Don Lancaster's What's New? 2008 Archive Blog

Pick your blog year...
1997	1998	1999	2000	2001
2002	2003	2004	2005	2006
2007	2008	2009	2010	2011
2012	2013	2014	2015	2016
2017	2018	-----	-----	-----

One of the handier "rules of thumb" that
sometimes apply some of the time and can be
enormously useful is this:

Very often, one percent of what happens
nationally happens in Arizona. And one percent
of what happens in Arizona happens in the Gila
Valley.

Thus, roughly, there are 300 million people in
the US, 3 million in AZ, and 30,000 locally.

While not super accurate, this rule can quickly
give you a rough estimate of an amazing variety
of events or tasks. Where you otherwise may
not have the faintest clue as to scale.

Naturally, the "rule" does not apply to anything
with a regional bias. I suspect Thatcher has
more cotton module fires than Bangor, Maine
does. And that walrus attacks may be rare in
Nebraska.

Supporting small electronic parts for eBay
scanning can get tricky. Sometimes simply
grabbing them with needle nose pliers will give
you enough flatness and mass for stability.

You can always take out the pliers with your
image postproc. To get fancy, you could add a
rubber band to the plier handles. Or actually
use a real hemostat.

The pliers ploy also has some nnobvious uses.
Such as making sure an object is horizontal.
Or showing correctly positioned lettering.

Never store carbide in a non-locking carabiner.

Expanded and improved our Gilia Hikes library page.

Not sure how many hikes total we will end up with,
but anything beyond 150 is proving difficult. At
least within our guidelines of "mostly easy drives"
from the Greater Bonita-Eden-Sanchez metropolitan
area and in fact a one day project.

BLM is actually working on a pair of new trails,
A Cottonwood Trail to loop the flying "W" with
the watchable wildlife platform and Bonita Creek.
And Arizona's first Water Trail providing a short
intro swimming, kayaking and whitewater route.
These are not quite listable yet.

BTW -- The "Flying W" has nothing to do with
ranch brands. That is the shape of the steel
supporting roof structure on the group picnic site.
And is far beyond instantantly obvious.

Please email me with any suggested further hikes.
No, not that one.

Once you have your grazing permit, you do have
have to actually eat the grass.

Such despicable stunts as land stewardship,
allotment retirement, or range improvement are,
of course, specifically forbidden by admin decree.

Your tax dollars at work.

Many ranches appear to be for sale on or near
Mount Graham. Some for as little as $45,000.

Only it is not clear to me exactly what you are
buying. Very often, there is no deeded land. The
fences and corrals may need major repairs, the wells
and springs might range from low to dry, and the
tanks and windmills often need major work.

You do, of course, gain the right to steal grass off
of the state or feds with grazing fees whose rates
are utterly, obscenely, and unconscionably low. And
basically comparable to outright theft. These are
typically only valid for a few months out of the year.

Grazing fees are approaching one tenth of their
admin costs and have recently been slashed.
And half of that is routinely rebated.

What you do with your cows ( the few dozen that are
supportable ) the rest of the year is up to you.

There is, of course, no grass because of the drought.

Thus, the payoff these days on both stock markets is
about the same.

I've been revisiting some of our local indian
ruin sites. Despite competent researchers,
the Gila Valley has lacked an obvious champion.
Yet its population and influence apparently
were among the highest in the American
Southwest.

The area is confusing to say the least, as
pottery styles represented Hohokam,
Salado, Anasazi, Mongollon, and Mimbres,
among others. Clearly complex trading
routes went through the area.

There were elaborate irrigation systems.
Both from the Gila River and riparian
side canyons. Not to mention many thousands
of water enhancing grids that may have been part of
an elaborate aguave to mescal booze factory.

Or possibly the world's very first paleo Dilbert
cubicals.

Many of the sites have been outrageously
pothunted. Others have been obliterated by
anglo constuction.

Modern agriculture has also  trashed the tops
of many sites. But they may lie pristinely buried,
awaiting subsurface radar techniques for modern
excavations. When all is said and done, plows do
not normally go all that deep.

Some interesting reading appears here, here, and here.
More on our Gila Hikes library page.

I've always wondered why we do not have
more heat pumps for water heating apps.

With low temperature differentials, a heat
pump can be as much as five times more
efficient than electrical resistance heat.

Sadly, as the temperature differential
increases, the COP drops dramatically.
Breakeven is around a 50 degree differential
and an electrical resistance backup has to be
provided for higher differentials.

A hot tub or a spa would seem to be an ideal candidate
for a heat pump. With a potential energy savings of
as much as 80 percent. But the available heat pumps
cost something like $3000 custom installed. At a dollar
a day resistance heat costs (10 kw heater for one hour
at ten cents a kilowatt hour ), the energy breakeven
would be something like ten years. Or longer than
a typical spa life.

It also seems faintly ludicrous to put a $3000 heat pump
on a $2500 hot tub. "Ten dollar horse and forty dollar
saddle" comes to mind.

Still, an integrated heat pump should add no more than
$100 or so to a high volume tub manufacturer. And
might pay for itself quite rapidly.

Meanwhile, there is a new $500 plus installation
heat pump for ordinary hot water heaters. But it
seems to me this system has a grievious flaw:
Since the air is cooled, the amount of space
heating has to go up during fall, winter, and
spring months. For an apparent ZERO benefit.

If the air being cooled is outdoors, the efficiency
will often be far lower. And even during air conditioning
season(s), the benefit is not clear, since the hot
water energy may be released into the air conditioned
environment.

Things are rarely what they seem when it
comes to energy alternatives.

The four "R's" of Arizona politics: Rightwing,
Racist, Reactionary, and Redneck.

In Arizona, a "political moderate" is anyone
who is moderately to the right of Atilla the Hun.

Updated and improved our GuruGram library page.

Seldom discussed is the fact that a pv energy farm
needs very little water. Which makes them a near
perfect use for government and indian lands in
the arid southwest.

If you start doing an economic analysis based
on highest and best use, a case can be made that
any site with adequate water is an inappropriate
location for a pv energy panel.

The rescheduled date for our BLM brown bag
lecture on 150 Gila Day Hikes is Tuesday
January 13th from 12 to 1 pm.

Much of the talk will be based on our Gila
Hikes library page.

The program is free and you are certainly
welcome to attend. Bring your own lunch or
snack and a drink. Location in Safford is
in the BLM conference room at 711 14th
Avenue. This is the corner of 14th Avenue
and 8th Street.

There seems to be a revival of interest in
wireless power transmission. The bottom
lines remain that most of Tesla was utter
bullshit, and that severe issues are likely
to forever remain with distance, efficiency,
safety, and sensor comm.

There are three probable approaches. The
first is to use inductive coupling via an air core
transformer. This is the approach used in electric
toothbrushes. The coefficient of magnetic coupling
in air is absurdly low, so distances of even a fraction
of an inch will totally trash efficiency. The amount
of power transferred drops as the fourth or fifth
power of distance beyond a very close in point.

The second approach is to go to a resonant system
similar to a Tesla Coil. Power levels as high as
60 watts have recently been shown. The coils have
to be physically large and their orientation is
critical. Again, efficiencies are low.

It is interesting to note that there was a well known
but little used resonance phenomen. If you place a
large resonant coil near an am radio, you can dramatically
increase the sensitivity if you tune the coil to the same
frequency as the station being sought. The large size
and awkward tuning largely prevented this from seeing
much popular use.

A third approach is to use a focused beam, similar to
a flashlight or a laser. The antenna or mirror doing
the focusing has to be many wavelengths larger than
the energy transferred for a minimially divergent beam.
Orientation of the receiver is super critical, for there
will be near zero power outside of the beam.

For even modest amounts of power, there are very real
safety issues with anyone coming near the apparatus.
In addition, because the efficiency is very low, some sort
of two way comm would be needed so that the system is only
active whenever the receiver (such as a laptop or cell phone)
is in need of power.

A practical result may be a foot square plate that you place
a laptop or cellphone or whatever on for recharge. Anything
larger or needing longer distances is likely to remain fictional.

I was asked what the "next big thing"
will be in high tech. It seems like the time of
year for predictions, so here goes...

   NET ENERGY PV PANELS -- The
   technology and learning curves are now
   finally in place for emergence of photovoltaics
   that are genuinely renewable and sustainable.

  WEB TELEVISION & WEB VIDEO --
  The advantages utterly overwhelm
  traditional cable and over-the-air broadcasts.
  There is no point for either to continue.

  USEFUL eBOOK READERS -- So
  much better than the traditional touch
  and feel of a book that books themselves
  will virtually disappear in an astonishingly
  short time.

  ROUTINE GOLDILOCKS EXOPLANETS --
  First by the dozens, then thousands. Caused
  by ongoing hitech breakthroughs and vastly
  expanded research.

  GYROS AND GPS -- extending to the lowest
  cost, simplest, and least elegant products.

  HD VIDEO AS A WEBSITE NORM --
  Creating a humongous bandwidth crunch
  and seriously impacting the very future of
  the web.

  TOTAL DEMISE OF TRADE JOURNALS --
  Along with most other magazines and newspapers.
  The economics are now ludicrous.

  IMPROVED WALL WARTS -- Modules only
  slightly larger than an ordinary ac wall plug and
  standardized "one size fits all" wireless schemes.

  UBIQUIOUS WiFi -- Free internet access at
  ultra high speeds available nearly everywhere
  all the time.

  CHEAP SYNCHRONOUS INVERTERS --
  Revolutionizing the economics of solar
  electricity. We are talking $9 here, not $2500.
  Including new max power optimization algorithms.

  EYEBALL SIPHONING -- New social web
  services continuing to drain interest in
  classic websites and most everything that
  has gone before.

  DISAPPEARANCE OF MOST CLUBS -- Caused
  by aging membership and easier web access to once
  priviledged information and resources.

  HDTV/MONITOR CONVERGENCE -- Since
  most television and video will be web distributed,
  the utterly abritrary distinctions betwen HDTV
  television sets and larger computer monitors should
  vanish entirely. HDTV's in particular must end
  up with greatly improved web nav.

  CHEAP SANTA CLAUS MACHINES --
  Dramatic price reductions are long overdue in
  three dimensional printers that can produce
  anything from a roast beef sandwich to a new
  girlfriend.

Google Chrome is now out of beta and has gone
mainstream.

Two of its features include a single window for
website access and searching; and blindingly
fast JavaScript.

Our fast Magic Sinewave Calculators now
appear to be instantaneous for many option
choices.

My own preference is to stick with Firefox
for all general web use and reserve Chrome
when ultra high JavaScript speed is needed.

Double Exposure techniques can be enormously
useful during image postproc. Typically, you can
go into Imageview32 or whatever and make a
second file whose image has its gamma, brightness,
or contrast modified. Or has been rotated.

In this example, the meter face is a double exposure
of darkened brightness and raised contrast. The
original was badly burned out. A modest amount
of retouch was also done to improve the appearance.

It is of crucial importance that you get the "blend
points" just right. Typically, a murky gray will
do the job. As few as two or three "wrong" pixels
can trash your final result, so extreme attention
to detail may be called for here.

In this example, the knob tops were much closer to
the scanner than the rest of the front panel and
burned badly. The gamma and brightness were both
reduced. The inner knob ring was hand adjusted
for minimum artifacts.

The individual knobs were knocked out to white
and Paint's "white-as-transparent" mode was then
used to overlay the second exposure.

In this example, the snap and the jack were not
quite vertical on the original scan. Each needed
rotated into a different vertical position. Their exact
position heights were also adjusted so they appear to
create a consistent image.

Normally, both exposures must be at the same
scale. Watch this detail!

Google Maps has just added Gila Valley
street views to their service.

As near as I can tell, the actual mapping
was done in early March of 2008. Judging
by a completed motel, empty cotton fields,
and pecan trees just barely leafing.

Satellite coverage remains much earlier
and some of it is uselessly low resolution.
It sure would be nice if Google included USGS
topo maps as an overlay option. Most other
topo sources are inconvenient one way or
another. Particularly when panning.

The Google street views seem to be catch as
catch can. Very large areas ( such as northern
and eastern Pennsylvania ) still seem to have
zero coverage.

The way to tell if you have street views of the
present screen is if the "view guy" is orange
( views present ) or white ( views not preseent ).

One immediate use TFD will have for the Thatcher
street views is to improve our sprinkler hookups
training program.

One of our recurring themes around here
is bashing pseudoscience.

Topics have included water powered cars
( an outright scam ), pv panels ( all remain
gasoline destroying net energy sinks ),
Tesla Turbines ( that demand inefficiency to work
at all ), on-board vehicle hydrogen ( a dynamic
brake in disguise ), the magic lamp ( incompetent
rms measurements run amok ), electrolysis
( guaranteed useless by exergy ), corn ethanol
( a worthless twelve billion dollar vote buying scam ),
and Brown's Gas ( bad labwork mixed with unusual
but expected results ), among many others.

There are, of course, other sides to these issues.

Often hosted by few-chips-shy-of-a-full-board
types who may be extreme litigants, paranoid,
nasty trick troublemakers, investors in denial,
just plain stupidly clueless, or worse. While others
may directly confront these types, I go out of my
way to stay "two levels" away from them.

By never directly confronting any of them, never
mounting an ad-hominum attack, and always
trying to present factual and unarguable fundamental
physical law info. That can be independently verified.

Otherwise, you may become what you are attacking.
"They", of course, will never be convinced. The message
should be aimed as education for others who might
get sucked into their ludicrosities .

The controller of choice for ac line operated
dimmers and such is often a triac. Triacs let
you control current in both directions and are
based on a gate pulse turning the main ac current
on and the next ac line zero resetting.

Shock hazards in such circuits are often
minimized by using a "baby" triac optoisolator
for triggering. Such as a MOC3010. These
same devices are ideal for high power ac
interface to a computer or PIC port.

It is important to note that there are two
"flavors" of triac optoisolators - zero
crossing switched devices that eliminate
transients, glitches, and pops, or non-zero
switched devices that are essential for lamp
dimmers or other phase controls.

If you are buying a triac based solid state
relay, it is super important to get the right
type ( zero switched or not ) for your use.

But triacs are hard to find in sizes above
25 amperes. Higher current needs will often
call for a pair of back to back Silicon Controlled
Rectifiers (SCR's) or an equivalent module.

But there is a crucial difference between
back-to-back SCR's and a triac that can
easily cause you grief: The allowable REVERSE
gate voltage on a SCR is often limited to
5 volts or less!

Thus, you should always provide some sort
of external gate protection diode. Or ( better yet )
always guarantee that only forward current
is provided in your gate circuit(s). Ready to
use SS relays will often provide this protection
internally. Plain dual SCR modules often do not.

Another rude surprise is that the gate sensitivity
of larger SCR's is often quite low. It is not unusual
for a 100 amp SCR to require 150 ma or more of
gate trigger current.

A critical analysis of where future energy
sources are going to come from appears here.

My own views are that both pv solar and
concentrated solar thermal are near to becoming
cost effective. And may be within a decade or so
of delivering net energy. Ignoring the staggering
costs of amortizing monumentally stupid subsidies,
of course. It would  seem mostly a matter of
engineering and learning curve economics.

For instance, there is no reason why a synchronous
inverter should cost more than $9. Let alone $2500.

And that the only reason for corn ethanol enthuasiasm
with its ludicrous conversion efficiency is that the Iowa
primary is earlier than other states. Creating a twelve
billion dollar vote buying scam.

I feel there is an utterly overwhelming advantage
of pv over nuclear -- pv adds zero ( or at most
very little ) new heat energy to the planetary
heat budget. pv solar also does not raise issues
over how the next five Cheyrnobles are
ging to turn out.

Besides decommissioning far easier.

It turns out that there really is a patron
saint of lost causes.

Received a second community college piece
of test equipment that was suspiciously easy
to repair. This one a power supply with "blows
breaker" written on the cover.

A visual inspection revealed smoke and fire
residue at the 110/220 selector switch. After
cleaning and testing, the fault apparently
self-cleared and things seem to be working
properly.

Just to be sure, the switch was replaced with
a 110 vac hard wired connection.

"Input overload" conditions in general can be
a bear to work on. As always, start with a
careful visual inspection. Then an ohmmeter
check, remembering that power transformer
primaries are only a few Ohms dc at most.

Next, get a variac. Preferably a fused or circuit
broken one with a current meter. Bring the input
voltage up to just the few volts needed to cause
the current to rise beyond what is reasonably
expected. Then use an oscilloscope to check diode
drops and electrolytic capacitor charging. '

I'm often aseked for my views on copyright,
having long been on both sides of the issue.

Copyright for a corporation should be
strictly limited to eighteen months and
not be nonrenewable.

Copyright for an individual should be
thirty six months and renewable once
after critical review.

The same should go for patents or music
or movies or videos .

Test equipment from aerospace auctions tends
to be super clean and well cared for with few
refurb surprises. Mil surplus stuff is almost as
good in that it is either well cared for or obviously
and utterly trashed.

But community college surplus test equipment
can include rude surprises. There can be intentional
or accidental damage. Either from students or
from incompetent repair attemps. And things may
rarely be what they seem.

The rule here is to always expect the unexpected.

Breadboarding systems are particularly suspect
because of the ease with which external damaging
voltages and currents can be applied. I got one
of these that had both the +12 and -12 volt supplies
with zero output. And due to a poor design, this also
disabled the internal function generators.

Surprisingly, the regulators checked out OK, but there
was no input voltage to them. Yet the raw dc supplies
worked just fine. Turns out there was a hairline crack
on TWO pc board traces. Along with a lifted bulge in
a third trace.

Suspecting vandalism, I inserted some 150 ohm resistors
in these traces to safely limit currents if other components
were trashed. Only to observe normal operation. Shorting
the cracks eliminated all problems. Or at least appeared to.

I've reluctantly concluded this was in fact an original
defect in the pc board manufacture that eventually became
a problem with age or flexing or whatever.

If you were to cut a pc trace during debug, you should,
of course, make the cut wide enough that it is obvious
and cannot inadvertently close by itself. Neither was
the case here.

The rule is to always assume student vandalism or
incompetent repair attempts. Then try to prove
yourself wrong.

Once again updated and expanded our Gila
Valley Day Hikes library page.

New additions include Grantham Cave,
Bramahan Cave, Gila River Whitewater,
Tollgate Road, and No Name Tank.

An outfit called Briel Computers now
has replica Apple I computers and Kim I
trainers available. They shortly will also
offer an Altair clone.

The Kim 1 was by far the finest of all
trainers of all time. We sold out of the
few we were able to obtain. But we
do still have some newer Heath Zenith
ET100 trainers and Apple IIGs computers
offered on eBay.

These remain hard to find and highly
collectible.

It usually pays to completely ignore any
customer who tries to nickel and dime you.

Chances are that, besides costing you time
and money, they will cause you more grief
later on.

We basically have a ten dollar minimum
charge on eBay. Which is around one half
of the normal breakeven sale price.

If you want less than our minimum order
quantities, an $89.95 pissing around surcharge
becomes immediately due.

We use the most appropriate shipping services
and absolutely insist on tracking being avaiilable.
Cheaper services cost much more in the long run.

We gladly combine shipping charges.  And your
best way to reduce them is simply to order more.

We are in the process of updating our
GuruGram Library listings. They should
end up more or less up to date sometime in
the next few days.

We continue to get requests for our water
soluble swimsuits. Which we long ago sold
out of. After long being the finest eBay
supplier.

The original source was mil surplus water
soluble hospital bags on a one time buy.
The supplier to the miilitary has long ago
also vanished without a trace.

However, these remain readily available
through the usual Google Search. The
key terms to use are PVA and poly vinyl
acetate. Kiddy slime is a related product.

They are not the greatest Bay item to
sell because they are rather heavy.

This Wall Street Journal article points out that
54% of US car manufacturers do not have any
trouble with sales or production and have not in
any manner requested any bailouts or help.

The bailout appears to be an outrageoust scam
of, by, and for losers who deserve to be flushed.

Such a bailout would severely penalize the winners.

A useful listing of 100 free digital artist web
tools appears here.

I was suprised that my favorite
ImgViewer/32v2.6 was not included.

Our own free image processing tools
are reviewed here.

Updated and expanded our Gila Valley
Dayhikes library page. There's now
over 150 listings.

Tentative date for the BLM brown bag
lecture is thursday January 7th at the
usual BLM conference room at 14th
street and 8th avenue in Safford.

I was asked for my favotire HazMat yarns.
In answer to an ever diminishing number of
requests, here goes...

   The hazmat rule of thumb: Hold your
   thumb up with your arm extended and
   close one eye. If you can still see the
   scene, you are too close.

   The all time "best" hazmat incident:
   The great Boston molasses disaster

   My personal "best" hazmat incident:
   The great Kentucky Fried Chicken
   gravy explosion. The original 20 gallon
   pot fire was put out simply by placing
   a lid on it. Much later and long after
   we thought the incident was over, we
   were pouring the gravy into a ditch
   in the back yard when it flashed
   spectacularly. Caused by oxygen in
   the air hitting hot buried oil.

  Neatest HazMat training effect:
  The Christmas Tree that creates a 20 foot
  diameter propane Dante's Inferno. Only
  to reveal spectacular ICE CRYSTALS (!)
  afterwards. The rapidly expanding propane
  freezes out moisture from the air under
  certain humidity and temperature ranges.

  Our best TFD "Golly Gee Mister Science"
  Our cotton module fires create their own
  oxygen through low grade pyrolysis. And
  require no air to continue. We fight these
  by the "Dentist Method" just like removing
  decay from a tooth. There are two teams,
  the "pluckers" and the "chuckers.

  Our scariest TFD hazmat situation:
  The HUNDREDS of sulphuric acid tankers
  PER DAY that go through our town. There
  have been spills and fatalities and it is CERTAIN
  to get a lot worse.

I continue to be amazed that it is not
possible to buy a television set!

At least not an economical LCD one in the 32
to 37 inch size that displays all web video
to 1080P.

Besides the outrageous pricing, the
apparent flaw in the Sony Internet Video Link
is that this is a buy-in controlled content service
that only works on the tiniest fraction of
available web video.

There are many conflicting goals beneath
these What's New? blog entries. To the
point where it is enormously difficult to always
have one new entry each and every day.

The obvious main purpose is to keep you
readers coming back for more. A hidden
purpose is to be an outline for future upcoming
work. And to catch any gross errors before they
become locked into a GuruGram or other paper.

I try to have a mix of new developments, energy
and electrocity debunking, humor, news, eBay
stuff, and obscure useful web links elsewhere .

And logicstics make it more attractive to
do three or more entries at once, rather than
single daily ones.

Your comments welcome on how these entries
can be improved.

Proponents of local pv power generation
( it ain't EVER happened ) tend to overstate
the amount and the "badness" of power
transmission line losses.

In reality, these losses are typically around
seven percent. And are an actual asset if
the distant baseload power station is enough
cheaper than local sources.

Despite urbanization and Californication,
the sounds of the old southwest remain if
you know when and where to look for them.

The top three would be the 3 am yip yarfing
of coyote pups, the creaking and groaning
of an Aeomotor windmill, and, of course, the
call of the Canyon Wren.

Purists might add the posture clicking of a
Coati Mundi or the rattle of a sidewinder.

One of the ruder surprises of eBay
is that the charges are based on the
total quantity offered rather than on
an individual sale.

Thus, the eBay charges can be outrageous
on high quantity auctions.

Here is how we deal with the problem:
We find the maximum quantity whose total
costs would be $49.99 or less and list them
at auction.

Then we place something like ONLY SEVENTY
FOUR REMAINING! in the auction description.

Someone who wants them all buys the eBay offer
off eBay and the rest direct.

More on similar topics on our Auction Help page.

Sony seems to have an alternate approach
to merging HDTV sets and web video
availability with their Internet Video Link
accessory.

This is basically a dedicated personal
computer that accepts a cable input and
outputs to 1080P rather than VGA
video standards.

Other manufacturers are certain to follow
suit.

Some recommendations for assertiveness are...

   1. The best way to get something to happen
       is to ask for it.  "Please" is a power word.

   2. The best way to stop something from
       happening is to just say no.

   3. The best way to defuse an awkward
       situation is to tell them how you feel.

   4. Exactly repeating yourself a second
       or third time can be devastatingly
       effective.

   5. Should things prove obtuse after three
       attempts, use ridicule: "I'm sorry about
       your hearing problem".

The best way to tell if your assertiveness
is adequate is if you have just been
arrested for aggravated assult.

The latest entry in the Browser Wars seems to
be Google Chrome. Now freely downloadable
in beta.

Two of its features include a single window for
website access and searching; and blindingly
fast JavaScript.

Our fast Magic Sinewave Calculators now
appear to be instantaneous for many option
choices.

Managed to find a rare General Radio GR
1862B Megohmmeter that measures as high
as 2,000,000 Megohms (!) for such things as
insulation resistance measurements. Besides
being a classic collectible.

This is the -B version that retains the original
black alligator livery and includes the
classic old style Weston meter movements.

It is reasonable to ask how much image
postproc is acceptable. One criteria is
the amount of postproc used in, say, a
Newsweek ad.

But there is a continuium between a gal
using makeup on a date and putting lipstick
on a pig.

Unless you are dealing with a collectible,
the criteria should be: Does the image
clearly show what the object is? Anything
that sharpens or improves background
should be legit.

As should anything the improves any
artifacts of the photo or scanning process.
Such as more legible lettering or making lighting
more uniform. Or sharpening focus to one
pixel accuracy.

Photography tends to emphasize flaws for
several reasons, so a Take half and leave
half approach to any defects can often be
appropriate.

It is also reasonable to defer any refurb or
final cleaning until the item is actually sold.
You can remove unwanted labels and such
a lot faster in postproc than you can in
an actual cleaning. Thus showing the object
as you expect it to be, rather that how it
happens to be now.

My main criteria on postproc is the more
the better. Consistent with the value of the
object being sold on eBay, the number of
items available, and the expectation of
future inventory.

Here is yet another example of how important
image postproc is and what can be done with
it. As in before and after.

Black on black does not make for the best
photo subjects.  Normally, this would be
scanned rather than photographed, but the
ears on the instrument sides made for
a useless three inch depth of field.

The first correction was a tilt adjustment
to make the sides perfectly parallel. A
small portion of background was replicated
for the total fill. The best edge was chosen
and enhanced. Then "chased around" the
entire instrument.

A double exposure was used with the meter
face having increased gamma plus retouching
of the burned out areas. The instrument itself
had its gamma reduced (blacks expanded).
This dramatically improved such things as
the main control knob and brought out the
texture.

But still was too far into the black on the
right side. Some good background was then
cloned left to right to unify the entire front.

A minor amount of lettering retouch was
done. Most of it was "good enough" on
final reduction and sharpening to not
need a total relettering. Although this
image would be a good candidate for
a total reletter if there was more than
one going to be offered on eBay.

Many of our image postproc tools can
be found here.

The latest GuruGram #101 is on A PV
Photovoltaic Panel Intro and Summary .

It is a "slideshow" type lecture style synopsis
of our earlier Energy Fundamentals and
More Energy Fundamentals papers and was
created with our PostScript PowerPoint
Emulator.

Sourcecode is available here.
And additional GuruGrams here.

One of the "sleeper" calculation techniques
kicking around the outskirts of computerdom is
using spreadsheets for distributed computing.

Here is one third party example.

Someone suggested spreadsheet computing could
also be used to solve the electromagnetic fields
using the rebounding technique of this GuruGram.

While this certainly seems like it would work, you
would lose the stunning graphics capabilities
of working directly in PostScript. Clipping to
areas of interest would also be difficult for
more complex geometries.

Additional GuruGrams are found here.

The ancient oriental art of TI WUN ON consists
of getting totally snockered but always doing so
in a professional and workman like manner.

On one hand, I am saddened by the imminent
demise of the trade journals. On the other,
their author payment has been obscenely low for
years and their technical content has become
abysmally bad. As has their telephone harassment.

Journals I still follow include Electronic Design,
Machine Design, E.E.Times, LED Journal,
Battery Power Products, Sensors, EDN, and
Machine Design.

I don't expect any of these to be around in a
few months.

One of the curious hassles of eBay selling
is that complaints and problems are inversely
proportional to the cube of the selling price.

Thus, you will have far more problems with
$9 sales than $900 ones. A good strategy
is to never sell anything for less than $49.

Several local auctioneers seem to be
reverting to sealed bid sales. These
have their own sets of gotchas and
opportunities.

At least one surplus source is a
stickler for exact conformance with
what gets sent in how. They seem to
delight is voiding submitted bids.

A preview is far more important with
a sealed bid. Which might mean you
need two trips to a distant town to
participate.

Chances are your final prices on any
in-demand item will be much higher.
Which means you should bid higher
to compete. And that you will win
fewer items and pay more for them.

Sealed bid sales often may attract
far fewer bidders. One useful ploy is
to ridiculously lowball bid everything
you could possibly make use of.
Chances are good that you will be
the only bidder on at least a few items.

The latest GuruGram #100 is on An Energy
Fundamentals Intro and Summary .

It is a "slideshow" type lecture style synopsis
of our earlier Energy Fundamentals and
More Energy Fundamentals papers and was
created with our PostScript PowerPoint
Emulator.

Sourcecode is available here.
And additional GuruGrams here.

We are flushing some of our eBay items that
have been very slow selling and have offered
them for one time ultra low prices.

Any of the closeouts remaining will end up on
the Alvin Pile.

There seem to be several annoyances
from CBS and others on our new Video
Downloads links.

The first of these is to drop (or omit)
the full episodes in favor of confusing
and uselessly annoying clips. The point
in this totally eludes me.

And the second, of course, is leaving a
giant pickle on the screen after the
ad that you have to manually remove
if you want to watch the rest of the show.

No, I am not making this up.

Slashdot recently had a review of places to
go to get the mechanical stuff for projects.

We did an ancient review on a similar topic
here. Start on page 37.

Obviously, these days, you start with eBay
followed up by online suppliers found thru
Google.

Being on a first paw basis with your local
junkyard dog remains important. Share
an occasional bowl of Purina together and
discuss the aerobic exercise potential
of Felis Domesticus. A decent local
hardware store is also of major value.

Sources I routinely use and find extremely
useful include McMaster-Carr and Small
Parts. Plus W.W.Grainger for all things
electric.

Sadly, many brick-and-mortar surplus
outfits are failing or pricing themselves
way out of the market. I still find Apache
Reclamation and Surplus useful, though.

It is reasonable to expect any seller's eBaysales
to to drop sharply in the near future. First and
foremost because there are likely to be fewer
buyers. Secondly because competitors will likely
be expanding their store offerings. And thirdly
because there are likely to be scads of new
sellers.

Some defenses might include...

     DOUBLE STORE LISTINGS - The
     more you list, the more you sell.

     FLUSH THE LOSERS - Get rid of
     anything that is not going to sell
     anyway.

    IMPROVE DESCRIPTIONS - Make
    sure your ad copy is accurate and what
    you sell stays sold.

    CLEAN UP YOUR ACT - Use time gained
    on less packing and shipping to improve
    inventory control and make your operation
    more efficient.

   FOCUS ON PERSONAL VALUE ADDED-
    Do what you do do best. Offer items for
    which you have unique personal expertise.

   LET YOUR INVENTORY DROP - To
   the point where you are nearly supply limited.

   OPTIMIZE STORAGE - Try to dramatically
   slash your total long term storage costs.

   DON'T BUY UNLESS - You see unusual
   opportunities that clearly can give you a
   30:1 sell buy ratio edge. Then grab them
   agressively.

A toned down version of Debby Does Dallas
is newly available as Doris Does Des Moines.

The price of oil has dropped 3:1 to its 2004
levels.

The price of copper has dropped 3:1 to its
2005 levels.

The latter totally demolishing our local economy.

I'm often asked why I no longer publish
electronic construction projects.

Two short form answers are that the
college cancelled the electronics program
because the football team needed the money.

And that, not only can nobody solder anymore,
ROHS has often made soldering illegal.

Very simply, there is near zero demand,
the risks are too high, and the rewards
waaay too low. I do have several construction
projects available; they have not remotely
begun to justify the time and energy that
went into them.

There is some electronic construction
interest remaining that is covered by
Circuit Cellar and Make magazines.

Plus the bizarre Steampunk movement.
And some Retro Vacuum Tube stuff.

Tentative plans are to do a BLM Brown Bag
lunch lecture on 150 Gila Day Hikes in late
January and a Discovery Park lecture on
Energy Fundamentals in early February.

Both are open to the public and free. You are
invited to intend. More details whenever.

Sometimes less is more. We bought into an
apparent disaster with a huge instrument that
was broken, outrageously heavy, and three
generations out of date.

You could not really call it a white elephant
since it was a perfectly normal gray elephant
size and color and weight.

Turns out there was a quarter ounce of diamond
tooling included that was readily sellable to
bail out all costs. And some nondescript accessories
related to toolholders easily and profitably sold that
worked just fine with the newer models.

About five percent of the unit was profitably sold.
The rest was simply scrapped.

The secret to home energy management is to
measure and evaluate your incoming power.

I've long proposed an ISOPOD solution that
snaps onto your incoming power line and wirelessly
reports your ongoing energy consumption.

Trouble is that I don't have these for sale. And
there are safety issues even when clamping an
insulated ball onto an insulated wire.

An intriguing possibility is to point a wireless
video camera at your power meter. This is totally
legal and easily done. Some special image recognizing
software would have to be added to extract the
actual energy use.

If there is one thing I cannot stand, it is
intolerance.

I'm not too sure what to do about my Incredible
Secret Money Machine. We sold out long ago,
but copies remain available on Amazon and
elsewhere. Some for outrageous prices.

The book did define an era of small scale home
entrepreneuship. But was pre computer and
pre web, so it is now laughingly out of date.

This was from the time when magazines threw
unlimited big bucks at just about any author.
And zillions of electronic hobbiests would buy
most anything in the way of a do-it-yourself kit.

And when owning access to a font was
a big deal. Fixed point size, of course. And
rarely could you afford more than two. And
when it was unthinkable for a degreed
engineer to frelance.

Chances are I'll make the ISMM freely
accessible on my website. Once I get
the Adobe CLEARSCAN working properly.

There is not yet a total solution to scanning old
books, but Adobe Acrobat 9 has apparently made
a giant step in the right direction.

What is ultimately needed is something that scans
book pages, analyzes to OCR style ASCII, and
then does a new layout that accurately approximates
the original font. Sans typos, glitches, and unrecognizables.
And with minimal file sizes. And character shape editable.

Acrobat 9's CLEARSCAN process does the scanning and
replacement, but leaves the unrecognizables as bitmaps
File sizes are smaller but nowhere near a pure text solution.
And the final fonts are improved but not optimal.

I can't seem to be able to get it to fix the unrecognizables
or let me tweak the font to improve its appearance. I
suspect an upgrade or two will finally give us a really
great way to capture pre-electronic books.

Two interesting and potentially useful
online book resources are http://openlibrary.org/
and http://www.librarything.com/.

The former attempts to create one web page
for each and every book ever printed. It then
gives you paid or free access per availability.

Librarything lets you post your own books to
find others of sinilar interest.

Science Magazine for October 24, 2008
vol 322 page 539-540 has an interesting
new topic that should make a great student
paper or be useful for fascinating future
research.

Titled Putting Electrowetting to Work.

Electrowetting is a nanotechnology effect
where an electric field can make a surface
wettable or nonwettable. As in hydrophobic
or hydrophillic. Important new uses are
liquid lenses for digital cameras and microsample
sized chemical analysis instruments.

Some other student paper ideas appear here.

It pays to cast as wide a net as you can
when searching for auction opportunities.

And not skipping the stuff that only rarely
pays off. Ferinstance, Craig's List recently
included a rare test equipment estate sale.
Which was findable on an "oscilloscope"
search. But which I missed by two days
because I had not checked for a week or so.

Some auction houses have outstanding
but very rare local auctions. Stopal in
particular has been a superb performer
for me. Especially with "contents of room"
and "contents of shelves" offers.

And second tier auctioneers with rare
offerings can end up outstanding sources.
Especially if they underpromote or are
simply clearing a building for a disgruntled
owner. Arizona Auctioneers has been most
useful to me for second tier offerings.

Auctioneers only "slightly" too far away can
also prove of occasional use. Such as
Dickerson in New Mexico, Mangold in
Wickenburg, Haynie in Snowflake, or
Harris in Prescott.

Much more in our Auction Help library page.

First impressions can be super important,
and the tiniest miscue can cause you grief.

Ferinstance, there is a local restaurant
that for years had (and still has) a super
sharp front door handle. Grab it and you
worry about getting cut.

Making the first contact experience
overwhelmingly negative.

Revised and improved our Arizona Auction
Resources library page. A guided tour is
available here and your own custom local
auction finder here.

Some sources for free online USGS topo maps
and similar include...

           http://mapper.acme.com/
           http://terraserver-usa.com/
           http://maps.google.com/
           http://www.digital-topo-maps.com/
           http://mapserver.maptech.com/homepage/index.cfm

An improved copy of my very first technical
paper has been uploaded here.

Since day one, my idea of the "norm" for a
technical paper has been 3000 words and five
figures. Less than this did not seem enough to
do the job, and more exceeded both reader
attention span and the outrageous costs of
editorial magazine space.

The web demands much shorter stories and it
can be a real challenge to try and say the same thing
in 300 words as once took 3000. See our
GuruGram #100 for an example that uses our
PostScript PowerPoint Emulator.

I guess a key secret to web communication is to try
and think like a cartoonist rather than a tech writer.

An impressive but possibly overstated breakthrough
in solar panel technology can be found here.

A new nanotechnology coating allows nearly all of
the incoming energy to stay inside of the substrate.
Apparently it works by successive bending of incoming
energy towards the norm.

A typical silicon panel might go from 67 percent
absorbtion or retention to 97 percent.

But note that this does NOT do anything for the
tracking angle. If the energy comes in at a 30
degree slope, it will still only have half the density.

All (!) that happens is that much less gets reflected
and much more absorbed.

An interesting map of nationwide gas prices
appears here.

As the price of gas went up, the amount bought
went down. Surprise. Surpricse. Or at least
it seemed to me where traffic into both
Phoenix and Tucson seemed to nosedive.

Meanwhile, the total trashing of the economy
probably prevents any increase in gas purchases
as the price goes down.

Karl Schmidt of Transtronics has created
a new web page with images of radial and axial
capacitors on them.

Once again, radial electrolytics are extremely
rare. But a lot of users seem to be confusing these
with the single ended axial style.

Apparently there is a dead-wrong Wiki Answers
post that is causing much of the problem.

Once again, radial capacitors have their leads
going out in a radial direction. Axial capacitors
have their leads going out along their axis.
They may go out the axis in one or two directions.

Per any 1929 electronics book.

Sometimes the simpliest "Duh" changes
in policy or the cheapest modifications can
make a dramatic difference in how things
get done or how much hassle they involve.

Always putting your reading glasses to the
right of the regular ones makes them easier
to find in the dark. A simple Velcro strap
prevents a cushion from forever jamming a
seatbelt latch. Extra hooks on the back of
a door or the side of a cabinet can significantly
reduce clutter.

Apparently the enormously useful Verify
Web Links in Document of Acrobat 7 has been
flushed in Acrobat 8 and Acrobat 9.

It seems to have been replaced by a more
versatile but klutzier Web Capture --> View
Web Links. This simply gives you a page by
page list of live links to check.

Considerably more effort seems to be needed
to get a list of all broken links in a document.
Hopefully the old code can be made available
as a plugin.

Any time you have a reasonably popular web
site, it is super important not to change its url
without keeping a redirect on anything earlier.

For doing so will cause grief for hundreds or
even thousands of other web sites that have
linked you.

Apparently Dilbert just changed syndicators
and failed to preserve their old links. Thus
trashing access from thousands of websites.
We have added the new link to our home page.

Btw, the Dilbert videos are available here.

The way you tell an extroverted engineer
is that they stare at your shoes rather than
their own.

Just made the amazing discovery that it is
not possible to buy a HDTV set at any price!

A useful HDTV would be in the 32 to 36 inch
size and have a minimum VGA resolution of
1280 pixels horizontally for web based tv
downloads. It would, of course have to include
a wireless keyboard and mouse with a minimum
of a 30 foot range to reach the receiving
computer.

Or have Firefox built in.

It would also have direct Netflix download
capability. The ability to receive on-the-air
broadcasts, of course, is totally useless.

Nobody, but nobody will sit still anymore
for a program that they cannot start when
they want, pause, fast forward, and rewind.

With all episodes instantly available.

Some day I will reveal my secret Fire
Lookout's recipe for boiled can.

The key secret is the 24 hour prep time
as the can should be boiled in yesterday's
dishwater.

We are STILL getting complaints from individuals
who cannot tell the difference between a radial
and an axial capacitor. How or why this continues
to happen is an utter mystery.

As any 1929 radio book will tell you, an axial
capacitor has its leads coming -- of all places --
out the AXIS of the capacitor. Two popular
styles of axial capacitor are the DOUBLE
ENDED axial where one lead comes out
each end, and the SINGLE ENDED axial
where a pair of leads comes out one end
for printed circuit thru-hole mounting.

A radial capacitor has its leads coming
out the radius of the cylinder. While common
in film capacitors used for thru-hole mounting,
radial electrolytic capacitors are quite rare.

We continue to check our listings for typos,
but this profound misinformation seems to
be coming from elsewhere.

Welcome to hell. Here is your accordian.

Another new source of free technical
and scholarly papers is http://www.arXiv.org.

They are now up to half a million papers
online.

It can be enormously difficult it is to properly
measure the power and efficiency of pulse waveforms.

It is trivially easy to grossly understate
the power and efficiency of these waveforms.

Any older meter is GUARANTEED to
read wrong, since it will be averaging
rather than RMS. The newest of
available lower cost meters are now
full RMS, but they still have strict
crest factor and bandwidth limits that
will still cause you grief.

While there are some new power meter
chips from Analog Devices and others,
you have to be very careful in how
and where you use these if you are
to get accurate and believable results.

A possible solution lies by using a pair
of identical high speed A/D converters
that are synchronously sampled. The
trick is to simultaneously measure the
voltage and current, go digital, and then
multiply the two together.

No analog solution can properly deal
with high crest factors.

Surprisingly high speeds are needed for
useful results. If you have a one millisecond
pulse you are measuring, a 100 microsecond
sample rate can only give you ten percent
accuracy at best. 10 microseconds can approach
one percent accuracy.

A 100 kHz, 14 bit A/D pair would be recommended
as a minimum. This could be build up from scratch
or should be available as part of commercial
lab data acquisition programs.

Few people seem to realize how earth-shatteringly
profound the IPTV format of distributing television
shows and videos is about to become.

You can sample these using the gray VIDEO DOWNLOADS
links on our home page. Or go to any of thousands of
other places on the web.

You now can start ANY tv program of ANY age anytime you
like, run it as long as you like, fast forward, rewind, pause and
hold, and otherwise be in total and absolute control of what you
are watching. All fast approaching HDTV quality and ( at
least temporarily ) with tolerably short and reasonable
commercials.

And all free.

What possible point could there be remaining for broadcast
network tv? For local theaters? For Blockbuster and other
video rental firms? For DVD's? Or (sadly) for Netflix?

And guess what is going to happen to web bandwidth
demands as the average web page shifts from a few
hundred text bytes to gigabytes of full HDTV?

>Clubs in general seem to be on the wane.

Memberships in some are sharply down,
while the average age in others is
dramatically increasing.

Obvious examples are ham radio now
being a ludicrous geriatric parody of its
one time greatness. Or at least some
caving clubs now having many older members
than newer. Or organizations like the AAUW
struggling for members in smaller rural
communities. Or the dramatic drop in the
number of computer clubs. Or in activities
such as model railroading.

The only clubs that appear to be bucking
the trend are the busywork ones in retirement
communities.

The web is somehow the blame, with
eyeball siphoning to immediate virtual activities
combined with instant access to information
whose value itself has dramatically declined.

Adobe Acrobat 9 has made dramatic advances
in the old problem of converting scanned text into
searchible and compact stroke defined fonts.

Trouble is that either it is not working or that
I cannot get it to work.

What is supposted to happen with their new
CLEARSCAN process is that a new font gets
created that looks pretty much like the original.
Meaning that any missed or misinterpreted
characters will still look about the same.

Sure enough, you get serchible text and a
stroke defined font. But they refuse to identify
the misidentified words. And there seems to be
no way to "improve" their created font for
bettter legibility and less rattiness.

Ferinstance, any character breaks could be
filled in on a substitute font for greatly improved
legibility. But they will not let you edit or substitute
the new font they created! There also seems to
be several possible letter shapes for a given
recognized character. Yet the text search seems
to work.

I suspect part of this is a bug that they will
repair in a future release. Meanwhile, they
are only part way solved on the "scan and
improve" old text problem

So far, I have not gotten very far with print to
disk as a sledgehammer cure.

Bad times create a dilemma: Do you tightly focus
on your own personal value added, aggressively
minimizing inventory and ruthlessly getting rid of
anything long unattended? Putting your energy where
it will do the most good?

Or do you make lowball bids on everything in sight
at a penny on the dollar or less, scooping up incredible
( but possibly utterly unmarketable ) bargains? Spreading
yourself thin on stuff you are no good at?

Several alarming signs that all is not well:
UPS locally has sharply cut back on employees
and employee hours. Because nobody is
shipping anything.

Many local restaurants have also cut back
hours, and it is now difficult to find Sunday food.

Several local auctions were cancelled mid show
( legally or otherwise ) owing to low bids and
lack of interest.

Local construction has been cut waaay back
with many projects literally stalled in their tracks.
New houses are selling at $120,000 discounts.

Our latest Craig's List property ad relisting only
pulled a third of its linkthroughs during its first
few days.

There is no reason to expect eBay to be the
exception ( per "pawn shops do good during
bad times" ). So far, our own sales have
slightly increased, but I do not expect this
to continue.

A millstone has been passed with the
1000th download on Wesrch .

Critical mass for Wesrch remains at
50,000 papers stashed. But there is
hope, considering its build rate.

Sites such as this one are the way that
research papers are gonna be presented
in the future. Working around the death
wish of all the major scholarly pubs.

One more time: ALL papers over three
years old should be instantly and freely
available without charge or restriction.

Here is a more or less independent verification
of our Magic Sinewave spectra done using
Sigview Math rather than our own.

We start with this data file, which is an ASCII
representation of fifteen cycles of an 0.53
amplitude original sequence n=8 magic sinewave.

Enter this file into Sigview by selecting import raw
binary, 8 bit unsigned, and 2,499,840 samples per
second. A display of fifteen square waves should
appear. Magic sinewave details can be temporarily
viewed by zooming by 65536 cycles. Be sure to
unzoom before continuing.

Set the Sigview defaults to subtract mean and remove
linear trend. Use a Blackman window and smooth
to Blackman 5. Temporarily select a linear display.
Do a FFT and zoom to 1024. A Magic sinewave
should appear with its fundamental, and zero even
or odd harmonics to the 30th. The 31st and 33rd
harmonics should be fairly strong.

While the linear display is impressive, the log display
is far more useful. Switching to it and rerunning the
FFT should show all of the intermediate odd harmonics
at least 65 decibels below the fundamental.

This is an original magic sinewave, so you can
zero out two more harmonics simply by going to
a best efficiency type instead. The data has been
quantized to allow a 10 MHz PIC on 60 Hertz.

A "shake the box" jittering of the quantized
data should reduce the unwanted low harmonics
by another two to ten decibels. Further reduction
can be done by increasing the samples per cycle
and the PIC clock rate.

I'll try to work up a tour and more details as
as upcoming GuruGram.

I'm not all that convinced that the all electric
car enthusasiasts have really thought things
out very well...

    FAST CHARGING - is well beyond any
    home installation and most utility distribution
    capabilities. Without staggering new costs.

   BATTERY EXCHANGE SERVICES - would
   triple or quadruple the de-facto vehicle
   battery costs.

   CERTAIN URBAN AREAS - ( Phoenix in
   particular ) have their entire economy based
   on vehicles with at least a 150 mile range.

  AIR CONDITIONING OR HEATERS - are
  absolutely essential in many areas of the
  country.

  ON BOARD PV PANELS - make absolutely
  zero sense. They might add enough energy
  for the power mirrors provided you did not
  adjust them very often. IF the air conditioning
  losses of sitting in the sun did not offset any
  possible gain at all.

  EXERGY CONSIDERATIONS - tell us you
  should start with anything but ultra high value
  grid, wind, or pv electricity. Destruction of
  energy quality during charging and maint
  and amortization is inherent.

  MANY MOUNTAIN WEST AREAS - could
  create a catastrophy on a remote dead battery.

  NIGHT DRIVING - considerations are not
  trivial. Some headlights consume 200 watts
  or more.

More energy fundamentals here and here.

Sometimes starting with the answer and working your
way backwards can dramatically simplify a problem.

This week's NUMB3RS math puzzler asks you to take
a 7 liter bottle, an 11 liter bottle, and all the water you want
to deliver 2 liters of water.

Working backwards, the 2 liters has to come from 6 liters
stored. The 6 liters has to come from 10 liters stored.
The 10 liters has to come from 3 liters stored. Bingo.

Not sure what their parallegram thingy is all about.
It might be equally cute.

Actually, two liters is a fairly easy solution. 2 begats 9.
9 begats 5. 5 begats 1. 1 begats 8. 3 and 4 are trivial.

Just did an upgrade to Acrobat 9 Pro. There were a
few surprises. The cheapest upgrades are via
educational software, providing you can submit a pay
stub or a student id.

A pleasant surprise is that many hundreds of top quality
fonts are now included. But that some older fonts may
vanish without a trace, per the details we looked at here.

The big recent gotcha, of course, is that the ability to read
and write most host files has been locked out as a default.

To use our Gonzo Utilities or otherwise write PostScript
code that can read or write any host file under Windows XP,
run distiller from the command line using

                             Acrodist -F

Then drag or drop your PostScript code into any open Distiller
window on the desktop.

 

There do not seem to be any new surprises with the Distiller
Params or PDFMark operators. By clicking on the help
box in disiller, you can access the current Distiller Params
or PDFMark operators.

But here is how to view them on your own...

          currentdistillerparams
          {exch == == (\n\n) print flush} forall

         1183615869 internaldict begin
         pdfmarkInternalDict { {== ==} forall } superexec

And a review of how to do transparency in raw PostScript
is found here.

Shortly after World War II, our sun became a radio
star with the start of commercial tv broadcasting.

Captain Video, Roller Derby, and Kukla, Fran & Ollie
became our goodwill ambassadors to outer space.

These signals have now gone out some sixty years, yet
remain at levels more or less detectable with our current
state of the art technology. Now approaching a full MILLION or
so CUBIC LIGHT YEARS of space. In which there are
likely to be at least a few candidate intelligent life systems.

Assume you are a junior level intern on Alderon VII monitoring
surprisingly structured vhf changes in a new minor radio star in
an obscure second rate arm of the Milky Way galaxy. As tends
to happen with astronomical observations, very rarely exceptional
viewing conditions may briefly result.

And you receive a lucidly clear ten second video clip of -- Roller Derby --
as the sum total of the knowledge of our civilization.

What report, if any, would you file with your supervisor?

How much would you say civilization has progressed meanwhile?

It is very easy to have your favorite fonts disappear
forever on any rebuild or update to Adobe Acrobat.

Some of the causes include...

   Dis-Installing an earlier version, especially
   on an Acrobat 9 from Acrobat 8 upgrade.

   Having your fonts on a folder whose access
   is blocked by security features on 8.1 or
   higher.

   Having a font folder within a folder.
  
   Having more than one user on the computer.

First, note that Acrobat Distiller on Acrobat
versions above 8.1 will only read or write
certain computer files. This was done as
a security precaution. If you want to read or
write from any file under Windows XP,
run distiller from the command line using

                      Acrodist -F

Then drag or drop your PostScript code into
any open Distiller window on the desktop.

The same restriction apparently prevents
you from defining certain files as font sources.

The safest place to put your favorite fonts
appears to be C:\Program Files\Common Files\
Adobe\Fonts.

Be sure your actual font files are "loose" in this
folder and not buried further in any subfolders.

Also, font locations may not get picked up until
after the next XP cold reboot. So always restart.

And ALWAYS keep CD or flash backup copies
of all of your favorite fonts.

A reminder that we have a new library page
on Gila Valley Day Hikes.

Most of these are an hour's drive or less from
the Greater Bonita-Eden-Sanchez metropolitan
area.

Just got yet another email from an individual
who was going to revolutionize wind power once
they got a patent on their possibly overunity and
paranoid secrecy concept.  All they needed was
millions of dollars for development.

They had no simulations or computer models of
course. Nor any small scale feasibility demos.

My guess here is that we have yet another fatally
flawed attempt at a nonsolution to a nonproblem.

It should go without saying that an industry outsider
will not have a snowballs chance in hell of getting their
concept accepted.

First and foremost, because their idea is almost certainly
totally worthless. But secondarily just BECAUSE thay
are in fact an outsider.

Patents, of course, are utterly and totally useless
for individuals and small scale startups. In that
they are GUARANTEED to result in a net loss
of time, energy, money, and sanity. Per these
details.

IF your concept has value, the first thing you
should do is have it independently reviewed
by industry insiders. Chances are that there
are fundamental gotchas that have caused
your idea to have been dismissed long ago.

We do offer concept evaluation services,
but we will not do so on a secrecy or NDA
basis. Because secrecy GUARANTEES
worthlessness.

Still getting a bunch of eBay questions on whether
pushbuttons are momentary or not. Virtually all
ordinary pushbuttons are momentary contact, meaning
that they are active when you push on them and return
to their initial state on release.

A class of special pushbuttons called "alternate action
pushbuttons" is available that work in a "push on- push
off" manner, sort of like a ball point pen. But these are
rare and expensive. Especially in smaller sizes.

It is however, quite easy to convert an ordinary pushbutton
into an alternate action one simply by adding two CMOS
inverters, two resistors, and a capacitor. Per the details found
in figure 56.1 of this link .

Many similar circuits in my CMOS Cookbook.

Free video or not, I continue to be utterly appaled
at the quality of traditional network tv offerings.

The demise of "Las Vegas" has cut the number of
watchable network tv programs by 50 percent, since
there is now only one left.

I tried watching the pilot for "Fringe". This one gives
new depth of meaning to "mesmerizingly awful". Now,
I am much in favor of hiring the handicapped and having
remedial video special ed classes. But I cannot believe
they did this one on purpose Surely they did not pay for it.

Mewanwhile, we now have " The Mentalist". Which stole
the concept, plot, and subplot from "Monk". While, of
course, totally trashing the humor and the watchability.

And the latest internationally popular video game show
seems to be based entirely on people smashing their
heads against a wall. No, I am not making this up.

But I can hardly wait for the next episode of CSI
Gila Bend.

Expanded our "free video links" download
section on our home page. The number of sites
offering extensive video downloads is now expanding
faster than we can add them.

Which should have dire effects on video rental
stores and theaters who have not upgraded to
full digital.

Please email me with your new favorites.

Just got yet another email from a clueless
epsilon minus about to get sucked in on a
"Browns Gas" vehicle electrolysizer.

One more time kiddies: Without exception,
these are a DYNAMIC BRAKE that will cause
the vehicle to come to a stop in less than its
normal coasting distance. They are EXACTLY
the same as driving around with your emergency
brake partially set.

Every 200 watts at the alternator costs you two
miles per gallon in reduced fuel economy. No
conceivable gain of an electrolysizer can ever
hope to overcome such a staggering loss.

Very simply, before the usual monumentally stupid
mistakes are made, the numbers do not add up by
a country mile.

After the usual monumentally stupid mistakes are
made ( such as using a fanbelt or stainless steel ),
the numbers become ludicrously and laughingly absurd.

A more detailed analysis here.

Usually credited to Mark Twain but really from
Josh Billings, an often better humorist:

"I've never known an auctioneer to lie. Unless
it was absolutely convenient."

There always seems to be some confusion between
Wordpad and Notepad. Both are simple text editors and
both are nearly perfect for programming in raw
PostScript using my Gonzo utilities.

Wordpad did let you have more than one font per
raw document. Notepad was something of a stepchild
in that it had a serious 64K total size limit. This limit
has long been removed and Notepad now works just fine
for most any general purpose editing.

A reminder I'll be doing a Mount Graham Tramway
lecture at Discovery Park this Saturday September 20th
at 6:30 PM.

Rebuilding a computer system from the ground up can
be a monumental pain, but it is sometimes the only
solution for extreme registry, stability, and blowup
problems.

BE ABSOLUTELY SURE YOU SAVE ALL OF
YOUR WORK FILES TO CD or Flash drive BEFORE
ATTEMPTING ANY REBUILD!

And don't forget the nonobvious hidden stuff. Such
as your sigfile from your email program, your address
book, and all of your Adobe fonts. The easiest way to
capture the later is to get into Disiller, then Settings,
then view Font Locations. This should give you a list
of what you want to keep.

I've just gone through a rebuild and thought I'd
review some of the maddeningly infuriating details
as well as go over what software I personally use why.

Much of my work is based on using raw PostScript
and my Gonzo Utilities along with custom written
and hand crafted code. Works for me.

Just reinstalling XP did not work for several reasons,
and my only XP upgrade refused to recognize HP Pavillion
recovery disks! So I started with a fresh OEM version of
Windows XP. Vista, of course, is only for Edsel owners.

It is very easy to trash the serial number on the OEM
software, so be sure to reseal everything when you are
finished.

The first order of business on a rebuild is replacing
their hideous screens with the blue "none". Useful and
crucial shortcuts to XP iteslf should next be added. I
use Paint extensively, so this was my first addition.
Followed by Wordpad and Notepad. Shortcuts to C: and
D: are easiest added by starting with a shortcut to something
else and then editing the destination.

Naturally, you IMMEDIATELY open up all hidden files so
you can view them. I overwhelmingly prefer a "bare metal"
approach to computer programming.

A Control Panel shortcut requires some sneaky tricks: You
have to go to the Folder Options and chose the hidden
"Show Contol Planel in My Computer". The System shortcut
is easily done by creating it out of the Control Panel.

Once you have all of your needed shortcuts in place, you
can then replace all of your original workfiles and folders.
It might pay to create a new "My Fonts" in Program Files
to simplify futuer font management.

Next in line is your networking and temporarily getting
Internet Explorer working. Needless to say, this is an
emergency standby procedure that you will flush at your
earliest possible oppotrunity.

We use a HP OfficeJet Pro K550 as our network printer.
Which has a maddeningly infuriating software feature: They
insist on force adding a "Photo Networking File" to your desktop.

CLICKING ON THIS FILE WILL BLOW YOUR SYSTEM UP!

The file serves no useful purpose whatsoever and falls in the same
class as infuriating "AOL Anything" bloatware.  And might have
been one of the main reasons I had to rebuild in the first place.

Removal is rather obtuse as you have to  run the Networking File
wizard to remove itself. This file is far beyond despicable.

Continuing our "secrets of an XP rebuild" thread...

Once your hidden secret XP shortcuts and your own workfiles
are all in place, and once your basic networking and printing
is set up, you can start getting serious over adding
software. Naturally, you start with Firefox and Thunderbird.

Be sure to go through your preferences on these programs.
such as turning URL underlining off. This feature is buried
under "color" for some reason. Whle underlining can be
tolerated in the third grade, it should be strongly discouraged in
the fourth and severely punished in the fifth Fortunately,
compulsive underliners eventually grow so much hair on the
palms of their hands that they can no longer type.

On Thunderbird, is it crucually important to click the "Leave meassages
on Server" box if several machines are to access your email
independently.

Additional details are reinstalling your sig file and reconnecting
your newsgroup reader. My reader preference is News.Individual.Net
which is convenient, friendly and cheap.

Naturally, you ALWAYS bottom post on any email or newsgroup response.

Next in order of buildability are some of the lesser software programs.
A FTP program is a must for website maintenence. and Ipswitch is clearly
top dog. I prefer to still use their ancient WS-FTP95LE which you can
get a free copy of here. I tried Filezilla and promptly flushed it.

I really like QuickPhrase and it is one of the few programs
I will gladly pay for. This lets you enter any text you use often with
a very few keystrokes. It is essetential for eBay seller listings.
Next, a PIC Programmer from Transtronics gets added, along with the
PIC Simulator from Oshensoft and a Fourier program from Sigview.

All of which at long last gets us around to Adobe Acrobat. There
free reader is largely useless, so I recommend the full Acrobat 9.
try to scam an educational version if you can. To do this. you will
need a fax of a paystub or a copy of a student registration.

You'll want separate shortcuts to Acrobat and Distiller.

I'm still looking for a decent website editor. GoLive is obsolete
and has serious size restrictions. Dreamweaver is an utter atrocity.

I am currently using Dreamweaver. You can partially get around their
infuriating slow response time (which was solved on Applewriter on my IIe!) by
working at the end of your file and then recopying to the beginning.

OK, kiddies, here's the secret to fixing the Dreamweaver code: You use
TWO files, one for everything  above the cursor and one for everything
below. With several cubic acres of free RAM between the two. Free form
entry does not require any remapping. Only when time critical data entry
is complete do you do any actual remapping.

I still do not known how to deal with their random throwing away
of half your attempted website links. But a cut-and-paste
and reedit is usually a lot faster and more convenient than
believing they are going to give you a link when they decide not to.

Continuing our "secrets of an XP rebuild" thread....

Two final details are getting your scanner and your digital camera to work,
HP seems really good about online software downloads, and you should be
able to install their latest drivers with no particular problems. It is probably
better to go with new software rather than your original, especially if your
scanner if obsolete or several years old.

Nikon returned a "Service Pack Incompatibility" error when I tried to reinstall
our original software on my Coolpix 5000. . Their website is somewhat
confusing, but if you try to ask them a question, they usually can second guess
what you want and immediately give you the link to a suitable driver.

EXCEPT - the driver STILL returns the server pack error. But it still lets you
install it and it seems to work just fine. Microsoft posts all sorts of dire
warnings about "does not meet XP certification". But I cannot imagine any
long term Nikon-Microsoft compatility issues of image format reading having
not been resolved long ago.

While I despise Microsoft code, I've found it necessary to add Word to do my
spell checking for me ( I find the program otherwise totally useless ), and
the eXcell spreadsheet for certain aerospace auction listings. I found Power
Point so distasteful that I wrote my own emulator instead.

While others place great import on antivirus software, malware detectors,
and registry fixers, I've often found these to introduce more problems than
they solve. My own approach here is to only install these rarely, fix
whatever specific problem I have to, and then deinstall them.

I do most of my word processing and electronic simulations in raw PostScript
using my Gonzo Utilities. Image postprep and such is usually handled with custom
routines I've written such as these.

Given the choice, I'll usually write my own tools to solve any problem. Some typical
examples are found here. In those rare instances where PostScript does not have
enough math acuracy, I'll extend it a little with my custom routines, or else go to the
Javascript fround in Firefox.

Wikipedia has come up with a fairly useful Fuel Economy in Automobiles
thread. One of their useful and otherwise hard to find figures is that
200 watts at the alternator translates to  nearly two miles per gallon drop
in fuel economy.

And that for a wide variety of vehicles, there is not much difference
in miles per gallon between travel at 20 miles per hour and 60
miles per hour. Apparently the higher fuel consumption gets
offset by the shorter trip time.

Ordinary high beam headlights consume 200 watts. Thus, running
with your lights on costs you two miles per gallon. And any
electrolysis fantasies will cost you two miles per gallon for
every two hundred watts of alternator input.

There is, of couse, no way in hell that any on board electroysis benefits
can remotely begin to dream of approaching these staggering loses.

There can be a 5:1 or greater difference
between the price of test equipment on
eBay. How can you guarantee your offers
always are near the high end?

    DOES IT WORK - There's no way to
    fully test an item unless you are a cal
    lab, but you should run enough tests
    to be able to say "appears fully functional"

   IS IT COMPLETE - Is the case intact (or
   at least superglued? Are all knobs present?
   Can the customer plug the unit in and have it
   do something?

   IS IT SQUEAKY CLEAN? Have you super
   detailed all of the knobs and removed all
   third party labels?

   FIFTEEN DAY INSPECTION - Always give
   the customer an opportunity to evaluate. And
   be willing to offer a refund at any time for
   any reason.

  EXCEPTIONAL PRODUCT PHOTOS - Always
  spend a minimum of two hours in image post proc.
  Architect's perspective, panel relettering, single
  pixel locking are all a must. Tools here.

  LIVE SCREEN SHOTS - At least one image
  should show the instrument alive and doing something.

  SPECS AND USER MANUALS- The buyer should
  know where to go to get free or cheap docs to use
  his instrument. Along with manufacturer's site links.

  IS YOUR LISTING LUCID AND NON-SNOTTY?
  Do you have a rep for delivering quality products?

  IS YOUR PACKAGING BULLETPROOF? A
  double boxing is recommended for most items.

  DO YOUR LISTINGS START HIGH? If
  a piece of test equipment sells on the first
  three tries, you did not charge enough for it.

  DO YOU FLUSH THE DREGS? If you do not
  feel the item is first rate, put it on the Alvin Pile
  rather than listing it?

Does it ever pay to flip someone else's eBay
item? The double shipping charges guarantee
you will be unable to remotely approach a 30:1
SBR. And the risk factors are off the wall because
the other seller may be flushing garbage rather
than making a stupid listing mistake.

Nonetheless, if you have very special expertise in
one area, there rarely might be times when flipping
sort of works. I happened to have a case left over
for a toasted piece of test equipment that we sell lots of,
and a flip just may be working on a broken case offer.

I also just saw an oddball "reverse flip" as well.
There is a horribly obsolete collection of machine
shop tooling that nobody uses any more. These items
remain outrageously expensive, and they are listed as
such on eBay. It is difficult to imaging anyone buying
one ( you need an entire collection ) at any price for any
reason.

An associate of mine picked up a pile of these free
and listed them at a ridiculously low price. One of the
other eBay hopefuls bought them to keep them off
the market.

Scene lighting at night has always been a serious
problem for fire departments and EMT services.

Thatcher Fire Department seems to be having very
good success with a new inflatable scene light that
covers up to 12,000 square yards. It is self standing
and self-erecting in absence of wind. And can be
guyed under reasonable wind conditions.

It is normally self-powered by a Honda Generator.

Prices range from $2000 to $4000.

Not surprisingly, it is locally referred to as the
"dildo light." I can see a fairly rapid turnover in their
pr department.

There seems to be yet another "more miles per
gallon" video making the rounds. As usual, the
most likely explanation is incompetent labwork.

If you are a legitimate researcher, any time you
get a "too good to be true" result, you should
drop what you are doing and immediately set
about to PROVE YOURSELF WRONG.

And then IMMEDIATELY ask others to do
the same.

It is enormously difficult to properly measure
miles per gallon outside of a major proving
ground. The odds are overwhelming that
your measurements are flat out wrong.

The other thing that these hubris bound
proponents never seem to pick up on is
that this is thoroughly plowed ground. Unless
you are bringing something genuinely new to
the table, the odds of an individual significantly
improving vehicle miles per gallon is zero.

I had Paint blow up again on an older machine, so
I decided to check out the alternatives. Bottom line:
I was not impressed.

The best of the also rans seems to be Ultimate
Paint. It offers great heaping bunches of ultra fancy
features and its learning curve is reasonably steep.

But it seems to have thrown the baby out and
drunken the washwater. Very simple tasks such
as single pixel work or fast symmetry or transparent
replacement seem cumbersome and awkward.

I suspect what happened is that they studiously
avoided the Paint algorithms because of IP
issues.

I very much feel that classic Paint is
grossly underappreciated. It continues
as my first choic in Graphics Programs.
Helped along by Imageview32 and my own
custom routines.

One of the other blogs asked what a "spandrel"
was. A Crocker Spandrel is a medium sized
hunting dog.

I may sorta have got some of this right,
so here goes: I've been wanting to update
our Regular Magic Sinewave calculator
but have been mystified with how to deal
with the two possible solution sets.

Simply automating the old "level II" glacial
calculator set probably makes the most sense.

We can start by noting that Newton's Method
lets you get a closer answer by adjusting your
present answer by the improvement times the
slope. In the case of a cosine, cos (a+x) gets
adjusted by -xsin(a).

We also have a trig identity of cos (a+x)
simplifying to cos(a) - xsin(a). Which is identical,
but we do have to remember that our initial x
has to be in radians. Or degrees * pi/180.

Our goal is to zero out our "pseudodistortion",
which is the actual residue at harmonics 3,5,7...
combined with our fundamental residue error. This
pseudodistortion is calculated in the usual square-
root-of-the-sums -of the squares manner.

It turns out the harmonic equations are not in
any manner independent. A fundamental change
in residue of "w" will generate a 3h change
scaled by sin(3a)/sin(a) and so on. Which
creates a big ugly quadratic polynomonial
that depends on "n".

The function apparently has a minimum which
we can easily calculate by setting its derivative
to zero. Thus, we should directly be able
to make a fairly simple calculation for a
best guess improvement. Instead of the horrible
iterations done back in the level II calculators.

The optimum appears to be of form...

    w = [k1 + k3s3 + k5s5 + ... ] /
          [ 1 + s3s3 + s5s5 + ... ]

where kx is the existing residue and sx
is the incremental change.

More as I've had a chance to further test.

As we've seen here, you should never make any
foreign eBay sales. Except to New Mexico.

Even going beyond the language barrier and the hassles at
customs, everything needs reloaded at the border
stations because the truck tires are a different size and
spacing.

Fortunately, reversible truck tires have become newly
available. These can simply be insided out at the border
station and the truckers can continue on their way.

Yet a third gotcha in low low temperature delta energy
recovery is that the amortization clock runs all the time.

If your solar heat is only there six hours of the day,
the cost of financing the project will be at least FOUR
TIMES higher than a continuous energy source.

A useful amortization calculator can be found here.
More on energy fundamentals here and here.

Sort of the latest GuruGram #96 is on Enhancing your
eBay Strategic and Tactical Skills VII .

Sourcecode is available here.
And additional GuruGrams here.

It can be interesting to try and second guess
some of the fundamental engineering numbers
behind the Mount Graham Tramway. They
apparently decided that one mile of 1-1/2 inch
track cable was the largest feasible they wanted
to mess with in remote and difficult roadless
vertical terrain.

This implied a minimum of fourteen (!) tightly
linked and carefully integrated tramways working
together in the total project. Such cables would
likely be 100 ton test and weigh four pounds per
foot or 22,000 pounds per mile.

The cable guides on the towers might have a
coefficient of friction of 0.1, meaning that 2200
pounds maximum ( or around 1100 pounds average )
might be needed to yank the cables in place. Such a
force is easily done with a windlass or other fairly simple
mechanical advantage.

Cable sag is obviously crucial with a maximum
effective tower height of only fifteen feet or so.
Probably set by project cost and what a mule
could drag in the way of timbers.

With an average tower spacing of 300 feet,
something like 1200 pounds of cable would
have to be supported. A single load might weigh
1800 pounds with a pair of 240 pound carriers
and its timber or sawn board load. Thus, our
sag would almost triple when a load was mid towers.

Cable tension was provided by tension stations that
placed enormous concrete weights on one end of
each track cable. Rock has a density of just under
three, so something like 200 pounds per cubic foot
of concrete + rock load.

Taking 7 x 7 x 4 feet as a typical counterweight
gives us just under twenty tons. Thus stressing the
cable to a sensible one-fifth its rating under normal use .

OK - Here is how you read the demo and proof
of our stupendous new digital sinewave synthesizer.
The code is ridiculously simpler than it looks at
first glance...

There are two elements to doing a DFT direct
Fourier Transform. To find the fundamental, take
the difference between the cosines of the start and
end of each pulse and add them all up . Then adjust
your answer by 4/pi for one quadrant or 1/pi for a
full cycle.

To find a harmonic n, find the difference between
n times the cosine of each start and end angle and
add them up. Then adjust your answer by 4/n*pi for
one quadrant or 1/n*pi for a full cycle. Then normalize
your harmonic by dividing by the fundamental.

The harmonics are presented as an array of
[fundamental h2 h3 h4.... hn] in the output PostScript
log you get by sending the file to Distiller.

Note that the first thirty harmoics are zero, the next
thirty harmonics are neglible, and the first uncontrolled
harmonics of concern are the 61st, 63rd, 121st, and 123rd.

Thus completely blowing away all known digital
sinewave synthesizers and doing so at the absolute
maximum possible efficiency.

As to the rest of the PostScript, you can ignore the
first half of the program as this is just a pretty
printer. There are two PostScript store definition
arrays that define the on-off pulse degree angles for
/regmagsin8.6 and /bbemagsin8.6. As in "Regular
Magic Sinewave, 8 pulses per quadrant, amplitude
0.6

There are four definitions for /quad1 thru /quad4
that let you mix and match which type of Magic
Sinewave gets used where. And finally, a pair
of definitions for fullfun and fullharm that finds
the fundamentals and harmonics for you.

I managed to "backfill" most of the missing blog
entries over the last six weeks or so. You may want
to go back to mid-July and pick up the latest
additions and updates.

A helpline caller asked whether there was some
way to use excess heat from his solar hot water
system to generate electricity.

In general, a double whammy prevents much
commercial happening in the way of low delta-T
electrcial generators.

The first is that there is a fundamental limit
to heat engines called Carnot's Law that
severely limits the theoretical maximum
possible efficiency ( never acheived ) to the
ratio of absolute temperatures.

Ferinstance, a source of 200 degrees F and
a sink of 100 degrees F would have a maximum
possible efficiency of 1 - (Tcold/Thot) or

      1 - ( 473+100 )/( 473+200 ) = 0.15 = 15%.

You do have to remember to use absolute
temperatures, which really trash the results.

The second problem is that there will always
be thermal impedances between your input
temperature and the hot side of your heat
engine and between the exhaust termperature
of your heat engine and ambient. These can
further trash your efficiency in low delta-T
situations.

Ferinstance, on a Peltier thermoelectric
cooler, it is trivially easy to get the temperature
rise of your heatsink to EXCEED your net
cooling drop. For a negative efficiency.

It is very hard to get thermal impedances below
half a degree per watt without going to forced
air or liquid cooling.

There are some interesting improvements in
thermoelectrics coming down, but it is ulikely
these can get around the fundamental low
delta-T limits for routine use.

Our latest and best Magic Sinewave Calculator
handles Best Efficiency, Bridged Best Efficiency,
and Delta Friendly types. Amazingly fast.

At present, this ancient and utterly glacial
caculator is needed for Regular magic sinewaves.
Which have suddenly become of extreme interest.

I've been working on an upgrade but this gets tricky
in a hurry. It turns out there are two solution sets
for regular magic sinewaves (based on how far from
the left the first pulse starts), and it is not clear how
to deal with this possible redundancy and ambiguity.

Stay tuned as the drama unfolds.

I still cannot claim that my "too good to be true"
double magic sinewave scheme is valid, but the
method does seem to be withstanding some
challenges.

A new version of my purported proof appears here.
It has been modified to show that the even harmonics
are all zero.

The claim is as follows: These minimal switch flips
off a dc supply will syntheize a clean power sinewave
whose first thirty harmonics are zero, whose next
thirty harmonics are negligible, and whose first
uncontrolled harmonics of concern are the 61st, 63rd,
121st, and 123rd.

While these unwanted harmonics are far enough
"out there" to be eliminated by simple low pass
filtering, they could also be dealt with by alternate
schemes such as resonant trapping, combing, or
bucking.

The implications for power sinewaves are in the
freeze dried holy grail in a spray can league.
With potentially billions of dollars of energy
efficiency and power quality at stake.

I once again invite you to try and prove this
wrong. Please note that direct Fourier Transforms
(e.g. plain old trig) are preferred to FFT because of
windowing issues and other subtlety. And please
note that successive quadradrants have to be
properly mirrored or inverted.

Let's hear from you.

While it is utterly amazing that the Mount Graham
Aerial Lumber Tramway was even built, the fact that
it only ran for a year strongly suggests it was an
economic failure.

The evidence on the ground suggests to me that
the down time was flat out excessive. Caused by
too small footers, using rock where concrete should
have been used, using sliders where real bearings
would have been more appropriate, using wood for
iron and iron for steel, and using "can do" local
attitude for genuine engineering.

Nonetheless, the story remains mind boggling. I'll
be presenting a paper on this at Discovery Park
on September 20th.

An obvious mod to yesterday's doubly magic
sinewaves that looks better and has two less
switch flips per cycle: Do the first half cycle with
an ordinary Magic Sinewave and the second half
cycle with a Bridged Best Efficiency Magic Sinewave.

At 0.6 amplitude, the first thirty harmonics will be
zero, the next thirty will be "very low", typically in
the 0.001 range, and the first problem harmonic will
be the 61st at 0.2 of the fundamental.

More details here and on our Magic Sinewave
library page.

Our demo proof is also updated.

Despite its "too good to be true" aspects, I am just
about to claim a major breakthrough in Doubly Magic
Sinewaves. In which unwanted harmonic energy ends
up astonishingly low and low pass filtering can be
dramatically reduced or possibly eliminated entirely!

A proof of sorts appears here. I invite you to find fault
with this analysis or flat out prove me wrong. What we
have here is clearly in the "freeze dried holy grail in a
spray can" class stuff.  

Consider a regular 8 pulse per quadrant Magic Sinewave
you can find from this older calculator that would give you
an amplitude of 0.6 and would zero out its first thirty low
harmonics. Its pulse positions would be...

            5.65029593062 on     6.29225152930   off
          15.81862798201 on   17.714530427221 off
          26.26333600962 on   29.361333130312 off
          36.84108032643 on   41.047780153511 off
          47.56589434941 on   52.747568100103 off
          58.46406338431 on   64.443154605331 off
          69.55560700493 on   76.107018445052 off
          80.84316472644 on   87.695888763311 off

Sadly, if used normally, its first uncontrolled
harmonic would be the 31st at a heavy 0.70119349921
of the fundamental. This would demand low pass
filtering in most apps.

Next, consider a Bridged Best Efficiency 8 pulse
0.6 amplitude Magic Sinewave using this newer
calculator. It would also give you 0.6 amplitude and
zero out the first thirty low harmonics. Using these
pulse positions...

            10.4353438671 on     11.7077750493 off
            20.9101395432 on     23.4181541929 off
            31.4636502516 on     35.1329284465 off
            42.1342741421 on     46.8514916513 off
            52.9576520410 on     58.5675969212 off
            63.9624867662 on     70.2640240671 off
            75.1632563302 on      81.9072380861 off
            86.5515348716 on      90.0000000000 off

This time, if used normally, its first uncontrolled
harmonic would be the 31st at a heavy -0.71971577316
of the fundamental. This would again demand low pass
filtering in most apps.

But note that this its 31 harmonic is almost exactly the
opposite of that of our regular Magic Sinewave!
The pulses magically "interleave" so that the uncontrolled
harmonics are used in an almost opposite fashion!

So, we use a regular magic sinewave for our first
quadrant and then a bridged best efficiency for our
second. Repeating the process for the next two
quadrants. The net result is apparently that harmonics
h31 throguh h34 "cancel out" to astonishingly low values!

This immediately sounds bogus, but the definition of
Fourier Series is integration over precisely one
complete fundamental cycle. And should be no different
than, say, the seventh harmonic piling up in one area and
unpiling in another.

What is really cute is that, after substitution by quadrants,
the first significant uncontrolled harmonic is the
SIXTY FIRST (!) down at ONE FIFTH the fundamental!
And possibly further reducible.

Your critical comments, please.

Got a call from an individual wanting to know whether
they should accept a job assignment developing an onboard
vehicle electrolysizer.

Naturally, they should RUN AWAY as fast as they possibly
can! The bottom line is that the numbers do not add up by
a country mile. Worse yet, seemingly innocuous changes or
cost cuts (such as using a fanbelt or substituting stainless
steel for platinum) push the numbers totally over the edge.

The PRIMARY output product of an onboard vehicle
electrolysizer is useless low grade heat. The hydrogen
produced is utterly and totally negligible in comparison
when using stock alternator and similar mainstream vehicle
components.

You thus have a DYNAMIC BRAKE that behaves exactly
the same as leaving your emergency brake set all the time!

A detailed analysis appears here.

Latest GuruGram #99 is on Some Possible Filterless Digital
Sinewave Synthesis Techniques.

Sourcecode is available here.
And additional GuruGrams here.

Latest GuruGram #98 is on the artwork and resources
for my upcoming Mount Graham Aerial Lumber Tramway
paper I'll be presenting at Discovery Park on September 20th.

Sourcecode for the newer material is available here.
And additional GuruGrams here.

The fire service sure has made a lot of major improvements
lately. But three items still sorely needed are...

     INTELLIGENT FIRE HOSE - Whose indestructable
     RFID number can be read in the dark from three feet
     away or preferably an entire hose bed at a time.
     After being drug six blocks and otherwise abused.

      THE PITOT GAUGE HAS GOT TO GO -- Give
      us a no-bullshit pipe that measures pumper or
      hydrant flow to 1500 GPM and 350 PSI. Wirelessly
      plus eight inch backlit verbal digits.

     SOLITON PULSE FIRE STREAMS -- High tech
     should give much longer range, tighter patterns.

Its a long story, but the last calculator we have available
for ordinary magic sinewaves is way back at glacial Gen 2.
One example would be here for eight pulses or here
for the rest of the gang.

I'll try to get these out of the dark ages. Meanwhile, high
speed modern calculators are found for most other
magic sinewave types here with their tutorial here.

Updated and improved our Arizona Auction Reources
Library page.

Your own local regional custom auction finder can be
created for you per these details.

A reminder that I have a pair of eminently collectible
and restorable commercial silent movie projectors
available from the 1908 era. Both Eastmans that
served in the Cliffton, Arizona theater for decacdes.

They are presently in pieces and are UPS shippable.
Please let me know if you are intereted in acquiring
these.

Got a phone call asking about the The Hexadecimal
Chronicles. This was probably my strangest and rarest
of books.

At one time, there was a continuing need to switch from
16-bit hexadecimal to decimal to signed decimal in the
Apple IIe world. You did not want to stop and run a special
program, nor could you afford the then outrageous hex
calculators.

So, this was mostly a big fat table lookup converter that
did the job for you. Much of it was directly set off a
Diablo 630 Daislywheel, so the production costs were
quite low. There were also ASCII lookup charts and
a circular slide rule for 6502 machine language relative
branch calculations.

I have a copy in front of me now and still occsaionally
use it for the odd ASCII control command lookups.

We will gladly combine shipping on our various
eBay aucttion and store offers. All you have to
do is email us for a quote. We most definitely do not
use shipping charges as a profit center.

Can a magic sinewave be created that needs zero
filtering? One that would set a fundamental, zero out
a chosen number of low harmonics, and only use absolutely
minimial carrier energy anywhere else?

The trick here would be some games that suppress any
higher frequency carriers. Done trhough cancellation or
spread spectrum, or whatever.

Consider making an 0.53 amplitude magic sinewave by
alternating "Original" and "Bridged Best Effeciciency
each quadrant. The first 30 harmonics would be zero on
each quadrant. And the difference between how the
carrier supression was done on each quadrant would
leave you with an unsupressed average carrier of 0.014
amplitude! And trivial to further filter.

While things would be spectacularly better for even lower
amplitudes, considerable work remains near unity output.
The possibilities here, if valid, boggle the mind.

More in GGMS.PDF.

Latest GuruGram #97 is on "GGMS" Fifth Generation
Split Magic Sinewaves.

Sourcecode is available here.
And additional GuruGrams here.

At one time, the finest audio foresnics gear in
the world was a Kay Sonograph II.

Which was a hard wired FFT computer that could
do such tricks as identify an individual bird call.

This two foot cubic beast sold for something like
$75.000.00. I've put one on the Alvin Pile and it
is likely to quickly go for pocket change.

But the emperor is wearing no clothes!

A brand new paper suggests some useful
new developments in electrolysis. These
use nanostructures and related technology
to approach the efficiency of conventional
platinum based electrode technology.

But, because of the staggering loss of exergy,
electrolysis is fundamentally a destroyer
of value. A kilowatt hour of electricity is
ridiculously more valuable that a kilowatt
hour of unstored hydrogen gas.

Even when done "perfectly", the process is
equivalent to 1:1 exchanging US Dollars
for Mexican Pesos.

There ALWAYS will be more intelligent
things to do with high value electricity
( such as grid, wind, pv, or alternator )
than instantly and irreversibly destroying
most of its value.

Additional analysis here.

If you are serious about being an eBay seller,
you simply MUST be willing to issue a refund
at any time for any reason.

If a customer is unhappy for any reason you
do not want them staying that way. Done
properly, the mere timing of a refund offer
will in fact elimiate any need for an actual
refund.

Most of the time, you should refund WITHOUT
demanding the item return. Chances are this
will prove simplest and cheapest in the long run.

Protection against unreasonable losses are
first and foremost your 30:1 SBR. Offering
a fifteen day inspection privelege on higher
ticket items (say $79 or higher) also heads
much of possible hassles.

Plus, of course, not trying to sell abject
and misdescribed trash to questionable buyers
in the first place.

More eBay selling secrets here.

When is something way too good to be true?

Our latest fifth generation self-supressing Magic
Sinewaves would seem to beat out freeze dried
holy grail in a spray can. But I sure could use your
help to make sure there is no fundamental gotcha here.

Magic sinewaves are newly discovered mathematical
sequences that have extreme potential for the highly
efficient digital synthesis of power sinewavs for such
apps as synchronous inverters, industrial speed control
systems, electric autos, PFC power quality controls, and
various aerospace and telecomm apps.

Magic Sinewaves allow you to force any chosen number
of low harmonics to precisely zero in theory and to
astonishingly low values in quantized practice. They do
so using an absolute minimum number of efficiency
robbing swwitching transitions. And are thus considerably more
efficicnt than conventional PWM .

Magic sinewaves to date shared a common characteristic in
that the first few uncontrolled harmonics ended up rather
large, although never significantly larger that its fundamental.
This has created filtering and high frequency energy issues.

For instance, an amplitude 0.20 8 pulse per quadrant Bridged
best efficiency MagicSinewave might zero out its first 30
harmonics. Its first uncrolled harmonics might have values of
h31 = -0.96731572096, h33 = 0.9350634087, h35 = 0.03182340473
and h37 = 0.00042586168.

Curiously, there is an "interleaved" magic sinewave variant that
seems to demand very nearly THE EXACT OPPOSITE in the
way of its low uncontrolled harmonics! A "normal" amplitude 0.20
8 pulse per quadrant magic sinewave might have its first uncontrolled
harmonic value of h31 = 0.96364176, h33 =-0.9388459125874, h35
= -0.0319192013367 and h37 =-0.000398802934595.

Which strongly suggests doing the first quadrant of your magic
sinewave with a regular 8, the second quadrant with a bridged best
efficiency 8, lathering and repeating.

Over half a cycle, the lowest suppressed carriers should dramatically
reduce. Perhaps from h31 = -0.96731572096 or h31 = 0.96364176,
clear on dowm to h31 = - 0.0037! Or an astonishing 52 decibels of
unwanted carrier suppression and that is BEFORE filtering!

Meanwhile, I'll try to work up a GuruGram on this. Please report any
possible problems you see with this strunning breakthrough. It seems
simply too good to be true.

Ferinstance, if all of the energy is now in in the sinewave and there
is negligible harmonic energy, where do the sharp corners come from?
Are there any rude surprises hidden in the really high harmonics?

It pays to thoroughly test any switches or relays
you put up on eBay to make sure they really are
what you think they are. An Ohmmeter with an
audio continuity feature is handy for this, along
with a veriable voltage dc supply or a switchable
wall wart.

You'll also want to spell out exactly how many
switch positions there are and which are sustained
and which are momentary and which are inactive.

Ferinstance, an on-off-(on) switch has three positions,
a sustained left, a center off, and a momentary
spring return right.

The coil on a reed relay will have several hundred
Ohms of resistance and should be easy to find.
Typically a 500 Ohm coil will be found on a 5 volt
relay and a 2000 Ohm one on a 24 volt relay.
A quiet "plonk" and "unplonk" should be heard
when enough voltage is applied to activate.

The need for an eBay image can sometimes
be avoided by linking to a manufacturer's web
site with appropriate data.

Here's one sneaky way to have an image without
having an image. Just place this HTML code at the
beginning of your eBay listing...

         <a href="http://www.rockwellautomation.com/
           products"><img src="https://www.tinaja.com/
           images/bargs/clickfor1.jpg"></a>

Be sure to be in the HTML coding mode rather than
the normal mode when you do this!

Naturally, you change the link to make it reach the
desired info source. And change the image for
your own "generic" one.

Which might look like this...

                 
The image will appear in the listing but not
as thumbnails or in the gallery.

We pride ourselves in having the finest images
on eBay, bar none. Secrets of image post prep
are found on our Auction Help library page.

I suppose the key centralmost secret is simply
this: Lock all horizonal and vertical lines to one
pixel accuracy.

Once this is done, all sorts of "sliding" and "symmetry"
tricks become possible. A summary of many of our
tools appears here.

The "FFT" fast Fourier transform is not particularly
fast. It also has some severe problems that can
cause grief. Needed windowing causes a mix of real
data and envelope artifacts. Plus the algorithm is
hard to write yourself from scratch.

For my Magic Sinewave work, I overwhelmingly prefer
to use the plain old textbook DFT. Otherwise known as
ordinary trig. The math dramaically simplifies when you
are only switching unity height pulses.

Your fundamental is found by taking each pulse pair
and adding up the difference between their starting
and ending cosines. You then modify the answer by
4/pi for one quadrant or 1/pi for a whole cycle.

Your harmonics are cound by taking each pulse pair
and adding up the difference between their starting
and ending cosines multiplied by the harmonic number.
You then modify the answer by 4/(n*pi) for a quadrant
or 1/(n*pi) for a whole cycle. You then further normalize
your harmonic by dividing it by the fundamental.

Utterly clueless U-Tube and Facebook users
continue to attack our eBay image background resource.

Kiddies, DO NOT USE A HUGE .BMP BITMAP FOR
WALLPAPER! Especially if you are stealing it off
somebody's elses site and gobbling their bandwidth!

You also might also avoid stealing an image that is
intentionally ugly! Only a tiny portion of this catalog
is normally cut and pasted as a working tool.

One more time...

    If you do not understand exergy, you SHOULD
    NOT be pissing around with electrolysis.

    If you do understand exergy, you WILL NOT
    be pissing around with electrolysis.

The outcome is not the least in doubt. Additional resources
can be found here, here, and here.

If you haven't touched an eBay inventory item
for over a year, chances are you never will.

And on the basis of a fundamental "shit floats
to the top" principle, they will end up gobbling
storge space and hiding all the good stuff as well.

So, we are finally in the process of filtering out
both the abject trash and the stuff that never seems
to get done out of our inventory. With the goal of
zero unlisted items just in time for the next
major auction.

Much of this will end up on the Alvin Pile where
unbelievable bargains should abound for the
next few weeks.

A reminder that Wesrch is an outstanding source
of new technical papers and insider industry info.

I remain totally mystified as to why anybody
would buy a pv solar panel. Except possibly to
steal government dollars.

Or otherwise scam funds. Or delude themselves
about the true facts of net energy.

These are all gasoline destroying net energy
sinks since not one net watthour of pv solar
electricity has EVER been produced. Nor will
any until several years after the price drups under
a quarter per peak watt in today's dolars.

Much more in our Energy Fundamentals and
More Energy Fundamentals tutorials.

We are phasing out both our crystals and our
crystal oscillator modules on eBay.

You'll find outstanding assortments for a penny
or two each. The modules make outstanding
plug-and-go signal sources for CMOS or TTL.

How many flavors of magic sinewaves are there?
I know of four at present...

    REGULAR - These are spaced 1/2-1-1-1-1/2
    and reject the first 4n-2 harmonics.

    DELTA FRIENDLY - A regular variant that
     uses half its transitions to guarantee three
     phase compatibility and rejects the first 3/4
     n - 1 harmonics. They need one half the storage.

     BEST EFFICIENCY - These are spaced 1-1-1-1-1/2
     and reject the first 4n harmonics and give the most
     efficient bang for the buck.

     BRIDGED BEST EFFICIENCY - Spaced 1-1-1-1-0
     and always end or start at 90 degrees. Reject
     the first 4n-2 harmonics with 2 fewer flips.

It pays to put your personal value added where
it can do the most good. With this in mind, we will
be making some adjustments in our eBay mix with
more stuff ending up on the Alvin Pile.

We don't have any special info on telephone stuff
or older video receivers, so we will be flushing these.
As will bunches of low end oscilloscopes that have
been taking up storage space. Our time delay relays
seem to be causing more unhappiness than they
solve, even at a penny on the dollar, so all but the
best of these will also be gone.

We had some problems with UPS totally trashing
an indestructable item, so we also will be rethinking
heavier items. Possibly limiting them to hand
delivery to certain Phoenix or Tucson locations.

We will also be trying more use of Craig's Lists.
There are now more Arizona ones, but sadly nothing
local as of yet.

A "minimilist" approach to photographic light
boxes and tents is to simply light an area with
four to six bare CCFL miniature fluorescents and
help them along with a white reflective sheet or
two.

Being sure to defeat the flash on your camera.

Such a setup can be particularly useful in reducing
hot spots on any "bright metal" areas on your items
being photographed.

The tools in Imageview32 can be enormously useful
in brightening dull or murky areas. Partciularly if
"multiple exposures" are in use.

A new movement underfoot goes by the name of
"Steampunk" In which you seek out a retro retro
polished brass and hand crafted ancient steam engine
look for the latest in high tech whatever.

With, of course the latest in machined titanium
thrown in for effect. Think "Old Remington
typewriter off a UFO " as a keyboard ferinstance.

Hand crafted pride in workmanship is foremost.

We occasionally have some stuff useful for
steampunk on our eBay store. Old timey style
binding posts, fifties dials, knobs, and drives, and
an occasional gem knife switch or whatever we
pick up at public utility auctions.

I've still got many thousnds of Apple IIe era files
I'd like to update and CD convert. I am amazed that
a totally obvious product -- An Apple IIe to USB or
memory card interface simply does not exist.

Yet this should be a very much in demand $19 product.

Proof to me that the upcoming total demise of books
and magazines will happen a lot sooner and a lot
more totally is that I cannot even give away copies of
Science Magazine that are over three months old.

Interfacing the bargain high power light dimmers
we now have on eBay should not be that big a deal.

These are intended for use with a DMX receiver. You
can find out more about DMX here or here.

The dimming rules are as follows: Each dimmer channel
expects a half millisecond "on " pulse. The time delay
of this on pulse from the last power line zero crossing
determines the lamp brightness. A very short delay
causes full brightness. A very long delay near 8.33
milliseconds causes very low brightness.

And no pulse at all leaves the lamp off. In normal
DMX receiver, an addressible number is sent
and decoded as 255 = max bright and 0 = off.

You otherwise have the option of relating your time
delay to a potentiometer position, a dc control
voltage or the average strength of an audio or musical signal.

A source of DMX receivers appears here while
this PIC based ap note gives you the crucial receiver
code for your own designs. Or here is an
ancient history solution of mine that uses nothing but a
hex inverter.

Got a call from an individual wanting to know why plain
old springs could not be used for regenerative braking
and similar apps. It turns out their energy density is far
too low. Typically being 1 watthour per kilogram or about one
twenty-fifth that of a lead acid battery.

The regulation is also usually poor and the round trip losses
from thermodyamic irreversibility are usually unacceptably
high. Which is why there isn't all that much spring wound
stuff around anymore. We might expect new aps with
very low power electronics, though, where poor efficiency
may not be all that big a deal.

More in our Energy Fundamentals tutorial and its
Latest Update.

Latest GuruGram #95 is on Spread Spectrum
Carrier Suppression SSCS of Magic Sinewaves.

Sourcecode is available here. An emulator and
demo here And additional GuruGrams here.

Got another helpline call asking about HP
oscilloscopes. I have no idea where the outrageous
urban lore myth that HP builds usable oscilloscopes
ever came from.

Their last decent scope was the 130C. It has been
steadily downhill ever since. To them, "Trigger"
was a horse.

HP oscilloscopes are, of course, banned in all civilized
nations of the world as cruel and inhuman
punishment.

Added some more video download links to our
home page. Mostly networks, some others.

There is no way that wide availability and
interest in on demand HDTV will not totally
choke the web on bandwidth. And soon.

Just put a very nice HP 8112A pulse generator up on eBay.

And noticed that it is one of the very fiew pieces
of test equipment with a "LOL" button on it!

Pulse generators have pretty much peaked in their
popularity, being replaced with far more versatile
arbitrary function generators.

Nonetheless, the 8112A is probably the latest and
finest pulse generator of all time.

Apparently the "water powered car" scams
are now going gangbusters. Complete with totally
bogus product reviews.

My take on them appears here.

He He He. The old "Big gold strike in Hell"
ploy just worked like a champ.

In which others at an auction "inadvertently
overhear" you and an associate talking about
the OTHER (and presumably better) auction
going on a mile or two away.

Whoosh. Gone. Bye. Bye.

The original "gold strike in Hell" did have a tiny
problem."   Prospector can't get into heaven because
of overcrowding. Sidels up to the bar and whispers
"Big gold strike in hell". All the other prospectors
instantly clear out.

Prospector doesn't stay anyway "I dunno. Could
be something to it."

Just had an inept quasi-governmental bureaucracy
insist I keep live fax access at all times for their once
per quarter possible activity.

This is a dilemma. Our ancient fax machine has to be
turned on before each use. A separate line for a maybe
opportunity is not justified. Yes, you can run fax from
inside Windows XP. But the latest computers no longer
include a modem!

And there is no way I know of to decide whether a
fax is incoming without picking up the phone. Which
causes hassles for all other users.

And those email fax for hire services violate our rule
of avoiding fixed expenses. Not to mention keytowing to
an obsolete bureaucracy. Dremanding fax for routine
transactions in this day and age is inexcusible.

The solution here appears to be social engineering.
You tell the bureaucracy the fax number and when
they bitch about not reaching it, you turn it on.

Farm sibling explaining why he kept feeding raw
pork to city slickers: "Its the only trick I know, Sis.".

Latest GuruGram #94 is on An Expanded
and Improved Magic Sinewave Calculator.

Sourcecode is available here. The revised
calculator here And additional GuruGrams here.

As we noted two days ago, some Magic Sinewaves
may have identical unsupressed carrier harmonics
for a given amplitude independent of "n".

This suggests that spread spectrum carrier suppression
SSCS technuqies may be useful to reduce the size
of the first few uncontrolled harmonics BEFORE filtering.

Effects would be highest on low amplitudes. Memory
size, efficiency, and program complexity would likely be
involved as tradoffs. Let's estimate the best that can
be done as a limit...

A best effciency low amplitude 8 ppq magic sinewave
can have its problem harmonics approximated by...

       27  29   31   33   35   37   39   41
        0     1    -1   ~0   ~0  ~0   ~0   ~0 

   ... for unwanted low rms harmonics of 1.4. Combine
one quadrant of n=8 and one of a bridged n=8.5...

      27   29   31   33   35   37   39   41
        0     1    -1   ~0   ~0  ~0   ~0   ~0
        0     0     1    -1   ~0   ~0  ~0   ~0 

   ... gives us unwanted low rms harmonics of 0.7 for
a 6 decibel improvement. Now, try 8, 8.5, 9, and 9.5
over one full cycle ...

      27   29   31   33   35   37   39   41
        0     1    -1   ~0   ~0  ~0   ~0   ~0
        0     0     1    -1   ~0   ~0  ~0   ~0 
        0     0     0     1   -1   ~0   ~0   ~0 
        0     0     0     0     1   -1   ~0   ~0 

   ...gives us unwanted low rms harmonics of 0.353
for a 12 decibel improvement.

And a full 8 different magic sinewaves over the
eight quadrants of two fundamental cycles should give
us up to 18 decibels of unwanted harmonic reduction
BEFORE filtering.

Some more math, simulations, and explorations of
tradeoffs obviously need done here.

Please welcome CCS Compilers as our newest
banner advertiser.

They have a line of innovative new C compilers
for PIC Microproprocessors and dsPIC
DSP devices from Microchip.

Our latest Magic Sinewave Calculator
seems to be giving an astounding result that
so far does appear real...

The critical uncontrolled 0.53 harmonics for
a n=8 best efficiency magic sinewave are...

    h31 =   0.00000000000
    h33 = -0.77885548201
    h35 =  0.57886655491
    h37 =  0.17991844863

The critical uncontrolled 0.53 harmonics for
a n=23 best efficiency magic sinewave are...

    h91 =   0.00000000000
    h93 = -0.77885548201
    h95 =  0.57886655491
    h97 =  0.17991844863

The same beyond ten decimals! And thus...

Apparently the carrier suppression for
a given amplitude is INDEPENDENT of n!

The results appear identical for both
best efficiency and bridge best efficiency
magic sinewaves as well. Minor exceptions
for very small "n" and some variation for
Delta Friendly.

Despite the repetitive calculations having
got there from wildly different routes and
totally different initial guesses!

This is both scary and "horribly significant".
I don't have the faintest clue what it all means.

Expanded and revised our Auction Help
library page.

Your own custom regional auction finder can
be created for you per these details.

Updated and improved our Magic Sinewave
Calculator. Many solutions up to n=23 are
now available for Best Efficiency, Delta
Friendly, and Bridged Best Efficiency .

Much of the code is now more cleanly
exandable to larger "n", and new debugging
tools have been added. With the latest JavaScript
in Firefox 3, typical speeds are well under a second
for all but the highest "n".

I'll try to work up a GuruGram on this shortly.
More in our Magic Sinewave library pages.

I've been recycling a copier coin interface
whose components will go on sale on eBay.

The project got off to a good start when I
paid $12 for it at an auction and it had
$68.85 in bills and change in it. It got even
better when the main problem seemed to
be a bad crimp on the power connector.

At any rate, these devices are incredibly
sophisticated. A very elaborate serial code is
used in vending machines to communicate
between coin mechanisms, bill changers, the
product machine, card readers, and the
controller electronics.

The apparently standard modern code is
called MDP/ICP short for Multiple Drop
Bus Internal Communications Protocol.

Its hard-to-find details appear here.

The system basically uses 9600 baud
serial comm in a host controlled master
slave request-response configuration.

I probably now have a good solution to the
coefficient blowup problem in Gauss-Jordan
high nxn solution of our Magic Sinewaves.

Changing the order of the angles from high
to low and alternating rows from high to low
harmonics seems to make a four decade or
more reduction in coefficient blowups.

The row sequence is now fully programmable
per these details. For really high n, you could
optimize each case for minimum coefficient
buldup.

Got a consulting request from someone who
"invented" a "new" integer square root and
would I please help them market it.

I promptly bailed when I asked how it compared
to Knuth, and got a reply of "Knuth Who"?

This topic got beat to death decades ago in
first generation pubs like Apple Assembly
Lines . And Google gives just under 3 million
hits on the topic.

Microchip has an interesting variant that
is division free. And this site is a diploma
mill that will write you all the new integer
square root algorithms you want to pay for.

Even if a new integer square root was for real,
its highest and best use would probably in fact
be publication in a Microchip or similar ap note.

Marketability would appear to be zero. There
are zillions of competitive solutions, each
with platform and situation specific advantages.

And, even if there were a groundswill of popular
demand for the new scheme, it would simply be
stolen rather than paid for.

Sadly, and much as I hate to say it, there
is very little demand these days for clean,
elegant, fast, short, and jaw dropping code!

It is far cheaper and simpler to throw another
million calculations or a gigahertz or two of
storage at the problem. And adding real floating
point is no big deal these days.

My favorite 16 bit integer square root is
table lookup. Let's see you beat the speed
on this one.

Certainly one of the more bizarre sagas in
the entire history of the Arizona Fire Service
seems to be finally winding down.

Some details here and here.

One major difference between "haves" and
"have nots" these days simply lies in how much
math they use and understand.

The available fast and cheap math tools and techniques
these days utterly boggle the mind. It is utterly trivial
to throw another hundred million calculations at
a problem. And even more trivial to pick up secret
info once available only to professional insiders.

Back in, say, 1963, a zillion dollar Olivetti calculator
available only to an aerospace firm could print out
a mortgate amortization for you. Or do one of my
first innovative math projects - finding out which
8 bit digital sequence could give you the least sour
musical notes. ( as in 116 - 123 - ... - 232 ).

With only a week's work.

These days, starting with a really bad math
background ( I nearly failed repeat remedial trig in
high school ) I'm routinely up to finding subtle
problems in 50 linear equations in 50 unknowns,
and exploring outrageously complex nonlinear
and ( apparently ) nondeterministic high order
math that may lead to some genuine energy
efficiency breakthroughs.

We seem to be making progress on our
expanded Magic Sinewave calculator and
are now up to 20 pulses per quadrant out of
a goal of 23.

It turns out that solving nxn equations with
64-bit math most definitely can cause
problems for large n. Outrageously large
coefficients can easily result.

Intermediate workarounds for the magic
sinewave math are to write the equation
rows from high to low angles and the
rows themselves as amplitude first then
higher harmonics working downward.

Further refinements in equation sequencing
and initial guessing may be needed for
even higher values of n.

Latest GuruGram #93 is on How to Totally
Trash an Onboard Verhicle Hydrogen
Electrolysizer.

Sourcecode is available here. And additional
GuruGrams here.

Updated our GuruGram library page.

Alexander Graham Kernatski was the
first Telephone Pole.

Larger nxn linear equations solved by
Gauss-Jordan reduction can in fact
easily blow up with JavaScript's 64-bit
math.

Some of the coefficients in our new
Magic Sinewave calculators were already
up in the hundreds of millions. Before
the higher nxn convergence problems..

Changing the equation sequence from
high to low angles and the row sequence to
amplitude then highest harmonics early
seems to be a useful workaround.

A heavier solution would be to pick
each new row for its best behaved
coefficients and then shuffle the
remaining rows for same.

Early large numbers can build up
at an alarming rate as they ripple
through.

More as testing progresses.

Latest GuruGram #92 is on Some Bitmap
Circular Lettering and Related Techniques.

Sourcecode is available here. And additional
GuruGrams here.

Here is my response to a recent media inquiry...

There IS credible published and peer reviewed
evidence that a 5% hydrogen injection into a
conventional ICE can have moderately significant
pollution and economy benefits.

However, I strongly feel the staggering losses
involved in onboard electrolysis utterly and totally
and ludicrously overwhelm any potential benefits.

And that is BEFORE amortization.

The point most of the enthuasiasts do not pick up
on is that the greater the electrical load on the alternator,
the greater the mechanical load placed on the engine.

You ALWAYS have to input a lot more energy than
you get back.

Because of inherent inefficiencies, an alternator driven
on board electrolysizer PRIMARILY produces waste
HEAT. And is thus a DYNAMIC BRAKE that works
pretty much the same way as having your emergency
brake partially on. All it can hope to do is REDUCE
your gasoline mileage.

I feel the whole process is clearly an electrocity.

A thermodynamic fundamental principle called
"exergy" absolutely GUARANTEES that electrolysis
from high value sources ( such as grid, pv, wind, or
alternator ) for bulk hydrogen generation flat out
ain't gonna happen. Because a kilowatt hour of electricity
is ridiculously more valuable and has a ridiculously higher
quality than a kilowatt hour of unstored hydrogen gas.

There ALWAYS will be more intelligent things to do with
the electricity. EVERY time. I feel that elctrolysis is
fundamentally a DESTROYER of quality and value.
Unavoidably and irreversibly.

The process is essentially the same as stupidly 1:1
exchanging US dollars for Mexican pesos.

Further, the available energy that can be sent through
an ordinary fanbelt is utterly trivial. The efficiency of
an ordinary car alternator is quite low. Often less than
50%. Thus, any conventional attempt at on-board
electrolysis is doomed before it even begins.

At most, I feel that only homeopathic quantites of
hydrogen could be produced. Which trips the
PLACEBO EFFECT many are purportedly observing.

Further, using stainless steel electrodes instead of
the required platinized platinum ones trashes the efficiency
even worse. Owing to the hydrogen overvoltage of iron
found in any intro electrochem book.

As does not recognizing that electrolysis is a CURRENT
DRIVEN process that requires exotic switchmode
techniques for efficient control. Other current sources
( or, worse, direct connection ) are inherently very
inefficient.

Yet another inefficiency source is that electrolysis can
be endothermic, exothermic, or thermoneutral. Higher
rates of gas production are almost always well into the
exothermic for further unavoidable losses.

Other approaches ( such as on board supply tanks or
exhaust gas reformation ) MAY eventually prove of
benefit. I feel this appears HIGHLY UNLIKELY to me
in that they interfere with other legitimate developing
means of improving ICE performance.

The fact that no legitimate automotive engineer has
come up with anything to date strongly reinforces this
belief.

Another gotcha: The EPA has made it a felony for an
individual to IMPROVE their gas mileage or REDUCE
their pollution on their vehicle! The law clearly states that
tampering with any pollution control device ( as hydrogen
injection would surely do ) is a punishable felony.

It does not matter in the least whether the tampering
makes things better or worse.

Not quite the latest GuruGram #89 is on A partial
History of the Gila Lumber and Milling Company.

This one is a tad different from our usual GuruGram
technical fare.

I have long been fascinated by the Mount Graham
Aerial Tramway. Which was a spectacular local
lumber tramway ( circa 1924 ) that operated over
one vertical mile of incredibly hostile terrain.
And, while uniquely engineered, turned out to
have been an economic failure, lasting only a
little over one year of operation.

At any rate, one of the key historical documents
was L.O. Martini's A Partial History of the
Gila Lumber and Milling Company. Which
I've transcribed here in machine readable form.

Sourcecode is available here. And additional
GuruGrams here.

Both real-world LED lighting and fantasy
electrolysizers share a common characteristic:

These are both CURRENT OPERATED
devices And, by definition, traditional current
sources are very HIGH impedances.

Thus, you can easily end up burning up most
of your input energy in your SOURCE , rather
than in your load.

Worse, if you stupidly try to drive either of
these from a voltage source, you can
easily end up with way too much current,
way too little, or highly erratic operation.

The secret solution lies in always applying
"just enough" voltage to produce the current
you need at any given time.

To do this, you start with an efficient switchmode
voltage stepdown converter. Then you sense your
output current as best you can and then use
feedback to hold your output current where
you want it.

Our point being that unless you have a very
fancy switchmode circuit between your power
input and the current output to your electolysizer
or LED display, you are GUARANTEED
rotten efficiency.

Consulting services available.

It pays to really dig deep into obscure
second tier auction resources. Lately I've
found these to be paying off big time.

Far better than the regulars.

Some recent ferinstances: A distant
community college provided some
exceptional big ticket items. A rare
aerospace action by a liquidator did
even better. A normally vapid auction
site for cities and counties yeilded
some highly useful items. But only
after a bluff was called on a reserve.

Several "over the transom" email
offers were lowball bid. And awarded.
Even after the few items that were grossly
overvalued, the rest of the lot continues
to generate long term profits.

A national auctioneer did not bother
with local ads on industrial inventory.
Single item bids did quite well on
a stock barn auction where nobody
else spotted value.

In another case, a bankruptcy attorney
decided to hold their own auction.
Very few showed up. There can be
a distinct advantage when you are the
only bidder.

The bottom line being continuing
vigilance. Avoid getting in a rut where
you only check the few obvious biggies.

And continue to seek out the nonobvious.

Much more on our Auction Help library page.

Your own custom regional auction finder can
be created for you per these details.

A reminder that we have two exceptionally
rare classic 1908 commercial movie projectors
available. These should restore beautifully
and presently are in UPS shippable pieces.

You can email me for more details.

What is a good strategy for online bids
at a mixed online/live auction?

You ALWAYS have a big advantage in
being there and bidding live, so rule number
one is to avoid all online bidding.

If you must bid online, make sure you are
approved at least a day ahead of time.

NEVER bid too early! All this does is
telegraph your intent and draw interest
in the item. Besides starting pissing
contests.

Early in the auction, it is a good idea to
place a token shill bid on something you
do not feel strongly about and have no
chance of winning. This makes sure you
are properly online and your bid machinery
is working.

Needless to say, on any item that you really
want, you place BOTH your current bid
of ONLY the amount needed to be highest
bidder. AND your proxy maximum.

Do this ONCE AND ONLY ONCE!

Since you are unlikely to have to pay your
proxy max ( you will almost always win for
less or be outbid ), your proxy max can
be somewhat higher than what you are really
willing to pay. Perhaps 30% more.

There is a tradeoff on your best bidding
time. Too early and others have a chance to
respond. Too late, and the usual auctioneer
online screwups may miss your bid as their
equipment chokes.

Something like seven lots or nine minutes
early may be a good optimum choice.

And avoid getting stressed out. If it was
such a big deal, you should have attended.

Much more on our Auction Help library page.

Your own custom regional auction finder can
be created for you per these details.

The solution to high value Magic Sinewave
Calculator blowups may be to reverse the data
sequence during Gauss Jordan improvement.

Consider going from 0 to 90 degrees. The
first line might have sine coefficients from
.07 to .95. The second from .21 to .92. Scaling
the top line raises the .95 to 13.57. And
subtracting 3 times the top line from the second
raises the .92 to 40.7 or so.

We clearly are getting a huge spread in our
intermediate data values. These could cause
major truncation errors for very large n x n.

Instead, going from 90 to 0 degrees seems
to keep many of the intermediate data values
in line. And thus should help bunches.

More as the adventure unfolds.

A recent helpline caller asked about standard
humidity calibration. It turns out there are
standard common salts whose saturted
solutions create known chamber humidities.

A good tutorial appears here.

Bottom line: Lithium Chloride is 11.3%.
Magnesium Chloride is 33.1%. Sodium
Chloride is 75.2%. Potassium Sulfate is
97.6%

There is a mild temperature variation
found in the Greenspan Calibration Tables.

Oops department. I was in the process of
extending our Magic Sinewave Calculator
and it started choking around 19 pulses per
quadrant. Convergence first became erratic
and then ultimately blew up on certain values.

The most likely reason is that Gauss Jordan
Elimination of 38x38 linear equations may
be beyond what 64 bit math can provide for
certain input coefficient combinations.

I'm hoping that changing the sequence of
variables can provide a useful fix. Our
older calculators can handle much larger
quadrant counts, but are excruciatingly slow.

More as this situation unfolds.

Apparently I erred in my estimate of car alternator
efficiency. It is a lot worse than I expected, being
typically 50 to 60 percent for partial loading and
much less otherwise.

These figures from Wikipedia, who quote the
Bosch Automotive Engineering Handbook

There is no particular reason you would
expect auto alternators to be all that efficient.
Primary design goals are low cost, small size,
light weight, and reliability.

And the actual losses are rather small when
compared against total engine horsepower.

Their performance could be upgraded by
going to synchronous rectification, more
and better iron and copper, improved
magnetic paths, exotic magnets, smaller
air gaps, redesigned windage, etc...

A reminder here why hydrogen from
electrolysis via a car alternator flat out
ain't gonna happen: (1) virtually ALL of
the input mechanical energy gets
converted to useless low grade heat,
creating a DYNAMIC BRAKE that
trashes your fuel economy; and (2)
the fanbelt alone limits input power
to several hundred watts at most,
leaving you with homeopathic quantities
of recoverable hydrogen gas energy .

Picked up an interesting Hot Knife Webbing
Cutter that we have refurbed and placed on
eBay. A Cutex TBC47-LH.

This microprocessor controlled automated
machine will cut any number of any lengths
pieces of rope, webbing, ribbons, or similar
textiles for you. The adjustable temperature
hot knife automatically seals the ends of
any "fray sensitive" materials.

The engineering here is interesting. The
internally heated knife is more or less stationary
but spring loaded. A one turn rotary platen pushes
the material up against the knife for cutting and
sealing. .

An interesting logo website is found here.

First sale doctrine lets you use somebody
else's logo specifically for an item resale.
Otherwise, extreme caution is recommended.

To change the color of a logo, try this trick:
Get into Imageview32, and convert to
grayscale. Then convert back to RGB
true color. Then dink with the color balance.

One of many "not even wrong" assumptions of the
alternator electrolysis crowd goes something like
this:

The alternator runs all the time anyway, so all of
that "wasted" electricity is best converted to
hydrogen. Rather than sitting around unused.

Uh, yes, the alternator does run continuously.
But the MECHANICAL LOAD the alternator
presents to the engine very much depends on
the electrical load drawn. The more energy
you take out, the more you have to put in.

You ALWAYS have to put in more mechanical
energy than you get out as electrical energy.

Worse, only a tiny fraction of the alternator
output ends up as recoverable hydrogen energy.

The overwhelming majority ends up as USELESS
HEAT that would NOT have been produced had
electrolysis not been attempted.

Thus creating a DYNAMIC BRAKE that LOWERS
your gas mileage and causes the vehicle to stop in
less than its normal coasting distance.

Further analysis here.

Updated and improved GuruGram #65 that is our
Executive Guide to Magic Sinewaves.

The main issue was that our dramatically improved
Ultra Fast Magic Sinewave Calculators came out
shortly afterward. Per this tutorial.

Minor update adjustments have also been made
to our Intro to Magic Sinewaves and Delta
Friendly Magic Sinewaves

Present Magic Sinewave activities include new
demo chips, extending the calculator to 23 pulses
per quadrant, and exploring other microprocessor
options that can more elegantly deal with 12-bit
or 16-bit delay data.

Partners and associates welcome on this highly
significant alternate energy efficiency breakthru.

One sure way to make your idea look good
and oppositions bad is to internalize all of their
costs and externalize all of yours.

Some new web posts claim that in-wheel electric
motors can eliminate the 15% to 20% transmission
losses in a conventional automobile. The only
tiny oint in the flyment is that the motor is unlikely
to be 90 percent efficient, the controller 85 percent
efficient, and the wiring 97 percent efficient.

Thus replacing 20% losses with 26% ones.
Nice thinking, kiddies..

Now, there are many obvious advantages to
in-wheel motors. Especially involving true
four wheel positraction, vehicle stability, and
regenerative braking. Plus clutch-free and
continuous variability.

As the railroads discovered years ago.

Offsetting these advantages are an inherently
higher ratio of unsprung to sprung weight,
exposure vuneribility, and the tendancy of most
motors to run better at high speeds and thus
need their own torque increasing transmissions.

I'll stick with my earlier prediction: The optimum
number of electric traction motors in a vehicle
is not four. It lies somewhere between 4000
and 40,000 micromotors instead.

Latest GuruGram #91 is on Some Bitmap
Perspective Lettering Algorithms and Utilities.

Sourcecode is available here. And additional
GuruGrams here.

And the working utility here. Plus a practice bitmap here.

Recently received a few emails and calls
over homopolar motors, generators, and
Faraday Disks. We covered them here and
here.

These are the only known rotary dc generators.
While they make excellent student papers,
they are otherwise useless because of their
extreme low voltage, high current operation.
Which usually makes their commutation losses
very high and their efficiency very low.

These devices are extremely hard to understand,
Faraday and Lorenz laws must be treated together!
Your frame of reference is not at all obvious.

And it is easy to wrongly convince yourself there are
some overunity possibilities. Which, of course,
there are not.

It turns out that special relativity demands that
you cannot tell if a uniform magnetic field is
stationary or rotating. You can prove this by
trying to rotate a magnet on a string with one
beneath it.

There are no such thing as "magnetic lines"
and any attempt at "cutting" them to explain
this class of machines is utterly futile. Besides
being not even wrong.

The best discussion of this is in the Feynman
Lectures. Physics Volume II, section 13.10
in particular.