omenti

who are you going to call?

Drop two pieces of bread into the toaster and push the lever down.  You go about your business getting the rest of breakfast ready and then, after a few minutes, two pieces of toast pop up.  Have you ever thought about what's going on?

 

We usually think about the toaster receiving the bread and giving it back to us after a few minutes - electric heating somehow.  Someone from a few thousand years ago might wonder where the bread came from and what kind of magic transformed it into something similar but different after a few minutes.  He might wonder if the lever was some kind of prayer or incantation.  We can examine the process at a variety of levels and can find considerable depth.  We know the cord is plugged into a socket which has a path though a series of transformers and wires that usually lead to a generator that is turned by spinning a turbine with falling water or steam superheated by nuclear fission or the burning of fossil fuels.  When the switch goes on the load on the generators increases by a bit more than we're using.  The steps required advances small and large over the decades in science and technology.   For example getting the of generation and transmission of electricity to work efficiently depends on an understanding of Maxwell's equations and a fair amount of math.  

 

We don't think about these things because we trust them. There's a solid bedrock of reliable knowledge in much of our technology even though we're not one hundred percent sure of the science.  Our trust in science is nearly universal even though some of us chose to deny certain aspects.  Global warming deniers and flat earthers have no problem using the Internet - computer mediated communication built on silicon based semiconductor technology, fiber optics and much more.   Their mistrust is often based on something in their personal value system that to first order may seem orthogonal to the scientific arguments.

 

Not many people have a good handle on how science is done and how it progresses.  Many of us had to memorize the steps in "the scientific process" in elementary school, but it turns out it's messy and there's no formal process.  In college you may have come across Karl Popper's notion of falsifiability and Thomas Kuhn’s paradigm shifts. Both of these models have been shown to be wrong.  The view among many scientists and philosophers of science is science is more or less makeshift, but produces reliable knowledge and tends towards self-correction and reliable knowledge over time.

 

A few years ago a few lectures by Naomi Oreskes at Princeton gave me the basis of something I'm more comfortable with even though it's not complete. She argues reliable knowledge is based on five factors:  method, evidence,  consensus, values and humility.  I won't go into these as each is a deep subject, suffice it to say that the last three are deeply social - something many "hard" scientists tend to have a difficult time admitting.  Science is deeply collaborative and these social aspects often determine what is worked on as well as how it, and the people who did it, are perceived. 

 

Although most people trust much of science, personal and group values can get in the way of trusting certain aspects as well as certain scientists.  We've witnessed this up close with the pandemic, global warming and evolution.  I worry that many scientists have made a mistake by hiding their values (that's changing with some of us) as well as separating science and technology.  The separation of science and technology was value based and began to become dominant after WWII. Somehow science was to be the pure pursuit of nature even though it's linked to technology at the hip.  Often technologists are well versed in the science underlying their work and some venture into applied science.  Scientists, particularly experimentalists, are by necessity amateur technologists.  I've done some technology and have a bit of a feel for it, but I'm certainly not skilled - I do much better on the science side.  I'm in as much awe of great technologists as I am of great scientists. 

 

A final point.  Who should you trust to do science?  If the light in your house keep going out when you plug the toaster in, it probably makes sense to call a licensed electrician.  When a pipe in your house bursts spewing out gallons per minute, you call a licensed plumber and not an electrician or dentist. A trip to the dentist and not the Ghostbusters is in order when you hit a cherry pit in a piece of pie and a tooth falls out.  And when you're thinking of the long term - say your children's lives - listening to experts on global warming makes sense while listening to Mobil-Exxon or the Farmer's Almanac doesn't.  In each of these areas there are mostly good players and a few bad ones. You find the good ones from the consensus of their community. 

 

a thread of ev history

There's a nice vantage point in  Point Loma in San Diego with a monument to Juan Rodriguez Cabrillo.  In 1542 he made the first contact with the indigenous people of the area.  Not much happened, but sailing Northwards a few days later he noted smoke hanging over what is now Los Angeles.  Enough that he called one of the bays  Baya de los Fumos - Bay of the Smokes.

 

It's not clear what caused the smoke.  The Los Angeles region hosts three types of temperature inversions, each capable of trapping air masses close to the ground allowing smoke to build.  The region may have had one of the highest population density in North America in the 16th century, so it's possible the smoke was from thousands of fires - the region certainly held smoke in later centuries.  It's also possible there were large fires in the hillsides fanned by Santa Anna winds.

 

Before WWII Los Angeles, apart from areas around chemical plants, had fairly clear air most of the time.   That all changed after the war and by the fifties a thick brownish much often dropped visibility to a few blocks.  The brown smog was different from either smoke produced by wood fueled fires or the smogs coal burning was know for.

 

While the LA area was a natural environment for temperature inversions, it also hosted centers of intellectual curiosity.  A few years after the war Los Angeles decided to tackle the problem and created the first air pollution district in the country.  Arie Jan Haagen-Smit of Caltech managed to figure out what it was by 1950.  Sorting out how it was produced took a few more years.   He was able to show eighty percent came from  the automobile.

 

Caltech became a breading ground for ideas to deal with automotive smog.  By the mid 1960s Haagen-Smit and others were thinking about modern incarnations of electric cars. Wally Rippel was a student fascinated by the idea of electrifying transportation. His aha moment came when he showed electrifying the automotive fleet would only require a twenty percent increase in electricity generation.  (the number today is considerably less than ten percent). Wanting to move things along Rippel did what any Caltech student would do.  He challenged MIT to a race.

 

The Caltech team added over a ton of lead acid batteries to an old VW van for the Pasadena → Cambridge trip, while MIT converted a Corvair for their trip to Pasadena. MIT finished about a day and a half faster, but they broke down several times requiring towing more than once. The van, on the other hand, just motored through without incident and were declared winners.  

 

The 70s saw further development at several engineering schools. Recognizing batteries were heavy,  Caltech worked on early hybrid cars that would drive short distances on power from smaller battery packs before starting up a gasoline engine.  Along the way they develope regenerative braking to boost efficiency.  The first energy crises hammered home the idea that efficiency was important.  A Scientific American article appeared in 1973 telling the story of efficiency and declaring a person on a bicycle one of the most efficient forms of transportation in nature .. Steve Jobs later riffed on that calling personal computers bicycles for the mind.  An electric bike would be about twice as efficient as a human powered bike.  That led to work on motors and controllers. Unfortunately the American book disappeared and plans for a safe cycling infrastructure died.  Electric bikes would have to wait a few decades.

 

 

 

In the 80s Rippel was working with Paul MacCready's AeroVironment, something of a Caltech spinoff, on a specialized, almost anything goes, solar powered car for the five day World Solar Challenge race in Australia.  GM footed the bill for Sunraycer team and the car buried the competition.  GM engineering worked with them and the collaboration continued.  Learnings from the solar races went into new motor and controller designs as well as a recognition that aerodynamics were very important gave people the feeling something practical could be done.  Enthusiasm within GM's engineering community talked the company into building the EV1 - the first production modern electric car .  GM corporate didn't know what to do with it.. much has been written on how and why the project was killed after three years of production and extremely enthusiastic customers. Other than its batteries, the EV1 was more sophisticated than current electric cars.

 

Rippel and a few others from the EV1 experience founded AC Propulsion, building a demonstrator called the tzero.  More a proof of concept, it was sportscar designed to impress wealthy tech types defeating a stream of Ferraris and Porsches along the way. Martin Eberhard tried to convince them to take it to production, but the jump in scale was too much for them.  Instead he co-founded a company called Tesla well before Musk entered the scene.

 

Electric car history has many threads.  I'm familiar with this as I heard bits and pieces in school and through people I know.   The deeper story requires many more angles, but it's sticking how much can take place with just a few people.

 

I'm a believer in efficiency and infrastructure.  Moving towards safe active transport (walking, cycling and e-bikes) offers a much greater improvement over conventional electric cars.  (long story, willing to argue) 

 

A bit of early history. International Rectifier funded a solar car project in 1960.  Silicon solar cells were still very new and exotic.  IR provided solar cells to the space program, but wanted to grow a commercial market.  Doing the math they realized powering a home would be incredibly expensive, but a short range vehicle for city use might be practical earlier.  To make a statement they sent a solar panel to Charles Escoffery who converted a 1912 Baker Electric.  The world's first useable solar car.

Finally the clean air work starting at Caltech lead to catalytic convertors, reformulated gasoline and any number of environment regulation that have saved many lives.  That work was an innovation. 

 

building blocks

Watching Apple's Mac Studio announcement yesterday took my mind to something Carl Sagan said in the Cosmos series.

The Mac Studio's underlying technology goes far beyond the many tens of thousands of person-hours Apple has invested.  The semiconductor industry that stands on the shoulders of Bell Laboratories back to the fundamental work of James Clerk Maxwell and before.  Apple and others are able to come up with advances that strike us as dramatic as they make an impact on our lives.  The impact of some innovations like the electric light, the telephone, the automobile, powered flight, radio, the atomic bomb and the Internet have changed how we see ourselves.   But take a look and they're all built on long chains of invention and innovation that are often forgotten.

 

This comes to mind with the horrors going on in Ukraine.  Russia has always had brilliant minds.  They've made stunning advances in physics and math, but they haven't been as successful as the West in building on that.  A good deal of their technology is derivative.  In the past 25 years this may be by design. Most of the export value comes from extraction - oil and other natural resources. These industries seem to fall under the direct control of Putin's friends.  More technical industries are still under oligarchs, but these aren't super high profit.  As long as the core players can control them, the real focus of the economy is what can be run by shear force and power rather than technology and sound business practices.  (note - this is speculation as I'm certainly not a Russian expert..  so take it with a grain of salt).  

 

High tech industry is required for the military and some consumer use.  Although that is supposed to be home-brew there is a lot of importing and rebranding. I've seen that in electronics used in physics experiments and am told is standard operating procedure.

 

So imagine you can't important and rebrand the CNC tools necessary to make jet engines.  Can you make the tools yourself. What if you can't even make precision bearings?  What if China doesn't allow electronics into the country (they probably will, but now they can be the Mafia dons)?  Sanctions can make a medium and longterm impact.  

 

You don't need to make your Apple pies from scratch - we have our universe - but few companies or countries have the capability to build objects of the modern world without technologies that exist outside.  All of these long threads few of us think about until they aren't available.  Think about all of those high value products waiting for fifty cent ICs that were designed twenty years ago...

__________

PS - I'd love a Mac Studio, but it's way beyond my budget. 

 

 

orbits

It's almost Perihelion Day!   I may wish people Happy New Year, but the beginning of a calendar year is artificial, so I celebrate a specific point in our orbit every year.

 

The orbit of the Earth is elliptical. Not terribly so, but enough we're about 5.1 million kilometers closer to the Sun in early January than we are in early July.  The closest point is the perihelion and the furthest, the aphelion.  The exact time of perihelion varies from year to year - this year it's the 4^th of January at 6:52am UT or 1:52am EST.  Californias get to celebrate on the 3^rd.

 

At perihelion the Earth is at its fastest speed in our orbit - 30.29 kilometers per second.  At aphelion, the slowest point, we've slowed by about a kilometer per second.  Given its closeness, the apparent disk of the Sun has about 7% more area at perihelion than it does at aphelion.  That seems to bother people as it's Winter in the Northern Hemisphere.  In fact the Earth is a bit more than 2°C colder at perihelion than at aphelion - an artifact of the distribution of the continents and a bit of physics.

 

There more land in the Northern than the Southern Hemisphere.  Water has a higher heat capacity than land.  Think about a desert.  During the day it can get scorching hot, but it rapidly cools off at night.  The ground doesn't hold heat as well as water.  In July the Northern Hemisphere has long days that allow the land to heat up for longer periods raising the temperature of the air above it and raising the average temperature of the Earth in the process.   In January the Southern Hemisphere is getting the most sunlight, but most of it falls on water, which doesn't heat up as much as the land. 

 

There are some interesting rabbit holes one can get into and much can be learned by studying the positions of bodies in our solar system and people have been doing it for thousands of year.   Some civilizations were very sophisticated in calculating and predicting astronomical events even though their models of the solar system were wrong.  An early computer that makes your jaw drop is the Antikythera mechanism.  People have been studying it for over a century and Scientific American has a fine summary of what's currently known about the device.

 

In 1900 diver Elias Stadiatis, clad in a copper and brass helmet and a heavy canvas suit, emerged from the sea shaking in fear and mumbling about a “heap of dead naked people.” He was among a group of Greek divers from the Eastern Mediterranean island of Symi who were searching for natural sponges. They had sheltered from a violent storm near the tiny island of Antikythera, between Crete and mainland Greece. When the storm subsided, they dived for sponges and chanced on a shipwreck full of Greek treasures—the most significant wreck from the ancient world to have been found up to that point. The “dead naked people” were marble sculptures scattered on the seafloor, along with many other artifacts. Soon after, their discovery prompted the first major underwater archaeological dig in history.

One object recovered from the site, a lump the size of a large dictionary, initially escaped notice amid more exciting finds. Months later, however, at the National Archaeological Museum in Athens, the lump broke apart, revealing bronze precision gearwheels the size of coins. According to historical knowledge at the time, gears like these should not have appeared in ancient Greece, or anywhere else in the world, until many centuries after the shipwreck. The find generated huge controversy.

...

 

A more technical discussion appears in a recent article in Nature (open access)

 

A Model of the Cosmos in the ancient Greek Antikythera Mechanism

Tony Freeth, David Higgon, Aris Dacanalis, Lindsay MacDonald, Myrto Georgakopoulou & Adam Wojcik

The Antikythera Mechanism, an ancient Greek astronomical calculator, has challenged researchers since its discovery in 1901. Now split into 82 fragments, only a third of the original survives, including 30 corroded bronze gearwheels. Microfocus X-ray Computed Tomography (X-ray CT) in 2005 decoded the structure of the rear of the machine but the front remained largely unresolved. X-ray CT also revealed inscriptions describing the motions of the Sun, Moon and all five planets known in antiquity and how they were displayed at the front as an ancient Greek Cosmos. Inscriptions specifying complex planetary periods forced new thinking on the mechanization of this Cosmos, but no previous reconstruction has come close to matching the data. Our discoveries lead to a new model, satisfying and explaining the evidence. Solving this complex 3D puzzle reveals a creation of genius— combining cycles from Babylonian astronomy, mathematics from Plato’s Academy and ancient Greek astronomical theories.

 

Back to now.  Happy Perihelion Day to all of you!

and a comment

I'm  not fond of the idea of the Earth making a circuit every year as, the Earth never comes back to the same spot in any reference frame.  It's roughly close enough for most, but in four dimensional space time it's something of an approximate helix expanding out along the time axis.  You can never go home. Good enough reason to look forward!

 

 

 

 

 

looking back on long term focused change

Apple introduced two new laptops this week.  Fourteen and sixteen inch versions of the MacBook Pro.  On the surface they show a company responding to a certain class of users who value functionality over style.  A serious reversal from the days when Jony Ive dictated a minimalist approach.  Very nice laptops that happen to be extremely powerful. As fast as the most powerful Intel laptops with a much lower power consumption. It is likely these will run quietly most of the time - useful on video calls.   And while Intel has been almost stalled on increasing performance, Apples M chip series is only beginning its ramp.

 

I'm not a historian, but here's a rough sketch.  

 

  The Motorola 6502 was simple chip that retailed for about $25 in the mid 1970s.  The price made it popular with hobbyists.   Steve Wozniak used it in the Apple I and the commercially successful Apple II.  This started down on a long relationship with Motorola.  Beginning in the 80s Intel supplied the x86 chips used in IBM compatible PCs.  During the 80s Apple had moved to Motorola's 68x processors.  By the late 80s Motorola was falling behind.  Apple was in the unenviable position of being tied to a substandard processor.   Some in the company wanted to design their own processor, but Apple lacked the necessary resources.  Instead they turned to IBM which made fantastic silicon and was working on a reduced instruction set processor family (RISC) known as the PowerPC (PPC) 

 

I won't go into the RISC vs CISC wars of the 80s and 90s (CISC is complex instruction set) other than mentioning there were two competing processor architectures.   Intel's x86 and the Motorola 68x family were both CISC.  Apple was internally interested in the higher compute power per watt performance with possible with RISC.  Interested enough that they became part of a joint venture that changed the history of technology.

 

ARM was originally the Acorn RISC Machine - an English home computer the BBC was pushing.   Apple was the majority holder of a joint venture between Acorn, Apple and a chip maker called VLSI technology.  The company was headquartered in Cambridge, England and took the then novel approach of licensing their design rather than making chips.  To say this was successful is understatement.  Apple used ARM chips in the Apple Newton.  It was a failure and taken off the market when Steve Jobs returned to Apple, but the desire to have their own processor was there and a few changes forced by the market made it possible in the longer term. 

 

Apple introduced their first Power PC machine with an IBM processor in 1994.  It was fine as long as it was running software designed for the chip.  The problem was most software companies were more interested in supporting the fast growing Wintel platform.  Attempted to keep some market, Apple created an emulator that made the PPC act like it was a 68x processor.  It was a beautiful piece of software, but 68x programs ran slower on the PPC machines than on three year old 68x based machines.  Apple stayed alive, but nearly went out of business.  Then Steve Jobs returned with a much better operating system and - well - himself. 

 

By the early 2000s Apple emerged with some great industrial design and a growing number of PPC native programs.  Then came a tiny RISC machine called the iPod that only played music.  It gave them experience with integrated hardware, software and a service.  It may have also marked a point where Apple came out with something not only surprising, but cool.  But IBM was failing to keep the PPC relevant.  Apple needed to jump. This time the only choice was going back to the CISC x86 Intel processors.   Mac OSX, Apple's Unix based operating system, had been running on Intel chips for several years.  The software was improving quickly, but getting important developers write native applications was difficult.   Once again Apple built an emulator.   This time they had some internal experience with the process.  It was called Rosetta.   PPC instructions were converted to x86 instructions..  It wasn't perfect, but was efficient and good enough that the platform was growing again.  Growing enough to convince developers to write native software.  And  Apple was giving them the software creation tools to do it.

 

The pain of two major shifts and some interesting new products on the drawing board.  They needed their own silicon.  They purchased a small silicon design company and built it in the US and Israel.  It took years, but over time  the results have been remarkable  Systems on chips for the iPhone, iPad, AirPods, HomePods, the Apple Watch, and a variety of small support computers used in other devices like a pencil and a keyboard display.  All of these run versions of MacOS, use the same development tools,  and many are integrated into similar services.  

 

About five years ago Intel began to run out of steam.  Apple was working on its third huge transition for PC class machines. This time going from Intel x86 to their own M series ARM (RISC) based chips.  The first laptop was introduced last year.  The M1 was basically an iPad chip in a laptop. The operating system and a lot of Apple and third party software runs brilliantly on it.  The emulator runs native x86 programs about as fast as they ran on Apple's x86 laptops.  

 

That brings us to the announcement a few days ago - the M1 Max and M1 Max Pro.  (scroll down a bit for the chips) Both use the same ARM based design The larger Pro chip and has 57 billion transistors while the original 1985 ARM1 chip had just 25,000 transistors.  The ARM1 was made with 3 micron process (very roughly the distance between lines and spaces in the lithography) - that's 3,000 nanometers.  The M1 Max Pro uses a 5 nanometer process.  

 

Here's a relative size comparison.  ARM1 in red M1 Max Pro in blue. (A real M1 is about 20mm wide and the ARM1 is about 7mm) Look closely between the two and you may see a red one pixel dot. That's how big the ARM1 would be if it was made with the same process used in the newer chip.  

 

A feature of putting so much on a single chip is subcomponents can be directly connected on silicon.  It offers  greater performance and is more power efficient than putting data on a bus and shipping it around from chip to chip and then back again.  Such integration is very difficult  if you're selling a general purpose chip. Phone and PC makers need to differentiate their product and that usually means doing it with a mixture of chips and the software that binds them.  Apple doesn't suffer that restriction.  And, until Intel's architecture, there appears to be a lot of blue sky for performance improvement.  The real competition will probably come from other ARM chips, but Apple has a lead.  Here's some very early commentary from Anandtech.

 

While not many people will buy high-end laptops, they continue to buy the iPhone -  arguably one of the most popular consumer products ever made - along with a large supporting infrastructure of apps and services. And they make smart wireless headphones, smart speakers, smart watches, smart pencils and more.  All of this is based on the same core hardware and software architecture.   As a friend pointed out it's like the same company makes Ferraris, ebikes and everything in between using the same fundamental design philosophy and somehow making it work.   Not a perfect analogy as Apple has a much broader range.

 

A Moore's Law observation.  In 1985 the transistor count for all of the computers Apple sold in 1985 was about 25 bilion - less than half the count on a single M1 Max Pro chip. 

 

 

 

 

repairability

Growing up our family always had Fords.  For most of grade school through junior high that meant a blue Ford Falcon.  My Dad's reasoning was simple - a neighbor had good mechanical experience with Fords, particularly the straight-6 engine in the Falcon.  He also had a nice set of tools. Much of it was simple enough that I was involved in part of the routine maintenance even before I had my driver's license.  Simple things like changing the oil every 1,500 miles, gapping and replacing the spark plugs, making sure the battery levels were ok, draining the radiator, testing and replacing the antifreeze, and taking off the air cleaner to spray ether in when cold-starting it when it was below -20°F.   

 

By the time I was driving legally  (I drove a pickup on a wheat farm during harvest the year before I had my license. I suspect that was common in Montana.  Learning how to drive a manual transmission amounted to me trying to not stall the pickup while my boss watched laughing for an hour.)  By high school I graduated to more sophisticated work.  Replacing the head gasket on my own and even replacing the rings under the watchful eye of my Dad's friend.  Everything was exposed and simple - most of it was just sorting out fuel, air and spark.   Sometimes, driving to King's Hill on a hot Summer day, the engine would quit.  That meant pouring some cold water on the fuel line near the engine and waiting a few minutes to break the vapor lock.  Each car had a set of peculiar tricks, but makes had personalities.  Changing to a Chevy or Dodge would have a learning curve.

 

While the Falcon was simple and to work on, almost everything was poorly made.  Perhaps not to the eye, but tolerances in the engine and transmission were sloppy.  A careful break-in period was necessary if you wanted an engine and transmission to last.  It was inefficient even though it didn't weigh much by today's standards.  Although it was a compact, it only got 18 miles per gallon on the road and 13 or 14 in town.  Engine and transmission oil had to be replaced regularly.  Tires went about 10,000 miles if you were lucky, brakes needed adjustment every few thousand miles, the chassis needed lubrication regularly and so on.  The body and chassis designs were primitive and non-crashworthy by today's standards.  Cars often remained fairly intact even in fatal collisions.  

 

The seventies brought more efficient designs along with increasing safety regulation.  More sophisticated tooling that produced dramatically improved engine and transmission tolerances that allowed more efficient designs.   They lasted longer, but it was becoming difficult to work on them.   Repair was moving towards replace.  The failure of a wiring harness on the mechanism for an electric window might be a fifty cent part, but the replacement module could cost hundreds and require several hours of work to replace.

 

Home electronics followed a similar path.  Our first TV was a 19" black and white Magnavox from about 1961.  It had a few dozen tubes, a lot of point to point wiring and hand-soldering, and mechanical switches.  If something went wrong the first step was to take off the back of the set and see if any of the tubes weren't glowing.  You'd pull the suspect out and test it at the local drugstore - most had a tube testing kiosk with a stock of replacement tubes.  That happened a few times before we got our first color TV in 1969 - a Motorola "Quasar" which only had two tubes including the picture tube.  It's main selling point was reliability. By the 80s integrated circuits began to invade home electronics and home repairs became next to impossible.  It didn't matter as we perceived our devices were improving faster than their failure rate.  Prices dropped so rapidly that throwing them away became popular.

 

This replace rather than fix mindset applies to many consumer goods, and many aren't engineered for a long life.  During the pandemic our eight year old refrigerator failed.  I was able to get it working again in about a day, but the repair was anything but easy.  The dishwasher also failed.  A gear on the motor, but the replacement module was $260 and special tools are required. It was eight years old so the leap to a new unit was an easy choice.  I probably wouldn't have tried to repair these things if I didn't have the background of repairing things that broke at home.

 

Ten years ago a mechanical engineer friend thought there might be a market for simple good design and quality manufactured items - mostly home appliances.  They would cost more, but last twenty or thirty years and be easily repairable.   He built a few prototypes, but was laughed at.   It turns out we're mostly ok with disposable.

 

The design integration in most smartphones and watches makes repairs difficult.  If compact, rugged and waterproof are goals, replaceable batteries are an issue even though you may need to replace them every few years.  A cynic might say it gives the user a push to get a new device.  

 

More important is the level of software integration across the system.  Apple's walled garden is problematic for many, but the value of more security and privacy (assuming you don't live in China), and mostly well-behaved apps is worth it to many.  Trading price and flexibility for stability and privacy.   

 

This raises the question of what should we be using?  Cars are much safer and efficient, although more expensive to operate, consumer electronics are mostly better and much cheaper.  Thinking of different ways to do things probably makes sense.  Getting rid of a second car and adding an electric cargo bike if your roads are safe enough is an option.  Three very sophisticated friends have traded their smartphones for simple "dumb" phones.  Well-made and designed clothing that lasts a decade.  Not everyone has multiple choices, but the pandemic seems to have people questioning many things. 

 

I'll end with a recommendation that applies almost universally..  staying away from IoT (Internet of Things) devices makes a lot of sense unless you know exactly what the security and privacy issues are in each case.  And a fundamental flaw in many - they have software flaws and the only way to update is to throw them away and buy something new.  And while it's beyond the main message, some of these IoT devices send information back to organizations that market it.  They turn you and your behavior into a product.

 

 

power different

Recently a friend sent a piece on an energy harvesting device that coverts some of your body's waste heat into electricity.  The description was sparse, but it mentioned the average person produces about one hundred watts of waste heat and went on to say that is enough to power a laptop.  

 

umm..  Both true, but there are some problems ...

 

The average adult metabolizes about 100 watts on average to stay alive.  That's about the same as eating 2000 calories of food a day.   The average adult has about 1.7 square meters of skin area.¹  This means we need to get rid of about 0.006 watts per square centimeter of skin area.  Six milliwatts of power per square centimeter.  The photo of the device showed a ring like affair with some fins - maybe two square centimeters of radiator area.  Let's round that to about ten milliwatts for the device - hardly enough to deal with a laptop, which are on the order of 20 watts or even a smartphone which are a few watts.  That assumes all of the waste heat can be harvested with one hundred percent efficiency.  Here things begin to go down hill quickly.

 

A thermoelectric converter on human skin has a theoretical efficiency of about 5% at room temperature.²  This reduces the six milliwatts per square centimeter down to 0.3 mw/cm².  Even if you covered your entire body, you'd only harvest five watts and you'd be shivering.   And that's unrealistically optimistic as these devices rarely have more than single digit efficiencies on top of their thermal efficiency in this temperature range.  Even optimistically we're down to about 0.03 mw/cm² or 30 microwatts/cm².  Something like an armband might have 100 cm² giving us 3 milliwatts.

 

This isn't reason to abandon hope.  There are hardware devices with sub-milliwatt power requirements. Small energy harvesting devices that harvest mechanical or solar energy also exist and make it practical to put tiny sensors with tiny transmitters nearly anywhere as long as your receiver is nearby. The micro part of the Internet of Things (which has its own issues).  It turns out this is a rich area for college engineering projects.

 

But what about larger power demands where it's inconvenient to change batteries?  Spacecraft that travel past the orbit of Mars generally rely on radioisotope thermoelectric generators (RTGs) because the Sun is too dim to make solar power practical.  An RTG puts a thermoelectric converter on something hot - a radioactive material that decays rapidly.  Ideally it's an alpha emitter so you don't need heavy shielding.  There are a number of candidates, but Plutonium-238 (Pu-238) is the most common American choice with a half life of about 88 years. Not only do these generators produce electricity, the extra heat is used to warm the spacecraft's warm.   The current generation of Martian rover is RTG powered. 

 

There are a number of RTG stories.³  I don't want to go on forever, but the nuclear powered pacemaker and artificial heart are among the more interesting.  

 

A problem with early pacemakers was longevity. Batteries had to be replaced every eighteen months or so necessitating expensive and risky surgery.  In the late 60s and early 70s two types of tiny radioisotope generators were manufactured for use by several pacemaker manufacturers. They had enough shielding to be very safe and would last longer than even very young patients.  They were also very cost effective and a few thousand patients received them. Then, in the mid 70s, the practice suddenly halted.   The worry was devices might not be removed before cremation from patients who had died.  This would destroy the shielding and turn an area in and around the crematorium into a radioactive hot zone requiring evacuation and a very expensive cleanup.  

 

The radioactive heart implant never made it into a patient.  To make enough power to run the heart, the RTG generated about 50 watts of waste heat adding to a normal person's 100 watts.  It wouldn't be bad if it could be spread over a skin sized area, but that would be impractical.  It would be like holding an incandescent light bulb.   One of those interesting ideas that you wonder about afterwards...

_________

¹ The approximation medical researchers usually use is from Mosteller:  (height[cm] * weight[kg]/3600)^0.5  a

² The theoretical maximum efficiency of a heat engine is 1 - Tc/Th  where Te is the temperature of the environment - the air around the device in this case - and Th is the temperature of the radiator - the skin temperature.  We have to use an absolute temperate scale so Th = 310°K and Tc ~ 295°K.  So about 4.8% efficient.. let's call it 5, although it is probably half that with poor coupling to the skin and air. 

³ One could go on for a long time .. the CIA's RTG in the Himalayas is particularly interesting.  And then there's the Apollo 13 RTG that required last minute maneuvers as the limping spacecraft approached the Earth.   The very hot RTG is somewhere on the floor of the Pacific.

 

a meltdown between two brownies

You need to read Om’s blog post Why iPhone is today’s Kodak Brownie Camera  It reminded me of my home-brew digital camera about 35 years ago (the Meltdown the ferret photo is in this post - probably the first digital image of a ferret).   

 

 

Digital photography - at least how we usually think of photography (more on that later) - was invented in 1975 by Steve Sasson at Kodak.  (a fun NY Times piece)  Charged coupled devices (CCDs) used to capture the image, had been invented at Bell Labs a few years earlier.  Sasson used one made by Fairchild for his experiments.  I had read a bit about digital cameras when I built mine, but decided to used a modified DRAM memory chip because I could get one for nothing and it seemed like a reasonable thing to try.  It was just that - fooling around with some hardware and writing a bit of software for fun.  The fact that it captured my only record of my ferret Meltdown made it special.

 

 

But there’s another form of digital image making.  Experimental particle physics people moved from detecting events with regular photography (bubble chambers and film emulsions) to electronic counters.  Around 1968 the multiwire proportional chamber revolutionized the field and set the stage for dramatic discoveries that lead to the Standard Model - still the best description of three of the four fundamental forces.  A MWPC consists of many planes of parallel wires.  They’re assembled in a chamber filled with a gas that ionizes when a charged particle flies by.  The wire closest to the passing particle detects the event and a signal is read out. These arrays are arranged at right angles in tightly spaced pairs so it’s possible to calculate a position in two coordinates.  As the particle passes through the chamber its x and y position is registered along its track.  A computer calculates a trajectory for each particle and all of the particles that registered form a 3d image of the event.  This information is then processed to reveal the physics of the event.   This type of photography has become more sophisticated and MWPCs have largely been replaced by more capable detectors, but it's still the same idea - just a fancy camera.    The people  who use these exotic cameras use their smartphones like everyone else for all kinds of personal images. 

 

 

Over the years I’d spent some time worrying about image and sound quality.  At one point I thought better was - well - better.  My epiphany came in the a perfectly restored 66 red Mustang.  It was near Cleveland with the head recording engineer from Telarc Records and a seriously good musician from the Oberlin Conservatory.  We were talking about where music reproduction might go with sound field reconstruction and dramatically more information than CDs could provide when a favorite Beatles tune came up on the 8 track (gasp! - it *was* an authentically restoration).  The volume came up and the two of them were singing along to the music in pure delight.  There is something transcendent about the music - even with awful reproduction in that noisy environment.   Portable music players with cheap headphones would be good enough for most.  As long as there's a personal touchstone.

 

 

Om points out there is a market for high end photography.  Perhaps the real distinction is high end photography is more of a verb than a noun.  You do it to learn how to see images as much as to capture them.   Hopefully it will never go out of fashion.

 

why you can watch I Love Lucy

A few days ago Om and I were talking about some technologies.  I mentioned I sort of predicted one with reasonable confidence about twenty years out.  It seemed a bit too scifi on the surface, but other people in research agreed it could happen if a lot of money was spent.  It seemed desirable, so why not?  There was only one major roadblock and that turned out not to be roadblock at all.  It needed a much better battery than anything we knew about.  A lithium-ion battery was in the ball park at the time, but we had no idea Sony was playing with the technology at the time. We were equally unaware of the fundamental work in university labs two decades before.  

 

It's interesting to think about areas of technology under rapid development and the importance of missing technologies. Television is a great example.

 

Television has been around for a long time.  Philo Farnsworth's all electronic television in the 20s was probably the most important breakthrough.   The lens of a camera focued an image on a photosensitive plate while an electron beam was scanned row by row across the imaged area.  A signal of strength that varied with the brightness of the region the beam was on would be transmitted as a radio signal to a television receiver.  The receiver had a cathode ray tube that, in lock-step with the beam in the camera,  scanned a piece of glass coated with a phosphor that would glow when the beam hit it with a brightness that changed as with the signal the camera was sending. All of this was done at about thirty frames a second - good enough that your mind perceives motion.

 

Television was such a great name - viewing a moving scene at a distance as it happened.  Shortly after WWII a few television networks sent their television signals between cities and then across the country.  Even though the receivers were expensive it quickly caught on and the networks were spending serious money on production. But there was this problem with time zone.  You put a program on at 6 PM in New York City and the television receivers in Los Angeles would display the image at 3 PM.  Not exactly good for creating the largest potential audiences.

 

Electronic recording would fix the problem, but wire and tape recorders were just getting good enough for audio and television signals had about a thousand times as much information.  It just wasn't possible.   The solution seems a bit crazy now - point a film camera at a television monitor and immediately develop the film. Eastern and Central time zones were live, Mountain and Pacific were delayed by two hours.

 

At first these kinescopes used  16mm film.  Once used most of the film was thrown away it degrades and is a fire hazard..  By 1954 television kinescope delayed broadcast became the largest consumer of movie film eclipsing Hollywood and the home movie markets combined. A bit before  I Love Lucy became a huge hit show.  Lucile Ball demanded higher quality so a switch was made to 35mm.    The larger format was so expensive and had to be stored in newly constructed fireproof facilities.  An archive began to emerge. 

 

A few years passed and Ampex came out with a video recorder able to handle the flood of information. - a monster that used two inch wide tape.  An expensive and initially temperamental monster that used two inch tape, but dirt cheap compared to the  kinescope process.  It quickly became the standard for time zone delay. The curious thing is that the television people didn't see tape as an archive medium.  For a few years none of the tape delayed shows remain as the expensive tapes were reused until they wore out.  No one thought long term storage was important.

 

The last part is interesting.  We don't think of television as remote viewing - rather it's a video stream that might be live, time delayed, manipulated,  archived, or all of these. 

 

Oh .. the prediction from 1993?   You will take a flat piece of glass out of your pocket and have a video conversation with a relative in India and the "call" won't cost anything.  We knew how to code video, mobile radio would easily have the bandwidth in 15 or 20 years, multi-touch screens were in the labs, lcd color screens and lcd cameras were only going to get better, the price of bandwidth was going to crater (our corporate management didn't like that message) ... Moore's law on everything but the battery.   Note:  I've made a few good predictions, but looking back at my old lab books can be an embarrassment - you tend to forget the dead ends.  

 

 

connecting the titanic, sputnik and dealing with forgetfulness

1955, Fred Whipple had just been appointed to head the Smithsonian Astrophysical Observatory in Cambridge.  The International Geophysical Year was less than two years away and he wanted to involve citizen scientists as much as possible.  Whipple was one of the few who realized a satellite was likely and set about creating a web of radio operators to monitor and relay transmissions from satellites along with groups of amateur astronomers who could optically track their orbits.  Operation Moonwatch was formed and by Sputnik's launch over 200 teams were in place. 

 

Tracking a satellite was done with a clock and a low power telescopes that could read azimuth and declination.  The National Bureau of Standards time signal broadcast on WWV in Fort Collins, Colorado used as the reference clock and most people into amateur astronomy had a shortwave radio. Readings from stations separated by hundreds of miles were sent by amateur radio operators to a computer at MIT that gave the project a bit of its precious time.  More clusters of readings would come in from around the world and a model of the satellite's orbit generated an ephemerides - predictions of where it would be. Later specialized cameras were developed by Bell Labs to automatically generate satellite ephemerides as well as tracking ICBM tests, but Moonwatch was a success and lit the imagination of a few people.  

 

In physics it's common to turn a problem on its head to gain a bit of insight.  Operation Moonwatch used known positions on earth and an accurate clock to measure a position in orbit.  What if you put synchronized clocks in space and measured how long it took radio signals to reach a position on earth?  With enough satellites with synchronized clocks you could triangulate a position anywhere on the Earth's surface.  The technical details were difficult details, but the basic idea was simple.  

 

It turns out accurate navigation is very difficult and expensive and a limiting factor in fighting the cold war.  It was so difficult that as technology progressed ARPA decided to push ahead with a satellite based global positioning system - the inverse of Operation Moonwatch.  It was a bargain for what it gave to the military.  Finally a good enough signal was made available to civilians and navigation changed.¹ 

 

GPS doesn't work everywhere.  It would be nice to track things indoors and at a lower cost. Bluetooth and WiFi are currently used -  usually measuring signal strength from several transmitting beacons and triangulate a position.  Sometimes it's just proximity to a single beacon - a lighthouse model.   The approaches aren't very accurate - three to ten meters is common.  Good enough to track your phone as you move around, say, in a mall or subway, but doing better would be nice.

 

This is where the Titanic comes in..

 

On April 15, 1912 the Titanic struck an iceberg.  Frantic radio signals went out, but there was confusion among rescue vessels and people on land trying to coordinate the rescue effort.  Sometimes it takes a disaster to create order and an effort to regulate radio frequencies began soon after. Transmitters were assigned narrow swaths of the electromagnetic spectrum that were large enough to effectively communicate with receivers that could tune to those frequencies. Regulations changed as larger chunks of this resource were needed - AM radio required more than Morse Code, FM radio even more and television much more. Government would parcel out these increasingly scarce resources, but power was also regulated and it was possible to give much larger chunks to low power short range communications - WiFi and Bluetooth are examples.  Most of radio engineering tried to pack as much information into as little spectrum as possible and to build transmitters and receivers that could be accurately tuned to specific frequency ranges  If you waited for some technology to catch up there was another way (actually several, but I'll only mention one) way to do radio.

 

You may have heard about ultrawideband (UWB).  In the mid 2000s it was supposed to be a good way to send large amounts of information over short distances.  Cheaper and higher bandwidth than the WiFi of the day. Often existing technologies are improved with scale and an embedded base making it difficult for new players.  Sophisticated coding technologies and enormous penetration (WiFi has greater penetration than smartphones) cemented WiFi as the prime high bandwidth low range wireless connection.  In a few years UWB went from redundant to lagging. Many thought it was dead, but it had a few advantages in some use cases.   

 

Here's how it works.  Rather than sending out a precisely tuned narrow band frequency, UWB sends out narrow (in time) digital pulses that "splash" over a very wide frequency range.  The signals are usually only one or two billionths of a second long and aren't sent that often - a hundred thousand to ten million a second are common rates. The frequency range is so wide that usually at least some of it can pass through objects that block signals like GPS and WiFi.  It's also so weak that narrowband communication can't detect it.²

 

 Imagine scattering a few UWB transmitters around where you know their location with some accuracy.  Now start transmitting synchronized pulses.  A receiver notices the time of arrival of each signal and, using the fact that light travels about 30 centimeters in a billionth of a second, can triangulate it's own position.  Accuracy is about ten centimeters - about four inches.

 

A use case I'm familiar with is finding the locations of players in a beach volleyball game.  In international FIVB games a very small (less than a half ounce) UWB device goes in the back of each player's  shirt or sports bra.  Their positions are tracked within ten centimeters about a hundred times a second and accelerometers and a compass send out rotation and orientation information on the UWB signal that the announcers, coaches and sports science people use.  Four transmitters are used and all four players are easily tracked.

 

It would also be useful for augmented reality and Apple has included an UWB chip in every iPhone 11 - the greatest deployment of the technology so far.  For now it's being used for data transfer, but wait a bit.  Someone could also make very cheap ($10 or less) postage stamp devices that might be powered by available light that could be location tags.  Put them on your keys, glasses, remote control .. anything you tend to lose track of.  We went through trying to manage dementia with my mother.  I would have paid a lot for this capability.  And thousands of other uses from amateur sports to tracking boxes accurately in warehouses.

 

There are probably some bad things that can be done with it.  We need to be thinking these things through, but our track record isn't exactly good. 

__________

¹ It also gives a very good frequency standard and a timing signal.  Without this our mobile phone system, high frequency trading and a host of other core technologies we depend on would break.  This is an issue as GPS is vulnerable to attack and we've done almost nothing on backup technologies. 

² The FCC defines it as taking up at least 500 MHz of spectrum somewhere from 3.1 to 10.6 GHz.  Power is under -41.3 dBm/MHz .. or 75 nanowatts.  Since it only deposits a tiny bit of power in any narrowband frequency, it's just below the noise floor of a narrowband receiver.  Fortunately it's straightforward to build a receiver that can detect the signal.