Cable pulling lubricant exists

A few weeks ago, I was attempting to run Cat 6 UTP cable for gigabit Ethernet from our house to the garage in a pre-existing conduit. After a very frustrating attempt involving a fishtape, various strings, a flashlight, and lots of scrambling around in the crawlspace under the kitchen, I was convinced that it was not possible to pull the cable through all the corners in the crowded conduit without damaging the cable. My next thought was, "I need . . . cable . . . lubricant. I bet that exists." A quick trip to McMaster yielded 7431K41, a quart bottle of Ideal industries' Yellow 77 Plus cable pulling lubricant.

The next weekend, I re-pulled a new cable slathered in Yellow 77 Plus. The force required was reduced by an order of magnitude. I was gleeful.

Anyway, I just wanted to help distribute global knowledge a little more evenly: cable pulling lubricant exists! Also: gigabit Ethernet to the garage!

Becoming a renewable energy engineer

Due to an unlikely confluence of world energy trends, my interest in engineering, and the internet, lots of people ask me how they can get my job as a renewable energy engineer. Most of them don't want my job literally, but they do want to work on the engineering side of renewable energy. I generally don't mind answering that kind of question, but I'm busy enough that I thought it would be worthwhile to summarize my answers to the basic questions. I realize that this will probably make more people email me rather than fewer, but at least we'll be able to skip the first few questions; the efficiency of the world should be slightly higher.

What do I do first?

Try as hard as you can to perform engineering immediately. If you can get an internship or an entry-level job, do it now. Your employer will wish that you had waited until you had more skills, but that's their problem, not yours. Do not wait until you have completed engineering school-- you might hate engineering, and school is expensive in time and money.

People hate engineering?

Yes, they do. Engineering is the impossible job, never completed, too complex to understand in full, always held up by details, perverted by marketing, unending in unexpected wrinkles. Years later, you realize your approach was wrong. Plus, it's boring.

You should assume that engineering is not the right job for you. You have to be indoors almost all of the time. You must sit at a desk and use a computer most of the day. Most of the results of your labor will be thrown away or, if you're lucky, recycled for the next design. You will be asked to resend spreadsheets to people who will not understand them. If you aren't truly obsessed with solving hard problems, it's not for you. If you cannot artificially sustain your interest in dry topics, it's not for you.

Yeah, yeah. What if I can't get a job right away?

There are two second-tier classes of tasks: figuring out whether engineering will satisfy you and learning basic engineering skills. If you pay attention, you can figure out whether engineering satisfies you as you develop your skills.

If you want to be a mechanical engineer, you have to learn a CAD program like Solidworks. It is extremely likely that your first job will involve a lot of Solidworks. You might end up using Proengineer or Catia rather than Solidworks, but most renewable energy companies are startups, so Solidworks, the cheapest of the lot, is most common. If you're well-suited for mechanical engineering, you're probably already building stuff in your basement or garage or living room; model your next project in CAD before you build it.

If you're leaning more toward electrical engineering, you have to learn to lay out circuit boards. The most common software packages for PCB layout among startups are, I think, Eagle and Altium Designer (formerly known as Protel). Eagle is popular because the free version is legal, but Altium Designer is more powerful and pleasant to use. Lay out a simple PCB and send it to AP Circuits for fabrication. For $50-100, you'll have your first set of PCBs.

What else do I need to learn?

Even if you want to be a mechanical engineer, you have to learn the basics of electronics. Read chapters 1-4 and 6 of Electronic Circuits and Applications by Senturia and Wedlock, even though it's from 1975 (thanks to Max Davis for the recommendation). Other people will recommend The Art of Electronics by Horowitz and Hill; I think Senturia and Wedlock is much clearer. If you can take a class where you get to use a soldering iron, that's a great step, before or after reading about the topic.

You also have to learn some kind of computer programming. The languages you might learn include: C, C++, Java, C#, Python, PHP, Javascript, Perl, and Ruby. I'd start with Python or Ruby, especially if you have no programming experience. If you have any interest in embedded electronics, like the brains in battlebots, you need to learn C. If you hate or fear computers, learning to use LabView is probably your best bet.

You should also learn how the internet works. Reading the chapter 1 of Richard Stevens' TCP/IP Illustrated, Volume 1, is a good start. If that goes well, read chapters 2-4, 9, 11, 14, and 17, plus the Wikipedia page on HTTP. I'd also recommend reading C. J. Date's Database in Depth. If you make it halfway through, you're ahead of most database programmers.

That sounds like a lot of work.

No, it doesn't. That sounds like what you want to do with your spare time anyway. If you don't want to learn this stuff, go do something else. There is a lot of work in the world that needs to be done; engineering is only the optimal solution if you're obsessed with engineering or bad at optimization. We need good dentists, scientists, nurses, welders, luthiers, lawyers, and cheesemakers as well. Especially the cheesemakers.

But what about engineering school?

You must go to engineering school. It will take 1-5 years, depending on your background and what school you go to. There is a very slight chance that you can gain enough proficiency in engineering to make it without engineering school, but it's much more likely that you won't have the skill or discipline to teach yourself engineering faster and more cheaply than engineering school would. You can probably learn basic mechanics of materials or some math on your own, but learning advanced topics on your own is painful and slow.

For renewable energy, make sure you pay attention during thermodynamics and heat transfer. You need to take a statistics course as well.

No, you should not get a PhD.

Wait, you've barely mentioned renewable energy.

Renewable energy engineering is a broad field. The major skill you need is the ability to understand the physical principles that drive complex systems. For example, when someone says to you on the street, "I have a remarkable new device that will allow this skyscraper to capture all the energy it needs from the wind or sun," you listen politely. You do not need to know the details of the geometry of their vertical axis wind turbine or their solar concentrating lens because you know that the power density of wind and solar energy flows is 100-1000 times less than the power density of a skyscraper. When they have stopped talking, you ask them what power density they expect to reach. They stare at you blankly, and you can continue on your way, certain that you have not missed an opportunity.

Even when you are not turning down equity opportunities on the street, you should expect your engineering to run in a pattern that starts with the application of broad principles. Once you have an approach that can work in theory, you develop the details.

Beyond a broad engineering education, you need to learn the basics of solar, wind, and biofuels, plus or minus whatever you're interested in.

I'd start with Vaclav Smil's books:

• Energy: A Beginner's Guide
• Energy at the Crossroads
• An excerpt from a 1999 Smil book, hosted, probably illegally, by Jeff Feddersen at NYU

Then I'd read in the subject areas.

Solar: Applied Photovoltaics by Wenham, Green, et al. Wind: Wind Turbines by Erich Hau is outstanding. Biofuels: Ahindra Nag's book is at least decent.

So I just read books for years?

No. As I said at the outset, most important is that you start engineering right away. Build something today, and measure whether its performance is in accordance with theory. The renewable energy field is rife with crackpots; we need well-educated engineers to solve our worsening energy problems. Get to work, now.

Jatropha and my ignorance: a retrospective

I don't know a damn thing about jatropha oil.

Some time ago, I heard that a friend of mine was starting a biofuels company that would import jatropha from Africa and turn it into biodiesel. At the time, I was skeptical of the plan. "Jatropha? What the hell is jatropha? Your feedstock is something I've never heard of-- how plentiful can that be?"

The next day, I was in the MIT library reading renewable energy journals, and I noticed an announcement that a 15 MW combined heat and power plant run on jatropha oil was under development in Belgium.

It's a little late, but I wanted to let the internet know that I was wrong, and my dear friend Jesse was right. I don't know a damn thing about jatropha oil.

Visible from space

I am excited to report a personal engineering milestone: something I had a hand in designing is visible from space. In the picture, the item of note is long, skinny, horizontal, and white. I wasn't the lead designer, I didn't build the thing, and I can't say what it is, but the first iteration CAD model created of this thing was made by me.

I hope it is not interpreted by aliens as a message. "__" is not as friendly a welcome as I would like to give to our new overlords.

What's new at the 4th Annual Conference on Clean Energy in Boston

I'm taking notes at the 4th Annual Conference on Clean Energy in Boston today; I figured I might as well share what I learn with the world.

The Massachusetts Technology Collaborative has funded MTPV, a start-up in the Boston University Photonics Center, with $500k. (I think this is actually not news to the world, but it was news to me.) MTPV has an interesting technology. Their scheme is a variation on thermophotovoltaics (TPV). In conventional TPV, you put a very hot plate next to a solar cell. The cell absorbs the radiation coming off the plate, allowing you to get electricity from heat. Unfortunately, the efficiency is quite low. MTPV does the same thing, but with the hot plate ridiculously close to the solar cell. Their claim is that when the gap between the cell and the hot plate reaches the micron level (the M in MTPV is for micron), the efficiency increases dramatically, and they have some test data to prove it.

Maintaining a consistent micron-scale gap is a mechanical challenge. They do it by putting small bumps on the surface of their hot plate and then clamping the plate to the cell. They have some clever geometry that minimizes the heat transferred through the bumps, as that is also an efficiency loss.

Investment pitch #1 Srikanth Gopalan from Boston University described a "solid oxide membrane" which takes in waste and steam and emits syn-gas and 7 cc/min of hydrogen gas per cm^2 of membrane. Dr. Gopalan and his colleague Dr. Pal have three year plan that requires $500k per year for device testing and scale-up by 2011.

Investment pitch #2 Scott Faris, CEO of Planar Energy Devices, says that Planar is making metallic lithium batteries. 4x capacity and 10x lifetime, 80% cost reduction. Doesn't catch on fire like metallic lithium batteries used in cellphones in the past. There are no liquids in their batteries. They bury the lithium anodes under solid glass electrolyte. Faris says that Planar is using "the Miley Cyrus strategy,"-- the best of both worlds-- using thin film technologies, but scaling up to the size of prismatic batteries.

Planar has two products in development at present: a small battery called the PowerPlane: 25 x 29 mm, 12 mAh, and the larger PowerBlade, 100 mm by 100 mm, 4.8 Ah, 520 Wh/L. The PowerBlade is being tested by the military at present. Planar is pursuing Series B funding for pilot production.

Side note: Faris followed the irritating standard of the battery industry and quoted capacity in amp-hours rather than watt-hours. Amp-hours is not a measure of energy capacity, but of the number of electrons that can be induced to flow out of the battery. For example, a 1200 mAh NiMH battery contains around 20% less energy than a 1200 mAh alkaline battery because the NiMH battery runs at ~1.2 V, while the alkaline runs at ~1.5 V.

I suspect that this amp-hours habit developed because lead-acid battery manufacturers wanted to hide the fact that their voltage dropped as the battery discharged, making the amps that come out near the end of a discharge cycle less powerful than the amps near the beginning. If you speak in terms of amp-hours, you can avoid this embarrassing truth. Generally, modern batteries have a much flatter discharge curve, so watt-hours ends up being a linear multiple of amp-hours. Modern batteries have much less to hide in this regard, though there is still the variation across chemistries to account for. Nonetheless, the habit persists throughout the battery industry.

Investment pitch #3 Andrew Dillon, CEO of Varentec, described their solid-state electronic transformer that operates at 20-50 kHz. Dillon claimed that their transformer is 10x smaller and lighter than those of competitors that operate at 1 kHz, and costs 25-40% less. Undersea transmission, as required of offshore wind turbines, requires DC for efficient operation; photovoltaics also produce DC, so making transformers will be a lucrative business in the next few years. They want $1M for their first 1 MW high-voltage DC system, which they hope to build in 8 months of 2009.

Investment pitch #4 John Thomson, President and CEO of Yield Energy. They are a biogas developer. There are already 5000 biogas plants in operation in Europe, but no utility scale plants in the US. They use expired food products from grocery stores and waste from restaurants as their feedstock. Negotiated exclusive rights to proprietary pre-processing technology from Fitec in Germany. They have 6 sites currently under development; the first is near Toronto, Canada.

Investment pitch #5 Chris Sauer of ORPC started with a bold statement: "We have saved the best for last today." They've raised $2M and have recently tested their kinetic tidal generator in the Bay of Fundy. The systems are "horizontally-mounted cross-flow turbines" that look like helical Darrieus windturbines or Gorlov turbines on their sides. The turbines are made of composites and foam so they float. They tie the turbines down to the sea floor for operation and release for service. Unlike their competitors in New York, Verdant Power, they have no cantilevered blades. Sauer says they can get 250 kW per bladeset in a 6 knot current; 1 MW in a module of 4. He claimed their costs are 6-7.5 cents/kWh for a location with a peak current of 6 knots.

If all goes according to plan, their first module of 4 bladesets will be installed in Q4 2010. Their first commercial installation will be in Q4 2011 in Western Passage up near Lubec. ORPC has raised $4.5M so far; they're hoping to close Series A funding $10M in Q1 of 2009.

One of their competitors, Natural Currents, also had a table in the expo hall.

Somerville locals Second Wind also had a booth in the expo hall. (Disclosure: my employer has worked for Second Wind.) They make various tools for measuring how much wind is available in different places. Their latest product, the Triton (540 kB PDF), looks like a megaphone the size of a dumpster, pointed at the sky, with a solar panel bolted to the side of it. The thing is absolutely amazing. It emits sound waves at a slight angle from the vertical. The reflections that come back are Doppler shifted, and staggered in time. The frequency shift is proportional to the air velocity and the delay is proportional to the height. They can change the direction of the signal, so they can tell wind direction too. Their range is around 160 meters. This is so much better than setting up a tower with a bunch of wind vanes and anemometers. I predict that these guys will own the market in a few years.

So that's how I spent my morning. The rest of the day was debugging a PLC in a warehouse.