz or 50 Hz. HV avoids ohmic losses, while low frequency reduces reactive losses up to the limits of some estimable built-out scale. Going to DC allows scaling out to longer distances than AC, even as ohmic losses are minimized by using HV (ohmic losses are elementally the same, whether AC or DC).

* Here [2] is a quick summary of recent improvements in HVDC transmission and high power "VSC" inverter techniques, citing [1], summarizing: "...there is now no doubt as to the feasibility of HVDC networks ferrying renewable energy resources from wherever they are in surplus to wherever they are needed."

[1] http://mainstream-downloads.opendebate.co.uk/downloads/WG2_R...

[2] http://spectrum.ieee.org/energywise/energy/the-smarter-grid/...

Agreed, stunning it is that so common and reasonably safe a thing as a self-serve gas pump is a 22 MW chemical energy delivery system!

Some reflections:

* Perhaps an EV might be 10x as efficient with its input energy as a gasoline powered car. Perhaps 220 KW would do.

* Perhaps a robotically mobile, rather than a counterweighted, manually lifted and connected cable is in order. The present day robotic car wash design where the car stays still while the motile, rotating washing machinery moves around it comes to mind.

* Electrical connectors give me more trouble than wires and cables. A replaceable, quick-disconnect connector capable of 20KA seems more of a design challenge than the cables that would go with it.

* High voltage insulation and wiring is "easier" than high current conductance (low resistance) wiring, for example: Common engineering practice with 30-volt solar panels is to wire them in series so that the total string voltage is 300 to 600 VDC at 8 Amps, rather than in parallel so that the current would be 80 to 160 Amps at 30 Volts. Air is a lousy and unsafe insulator, but there are good liquid insulators which could easily, temporarily fill a two-pole, ground-surround high-voltage connector (displacing the air) before it's allowed to be electrically energized and ramped up to a high voltage. Precise ground fault current detection could make it safe from fault currents (i.e. shocks & shorts) in the milliamp range even as the cable carries hundreds or thousands of Amps. But any HV supply requires up-conversion at the station and down-conversion in the EV; whilst these can be 95% efficient, the effect of each conversion inefficiency is multiplicative, and they add weight.

* Then there's the charging station itself. If it has eight charging stations at, say 250 KW each, that's 2 MW draw from the utility (or an underground group of batteries recharged at a lower rate from utility power?) when all stations are charging cars. Not trivial; just the 500 KVA (call it 500 KW for discussion purposes) utility distribution transformer outside building where I work is the size of four refrigerators, not including its switchgear, all of which is enclosed by a 20 foot tall fence surrounding about 200 sq feet. Multiply that by four in volume to get 2 MW supply required from the utility.

* Consider that a few days ago there was a discussion of what Denmark's going to do with the excess electrical energy their offshore wind farms are generating [1]. They're dumping some of it off as heat into district water heating; it could be charging electric cars.

* As noted above, electric car charging generally would happen at night, so its 8 KWH of energy draw would not happen at the same time as most of the other home's average draw of 20 KWH. KWH is energy, whilst KW and GW are power (energy over time). Distribution and service are built first to a power (KW / GW), not energy (KWH / GWH) demand specification. One must integrate varying power over time to get the resultant energy involved. So the comparisons of KWH to KW seems to me like trying to compare areas (energy) to distances (power); there's a dimension missing on one side of the two quantities, so they can't be directly compared.

* I don't think California is all that special in that it draws the least of its statewide power during the night [2]. Looking at today's demand graph [2], 11 PM to 7 AM appears to be below the average for the rest of the day; that's 8 hours to charge those 8 KWH, which means a 1 KW power draw (that's less than 5 Amps at England's 230 mains voltage).

* 1KW seems to me a not-unreasonable additional night time power demand once everyone's in bed -- though the licensed electrician who inspects your electrical service before you commit to an electric car and the requisite charger should be consulted first. 1 KW will charge that 8 KWH requirement in 8 hours (ignoring conversion inefficiencies for discussion). If the car charger varied its draw subtractively against the whole-home's power draw, then charging could start slow but when one gets home then ram up (increase power demand) as lights, TVs, and other home loads were turned off, keeping home demand constant (though of course using more energy than without the car, so you'd of course pay for that extra energy). If smart meters reported their total-home draw at six second intevals to the consumer, then the consumer's car charger could self-regulate its draw. None of the smart meter reporting I've researched informally report anything better than on 15 minute intervals to the consumer when they do so at all. You could get a home energy monitor installed to get that information yourself (cited in [1]).

* Even at the largest scale, of utility scale generation, night time charging evens out the load rater than adds to peak demand -- as long as charging starts after hours. If so, this becomes an opportunity, not a problem for the electric utility. They can make more money selling more power per day because with electric cars, now there increased demand at night that wasn't there before.

---

[1] https://news.ycombinator.com/item?id=5547979

[2] http://www.caiso.com/outlook/SystemStatus.html

A quick check on wind turbine design for power generation [1] confirms to my satisfaction that all wind turbines should be throttable down to zero electrical output and then with good mechanical brakes [2] taking away the remaining kinetic energy sufficient to stop then lock the rotor. I've read elsewhere (sorry, no reference) that stopping the rotor can be necessary in emergency windspeed conditions. Per [2]:

  >  Let’s compare the emergency braking requirements of a
  >1.5 MW wind turbine under maximum wind conditions with 
  >those of a 40-ton mining truck. Imagine driving a 
  >fully loaded truck down a steep gradient of 25% (1:4)
  >at 85 mph when a road sign warns of a cliff a quarter 
  >mile ahead. The engineering required for effective 
  >braking in both cases is much the same. Braking for 
  >the wind turbine is, in fact, more demanding.

Beyond mechanical brakes, the available control strategies and actuators normally applied toward maximizing power output per [1] can instead be applied toward bringing the power output to zero and the rotor to a halt, such as:

* Apply generator torque control (dumping power to the grid) sufficient to slow the blades down to near zero

* Then or as the blades slow down, stall or furl the blades using pitch control

* Apply the brakes to lock the rotor

* Yaw the whole machine 90 degrees off the wind

OK so if they're stoppable, what is up here?

Others more familiar with Danish market, operators and regulatory conditions might speak better to this, but I'll surmise that the turbine operators get paid when they produce energy, whether Denmark needs it or not. So the turbine operators never stop their turbines, night nor day.

When that creates an excess of energy the grid engineers must do something. The article mentioned slowing hydro output elsewhere and heating the water used locally for district heating.

This creates an environment where there should be a financial incentive to create technologies for storage or highly variable use of this nightime and other times highly variable excess electrical energy.

Hundred-megawatt data centers [3] with quickly variable compute capability come to mind, but making stored hot water hotter (the district heating 'dummy load' approach) seems to me a lot simpler, cheaper and immediately practical, if a bit wasteful of the low-entropy energy that electricity is.

---

[1] https://en.wikipedia.org/wiki/Wind_turbine_design

[2] http://news.altramotion.com/?p=242

[3] http://gigaom.com/2012/01/31/the-era-of-the-100-mw-data-cent...

Not exactly. Since it was a subsidized system, with 'Net Metering', my utility sets the billing rules. Excluding the finer points, here's the gist of how the billing works:

* My utility (Los Angeles DWP) accumulates a 'dollar banking' credit for the present value of net energy [signed power flow integrated over the two-month interval of my utility's billing period] fed back to the grid when a billing statement period is net-energy-productive (i.e. the PV system made more energy than my home consumed over the statement-period of two months). I pay a $22 'minimum charge' (normalized: $11/mo) for, as I see it, my share of utility equipment amortization and operations I'd otherwise be paying as a consumer. Fair enough: I do draw energy when the sun goes down. It's co-operative.

* In the summer, my 5-ton A/C use exceeds my PV system's production capacity; I become an energy consumer during one or two billing periods. My utility calculates the price for the _net_ energy I use per period, then debits the 'dollar banking' account. If that gets exhausted, then I pay the residual, otherwise my 'dollar banking' balance just goes down.

Here's a PV power estimator: http://rredc.nrel.gov/solar/calculators/pvwatts/version1/

> Don't even consider solar panels until you've reduced energy needs first.

Absolutely, asynch13! That's what I did. It took over two years, starting with a home energy monitor.

Using a home energy monitor such as [1] or [2], one can get frequent measurements, every six seconds for [1], then identify individual loads by switching them on and off during a period of otherwise steady demand. Other loads silently switch on and off of course;

Then one can start making evaluations and upgrades. Some can be surprising; I was able to reduce my demand by well over 20 KWh / day; details here [3].

After that, a 5KW / 8000+ KWh-per-year grid-tie no-storage solar PV system went in. My utility charges $11 for a month when the PV system produces more than the home consumes during a billig period.

The utility did change out their transformer (to a smaller 25 KW unit feeding 3 other homes) on the power pole in my back yard. But home and location is 40 years old, so I can imagine the old transformer was fully amortized long ago.

---

[1] http://currentcost.net [2] http://www.theenergydetective.com/ [3] https://discussions.zoho.com/powersave/topic/how-much-i-ve-b...

* Confirmed, an F-1 engine stands tall [1] outside Rocketdyne corporate HQ on Canoga Ave just north of Victory Blvd in the San Fernando Valley of Los Angeles. One can walk right up to the engine and have a gander; it's 18 feet tall. You can admire the maze of cooling pipes on the upper half of the nozzle and imagine the heat and roar.

* In decades past, during development of the Apollo space program, Rocketdyne tested engines at its Field Laboratory test facility in the Santa Susana Mountains a few miles West. The sustained outdoor roar of those engine tests was audible better than 7 miles away. Source: My elementary school classmates, blase about the whole thing, saying "Oh, that's just Rocketdyne testing again".

* Lots of facility pictures of several of the Rocketdyne facilities (with engines under assembly) in the Rocketdyne Archives [2]

* It's a safe bet the name "Rocketdyne" would be a portmanteau of "Rocket Dynamics".

* Some rocket equations and stability discussion at [3]

[1] http://www.dailynews.com/ci_21138479/pratt-whitney-rocketdyn...

[2] http://www.rocketdynearchives.com/canoga.html

[3] http://earthweb.ess.washington.edu/~space/ess472/Lecture_Roc...