USB microscopes for very small SMT

[JosephKK]

I annoy kids when I come up with numbers in a couple of seconds, while
they're still fiddling with their Blackberry.


We still do a lot of arithmetic in our heads, at meetings and such.
But it's not classic adding/subtraction of digits, it's more like a
mental slide rule, rough "analog computing" estimates or rounding to
cardinal points (like 3/8 = 0.375) and then tossing in an estimated
interpolation to get a little closer. There are times when a 20%
accurate estimate is still very valuable.

But yeah, I'm not sure if I can still do long division, and even
subtraction seems silly by hand.

John

Each to their own, i still do add & subtract to 5 digits in my head,
2-3 digits by 2-3 digits multiply in my head, divide on pencil and
paper (up to 3 digits divisor and up to 8 digits dividend and
quotient), low accuracy (~2 digits) log, antilog in my head, low grade
circular functions (+/- 20%) in my head, and square roots on paper. I
used to do much better. Then again, my memorized data tables are
dying from nonuse. Just the same i still grab a calculator (or flip
to the one on my computer desktop, yes, it is always present) if more
than that is wanted.

 

[JosephKK]

The order of magnitude is usually good enough in such
circumstances. One significant digit is almost always enough.
That sort of "accuracy" was what I lost by using calculators as a
crutch. Annoying but I wasn't a bad trade, IMO.


I never worried too much about multiplication and division beyond
one digit. That's what the slip stick was for. One digit was
enough to verify the process, something now missing with calculator
(or spice) results.

That barely begins to be able to address the opportunities for
analysis that become available by using composition and decomposition.
Take any tough problem, break it into chunks that are easier to
understand, and then reassemble them and work with the assembly to get
to system behavior. This is the very basis of finite element analysis
style computer programs. This also describes SPICE in some ways.

 

[JosephKK]

No, not *your* work, I mean your negative feelings about 'programmers'.

I have to say, that I find C++ coders a strange species.
I think C++ is language disability brought forward by Stroustrup as he found
a way to work around his disability to program (think sequentially perhaps).
;-)
Blame this man ;-):
http://www.research.att.com/~bs/

Anyways I was so happy when Linus decided the Linux kernel should stay C,
else Linux would have been dead years ago.
And, on that subject, I prefer to look at the projects that succeeded.
You can learn little from bad hardware, but a lot from good designs.

Ah, we are getting to the point some here:
The dumbest do not learn from their own mistakes.
Only learning from your mistakes is not smart yet.
The beginning of smart is learning from others mistakes.
The journey level of smart learns from others successes as well.
The master level is when you embody learning by all means possible and
teaching accordingly.

Sadly, the world is now quite deficient of masters and journey level
people.

 

[Spehro Pefhany]

One person rice:
Small bowl, fill with water, 1 minute 40 seconds to heat it.
Now add 3 or 4 spoons rice.
Then, depending on the type of rice,
White fast cooking 15 minutes at 50%.
Brown rice, 45 minutes at 50%.
etc.
If you took the right amount of water, then the rice will be just right and dry,
you can then simply eat it without having to poor of any water.

These guys make nice fully automatic rice cookers that work that way--
perfect rice every time, and they keep it warm after making it:

http://www.zojirushi.com/ourproducts/ricecookers/ricecookers.html

I have the Neuro-Fuzzy controlled conventionally heated model... the
induction heated ones are probably a further improvement.

They're not insanely expensive-- newegg carries them.

Cheap knock-off types from Black and Decker etc. work okay too, just
not quite as nice. As cheap as $14. I bought a National
(Matsushita/Panasonic) type in University that still works okay. This
type uses a snap-action thermostat to detect (over) heating at the
middle of the cooking bowl and shut the cycle down. There's a round
combined sprung heater and thermostat that contacts the bottom of the
removable bowl. Mine is so old the bowl isn't even non-stick, which I
wouldn't recommend.


Best regards,
Spehro Pefhany

 

[JosephKK]

Yes, much increase in size comes from using libraries.
Sometimes where there is no need for one at all.
Examples is Qt4 in Linux, it is huge, while for many if not most
applications you could use for example xforms, that is very very small (I use nothing else).
And other big joke is if people start using SDL for every application that has audio
and or video.
But the advantage of the libraries is that they are so often used that they are pretty much bug free.
And they *can* add very powerful features.
So indeed the increase of bugs may not be that big.
If I was to code a formatted print in asm, or use C and libc with printf(),
then clearly the last method is what gives least bugs (and development time).

It is the same with hardware I suppose, I re-use circuits I have designed and tested in the past.
that saves time and increases reliability, although perhaps in some cases
you could, if you started from scratch, save a transistor....
In fact all integrated circuits are like software libraries, you may not
use all their features, but they provide instants solutions.
I do still remember designing my own video ADC in the seventies..
It is only because of the ever smaller transistor sizes that it has not become 'hardware bloat'.

But it has. There used to be a joke in the IC world "What do you do
with a million transistors?" Perhaps a 486dx2? a 512kbit memory? It
is within working professionals memory that 1 GB of ram was
supercomputer territory, now you need that much for a home PC.
Hardware bloat and software bloat are going hand in hand.

 

[Joerg]

But the quality of toasters has improved significantly with microcontrollers:
I remember as a kid, it was uncommon to find a toaster that would let you run
through multiple pieces of bread without them getting lighter and lighter on
each cycle, until by the 4th or 5th cycle the entire toaster was so hot it
would only "toast" for, say, 10 seconds before popping up the bread again even
on the darkest setting.

Our first family toaster was the kind where you had to open the sides
"just so". Then the slices would flip. If you did it wrong and wedged it
smoke would come out. It had no thermostat. The tool you used to
determine slice tan was on your wrist and said Timex or something on
there

To some extent this was probably due to our buying cheap toasters... we
probably went through a half-dozen sub-$20 toasters until my mother one day
bought a nice Braun unit... but of course in the store you have no way of
knowing how good the toaster really is, regardless of the price.

Braun used to be good. But we found their coffee makers didn't last more
than 1-1/2 years lately. So why spend extra for a name brand anymore?

Today most toasters just don't have this problem anyway. It's a case where
while, sure, 25 years ago there certainly were good toasters, there was also a
lot of crap... whereas today I believe that dirt-cheap microcontroller-based
toasters perform just about as well as the best anachronisms you might sitll
have around.

Until shortly after the warranty expiration date ...

They perhaps didn't do much torture testing before they started selling it.
I've worked at several places where, if the widget worked at room temperature,
you shipped it... and let your customers tell you -- via warranty
exchanges! -- thaat it didn't work so well in, e.g., Alaska or Arizona.

From an engineer's POV this thing isn't designed right IMHO. But
supposedly makes good coffee. Died on its first run :-(

 

[Joerg]

The Grand Ole Opry is performing at the historic Ryman Auditorium for
the next couple months. t was built long before electronic
amplification, and has some amazing acoustics. It was originally built
as a revival hall, and designed by the world's best acoustic engineers
of the day.

http://www.wsmonline.com

Hmm, tons of ads, even in the archives. But the tunes are cool and now I
know what Wompers Fried Sausages are

 

[Vladimir Vassilevsky]

Well, this is the day we disagree it seems.
I seems to me it is *as hard* for the hardware man to write bug free code,
as it is for any programmer.

There is a big difference between the projects for 2-3 developers and
the projects for 20-30 developers. In the later, there is no such single
person who has good answers for all technical questions. The overhead
increases like a square of the number of people involved. The law of the
big numbers comes into play also.

Have you seen like the ants are building the ant hill? Many ants are
pulling a straw in one general direction however each ant is pulling a
straw in the direction of his own.


No. This is what the market demands. Customers don't want to pay x10
amount of money and wait for x3 of time for the code to be completely
free of bugs.

They need a new language and a new discipline.

BASIC, Java, C# and ADA are there since quite long ago.
It is a mistake to think that a non-technical problem can be cured by
technical means.

All you need is love love, love is all you need.

....Until you get hungry or thirsty.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

 

[Joerg]

On Fri, 05 Dec 2008 23:24:34 +0000, Jan Panteltje wrote:
Anyways I have a way to make rice in the microwave that really works.
Do Tell! I do it on the stovetop and usually burn it. )-;

Thanks!
Rich

Two cups of water in a small pot. Bring to boil. Add a bit of butter
and a little salt. While stirring, add rice. Reduce to very slow
simmer, cover, and wait 10 minutes. Turn off the heat and wait another
10 minutes.

Jasmine rice is best.

That's it.

I've become a big fan of my little Zojirushi cooker. It just works. I'd
really recommend giving a real [*] rice cooker a try.

Too many gadgets! If the power goes out in the next earthquake, I can
still cook, make coffee, make rice or grits, bake potatoes, fry eggs,
do most anything on the gas barbecue.

But when it's really bad and the propane truck can't get through anymore
you'd be back to granola bars until depleted. I can go into the woods
and cut off some dead wood, then cook

 

[Joerg]

I once had a plastic filter cone; just the cone with a flange at the small
end; it could sit on top of a mug or on top of an ordinary coffeepot. It
was about $1.99 for just the cone. I've looked up "Mellita" and didn't
immediately see the one I had, but they might have one in the "coffee"
department at your local store.

Spelling was wrong, probably it's on this page:

https://shop.melitta.com/search.asp?SKW=MACM

I got a 15 year old electric Melitta coffee maker from my parents when
heading out to the university. A plain plastic version, nothing fancy.
It kept running the whole time, and then some. At least back then this
was a great brand. Most of the new stuff from other "name brands" is in
rather sorry state right out of the box.

 

[Joerg]

Maybe I'm having selective memories, but I remember seeing cars get
mileages in a really close ballpark as these EPA numbers; they just
kept them tuned properly!!

I had a 1.6L Pinto once, and routinely got about 30 on the highway -
that's a kinda rough estimate; I had an 8-gallon tank and could go
at least 240 miles. Admittedly, it's a small car, but I know how
to drive, so it was never in any danger of getting run over. >:->

Absolutamente. If you look real hard you can still get such cars but it
won't be a US model. Unless they took an Asian design under license like
GM/Geo used to do. And not all Japanese cars are really imports, for
example my wife's Toyota was built in Fremont, CA.

 

[Joerg]

There is a big difference between the projects for 2-3 developers and
the projects for 20-30 developers. In the later, there is no such single
person who has good answers for all technical questions. The overhead
increases like a square of the number of people involved. The law of the
big numbers comes into play also.

Have you seen like the ants are building the ant hill? Many ants are
pulling a straw in one general direction however each ant is pulling a
straw in the direction of his own.



No. This is what the market demands. Customers don't want to pay x10
amount of money and wait for x3 of time for the code to be completely
free of bugs.

Mine do.

BASIC, Java, C# and ADA are there since quite long ago.
It is a mistake to think that a non-technical problem can be cured by
technical means.

It can. But there needs to be a manager who enforces diligent use of
such technical means. For example, proper version control. Plus the use
of <gasp> a word processor by SW designers. They may hate it but unless
they do it'll quickly end up in a chaos. In my line of work agencies
such as FDA or FAA mandate that proper design history is kept. If you
don't, your product will not be granted permission to market.

Or to say it with the words of the R&D chief at my first employer: "If
you didn't document it then it didn't happen." He is a 12-year army
veteran and those guys know.

[...]

 

[Joerg]

My kettle, cone, and pot have been in daily use for about 15 years and
are holding up fine.

The guts of a coffee maker, a toaster, or a stove - temperature
extremes, steam, crumbs, oil - are horrendous environments for cheap
electronics.

It used to not be that way, back when more engineers knew about analog.
For example, one classic mistake I find in my consulting practice is the
use of too high impedances in low-energy sensing circuits. A little
moisture, a little dirt, and it'll be on the fritz. The right way is to
use low impedances and then pulse-measure. But it seems universities
don't teach that stuff anymore.

Yesterday a friend told me a sad story: They have a double oven, friggin
expensive, IIRC it's a GE. The controller up top is, unfortunately,
electronic and expensive, In ten years the third one has died so now
they consider this oven a total loss. That is pathetic.

Our double oven is over 35 years old. When it dies I will go to great
lengths to find a replacement that is controlled via bi-metal thermostat
switches. I don't want any PWM or whatever in there. Just relays,
bi-metal and switches.

 

[Joerg]

MANY years ago I designed a chip that used a combination of
temperature and color reflection to do toast "just right" ;-)

My experience, over these many years, is that "electronics" per se is
not bad, just that anything controlled by a uP can't be trusted...
unless an analog guy supervised the algorithm development ;-)

...Jim Thompson

.... _and_ unless it's mask-programmed. Flash retention years in hot
environments can easily drop to single digits.

 

[Vladimir Vassilevsky]

Mine do.

Joerg! Do you realize that the people like you (and me, too) are the
worse disaster to the economy then the dot coms altogether with the
toxic mortgages? I doubt you can drink enough of beer to justify your
existence as the member of the consumer society.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

 

[Vladimir Vassilevsky]

On Sat, 06 Dec 2008 11:12:12 -0600, Vladimir Vassilevsky



Banks still use COBOL.
Accounting types can program in COBOL, which is
really the point: a language that allows accountants to code
applications is a far better idea than trying to get a bored C++
programmer to understand accounting.

Certainly there are the specialized languages like Matlab, LabView etc.
where the main idea is liberating the non-specialist from the minor
technicalities. However 1) This can't prevent the mistakes in the
algorithm 2) The downside is the tremendous overhead and the inflexibility.

The three best programmers I know
were *not* formally educated as programmers.

"There is no and there can't be any replacement for the intelligence,
the experience, the common sense and the good taste" (B. Stroustrup)
One of the best circuit designers that I know is a medical colledge
graduate with no formal EE education.

It would take a combination of a new language and a new culture to get
us past the mess that programming is today.

Oh, come on. Programs are no conceptually different from the stuff sold
at Wallmart. A programmer is a blue collar job. "Better programming" is
a synonim of "cheaper programming". As the nearest example, look at the
lousy windows software used on Tek and LeCroy expensive (!) scopes.

A computing profession
that does everything carefully, on time, and bug-free would look
completely different from what's happening now.

Sure. Back in 70-x, they promised us that the true communism will be
built in 20 years.

Good reads:

"Dreaming in Code", about a group of world-class programmers who spent
years working on a product that was mostly undefined.

Why would the world class programmer waste his precious time on
something undefined?

"Showstopper!" about the coding of Windows NT.

"Road to the future" by Bill Gates of Microsoft
"Only paranoids survive" by Andrew Grove of Intel

VLV

 

[Joerg]

Joerg! Do you realize that the people like you (and me, too) are the
worse disaster to the economy then the dot coms altogether with the
toxic mortgages? I doubt you can drink enough of beer to justify your
existence as the member of the consumer society.

Oh, we do our part to keep the economy going. It's just that I'd rather
spend my money at the Thai place of the Japanese restaurant or other
local places than waste money on products with x1 design margins.

Or as my grandpa used to say, it doesn't pay to buy cheap stuff.

 

[Joerg]

Oh, come on. Programs are no conceptually different from the stuff sold
at Wallmart. A programmer is a blue collar job. "Better programming" is
a synonim of "cheaper programming". As the nearest example, look at the
lousy windows software used on Tek and LeCroy expensive (!) scopes.

Which brings up a question: Is there any 3rd party tool that I could use
to view Tektronix logic analyzer files instead of their TLAVu? IMHO that
is one of the most horrible pieces of SW I've ever had to use.

[...]

 

[JosephKK]

C is the only language that is portable to almost every processor.
It is close to the hardware, and was developed to write drivers,
to make writing drivers easier.
It is at a high enough level, especially when you count all the available
libraries and tools, to do very sophisticated programming.



After having coded the same applications again and again in asm for
different micros, for me the language *does* matter.



No 'goto' in C for me (but one could).
Asm is all jumps...



Well, this is the day we disagree it seems.
I seems to me it is *as hard* for the hardware man to write bug free code,
as it is for any programmer.



All you need is love love, love is all you need.
Beatles.

You and John are talking past each other.

Me i use what language is needed, if the mandated language is not up
to the task i say so.

Issues over pointers are red herrings.

 

[Joerg]

They're concentrating on "low power" (what every laptop needs, right?) so the
students naturally gravitate towards high-impedance designs...

Laptop designers seemed to have "un-learned" that as well. Until
recently when the Intel Atom came out and (some) laptop designers
"re-learned" the art. When Howard here in the NG mentioned the Samsung
NC-10 and I saw it gets >6h, like my old Compaq Contura used to, I
honored that laptop performance with an order. Seems others have done
the same since I initially received a backorder notice despite the fact
that "in stock" was mentioned. Should arrive in a few days

I often encounter folks who are quite unfamiliar with the scheme of
saving battery power via pulse action instead of high impedance.
Students need to re-learn the old art of obtaining some of their wisdom
from non-academic sources. I'd venture to say that >50% of the knowledge
I use in my practice is from non-academic sources. App notes, Internet,
ARRL books and such.