conservation of Euros

[John Larkin]

On Sun, 30 May 2010 12:06:42 -0700 (PDT), Bill Sloman
<bill.sloman@ieee.org> wrote:

On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplier would be an easy start and would be interesting to a
number of posters here.

John

 

[Bill Sloman]

On May 24, 11:26 pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

and
Bill *knows* business.  Massive investment that pays off zero-to-one
times is better and less risky

It can be a lot more profitable - the margins on turn-key projects can
be very high - and I never claimed that it wasn't risky.

 than modest investment that pays 100x.

John Larkin made it perfectly clear that his less ambitious projects
don't always lead to successful products. A one hundred-fold return on
development investment would be remarkably high. A successful and long-
lived product might make it, but probably not if you discounted your
cash-flows correctly. And how many products do you have to develop
before you find one that is popular enough to sell persistently and in
volume, without attracting the attention of the larger-small
manufacturers in the area, who can afford to develop an ASIC to handle
most of the electronic function to let them sell something that does
the same job a lot cheaper.

But James Arthur "knows" business - with the same sort of confidence
with which he "knows"  economics - and this sort of consideration
passes him by.

He does stuff.

Really?

You don't.

Not at the moment. If I did manage to find myself a job, I'd have less
tine to point out when you were spouting nonsense, so I can understand
your enthusiasm for seeing me more profitably occupied, but don't try
and persuade us that my incapacity to get a job makes your output any
less ill-informed.

--
Bill Sloman, Nijmegen

 

[John Larkin]

On Sun, 30 May 2010 17:06:35 -0700 (PDT), Bill Sloman
<bill.sloman@ieee.org> wrote:

On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplier would be an easy start and would be interesting to a
number of posters here.

What c-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

John

 

[Bill Sloman]

On May 20, 7:14 pm, dagmargoodb...@yahoo.com wrote:

On May 20, 9:57 am, John Larkin wrote:

Bill Slomanwrote:

You might be better off turning the company into a cooperative. At
least that way the poeple who inherit the company will have some
understanding of how it works and why it works that way.

Ah yes, free love and drugs for everyone.

The John Lewis Partnership and the Co-operative Group in the U.K.
don't work that way

http://en.wikipedia.org/wiki/John_Lewis_Partnership

http://en.wikipedia.org/wiki/The_Co-operative_Group

but a parochial Californian like James Arthur might not know this.

There was a fad in the '70s around here, centered in Berkeley of
course, for co-op buisnesses; I still have a couple of the books, like
"We Own It!" It was an interesting experiment. There seemed to be two
available outcomes, sad failures and hilarious failures.

The idea goes back almost a century earlier, and has had more
competent exponents. Not that your sources would have told you about
them.

A very few are still around. There's a co-op bakery on 9th avenue, not
bad stuff actually. They close down one day a week just to meet and
talk. And talk. And talk. I hear that it's painful for the majority.

If it actually were a co-operative, the majority would have worked out
how to curtail discussion. It's not difficult to persuade a loquacious
minority that their audience isn't all that interested - though the
techniques that work when the audience is physically present don't
work here.

I saw a cool thing on PBS just a few days ago about that era.

Not so much "cool" as "consistent with your prejudices".

A fellow was explaining how he and a bunch of fellow college students
with liberal educations surged out, full of energy and socialist
utopianism.  They fled to the hills (e.g. Foxfire), to live together
in peace, equality, and free love.  A commune, where all is fair and
free.

They quickly learned just throwing seeds in the ground did not a farm
make, and that equality sucked.  The chicks split, and then the guys
soon after.

The guy winced, sheepishly, explaining/defending: they'd had their
eyes opened, only wasted two years doing it, and didn't hurt anyone in
the process...

If they'd applied their ignorance to creating a capitalist utopia -
something based on Ayn Rand perhaps - they could have failed just as
abjectly.

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 20, 8:37 pm, Greegor <greego...@gmail.com> wrote:

On May 20, 12:14 pm, dagmargoodb...@yahoo.com wrote:



On May 20, 9:57 am, John Larkin wrote:

Bill Slomanwrote:
You might be better off turning the company into a cooperative. At
least that way the poeple who inherit the company will have some
understanding of how it works and why it works that way.

Ah yes, free love and drugs for everyone.

There was a fad in the '70s around here, centered in Berkeley of
course, for co-op buisnesses; I still have a couple of the books, like
"We Own It!" It was an interesting experiment. There seemed to be two
available outcomes, sad failures and hilarious failures.

A very few are still around. There's a co-op bakery on 9th avenue, not
bad stuff actually. They close down one day a week just to meet and
talk. And talk. And talk. I hear that it's painful for the majority.

I saw a cool thing on PBS just a few days ago about that era.

A fellow was explaining how he and a bunch of fellow college students
with liberal educations surged out, full of energy and socialist
utopianism.  They fled to the hills (e.g. Foxfire), to live together
in peace, equality, and free love.  A commune, where all is fair and
free.

They quickly learned just throwing seeds in the ground did not a farm
make, and that equality sucked.  The chicks split, and then the guys
soon after.

The guy winced, sheepishly, explaining/defending: they'd had their
eyes opened, only wasted two years doing it, and didn't hurt anyone in
the process...

James

These experiments resemble cults, where
they attempt to shut out REALITY and
replace it with their BELIEF!

Americans who have never heard of successful cooperative organisations
find it difficult to believe in them.

http://en.wikipedia.org/wiki/John_Lewis_Partnership

http://en.wikipedia.org/wiki/The_Co-operative_Group

<snip>

Only the most CULT LIKE liberal zombies
refuse to learn from the reality they
discover, and refuse to acknowledge
the real world limitations that came up.
Usually they blame external causes.

Americans who get their news from a right-wing press don't have access
to all that much reality.
Greegor does seem to value his ignorance.

All of this makes it even more bizarre
that Bill Slowman is over 50 years old
yet his political rantings have the
maturity of idealistic and inexperienced
18 year olds in the USA.

I think Gregor is saying that in the US anybody over 18 realises that
it is wise to act as if right-wing propaganda should be taken
seriously, rather than rejected as transparently self-serving drivel.

It's also bizarre that he obsesses about
United States economic politics considering
he claims to be an Aussie ex-pat living
in Netherlands.

Some members of this user-group post a lot of transparent drivel. I
amuse myself by pointing out the more nonsensical aspects of their
posts.

For a NON-PRODUCER living off his wife
to urge socialism is cartoon like.

Most retired people have a hobby. Mine is reviling nitwits. If this
user group had socialist nitwits, I'd revile them too.

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 22, 6:35 am, Bill Bowden <wrongaddr...@att.net> wrote:

On May 21, 3:24 am,Bill Sloman<bill.slo...@ieee.org> wrote:



After that you try to say you're
not trying to ""sell"" socialism??

Not really. Americans ignore the way the rest of the world does
things, despite the fact that some ways of running a country are
better managed outside the USA. Health care is the the classic example
- US health care cost half as much again per head as the best foreign
systems (in France and Germany) while providing no better health care
for prosperous employed Americans than the French and German systems
provide for everybody, while providng much worse health care for the
less well-off part of the US population.

Actually, health care costs in the US are inflated due to the
additional R&D costs other countries don't pay.

http://www.studentnewsdaily.com/commentary/the-cost-of-free-governmen...

"Countries with government-run health care save money by relying on
the United States to pay the research and development costs for new
medical technology and medications. If we adopt the cost-control
policies that have limited innovation in other countries, everyone
will suffer."

The bulk of the R&D costs of drug development is paid for by the
companies developing the drugs, and recovered from the people who buy
the drugs, many of them outside the US. The US healthcare cost per
head is directly comparable with the healthcare costs per head in
Frnce and Germany, which are only two-thirds as high. The French and
German doctors use the same drugs as their US equivalents, and since
all of their patients are fully insured - as compared with 65% of US
patients (another 20% of US patients are under-insured) they probably
cover a larger proportion of the drug development costs than their US
equivalents.

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 17, 7:23 pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Mon, 17 May 2010 04:28:09 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 17, 5:43 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 16 May 2010 19:05:54 -0700 (PDT), dagmargoodb...@yahoo.com
wrote:

On May 14, 2:31 am, Martin Brown <|||newspam...@nezumi.demon.co.uk
wrote:
On 14/05/2010 06:16, dagmargoodb...@yahoo.com wrote:

   Of course Marx himself was a n'er-do-well who never earned his keep,
a pseudo-academic parasite sponging off patron Engels.  Engels in turn
coasted off the family business.  Marx made his living guilt-tripping
Engels with econobabble, a fine tradition carried on by Marxists
today.

Engels saw first hand what greedy industrialists were doing to their
workers in the Lancashire cotton industry. Boiler explosions were
commonplace up until the Vulcan insurers made a stand and insisted on
proper boiler safety inspections. And in cases of tampering with safety
relief valves they would not pay out.

It was common practice to overstoke the fire before the first shift and
add weight to the pressure relief valve - this resulted in several large
scale boiler explosions destroying big mills in the early morning and
killing many workers in the Lancashire cotton industry.

Destroying your factory is a bad business model.  That quickly self-
limits.  Besides, nowadays we sue or jail those people.  Too much, in
fact.

http://www.camdenmin.co.uk/technical-steam/historic-steam-boiler-expl...

Articles on the history of boiler insurance show that the US had a worse
record despite having the advantage of seeing the innovations in UK
boilers. Some element of NIH played a part but mostly it was that
industrialists greed was paramount and the workers powerless. eg.

http://www.casact.org/pubs/proceed/proceed15/15407.pdf
first page and page 7 under Normal Loss Hazard

Interestingly and ironically enough, that emphasizes the need to
identify defects and eliminate high risk insureds to minimize
underwriting loss rates.

"Experience has also shown  that  the scientific  examination  and
inspection  of  insured  boilers  produces  a
declining  loss  ratio."

   "To each according to need" really means "From you to me."  "Dear
Fred, I need that grocery money, and I deserve it, luv Karl, xoxoxoxo
P.S. Stop exploiting me! KM"

It makes reasonable sense to pay your workers a living wage for the work
that they do rather than pay them less than they can sensibly live on.
Ford was about the first in the USA to actually do this.

In the UK there were some decent industrialists mostly of quaker
families who did treat their workforce fairly - examples include some
household names like Pilkingtons, Cadbury, Bournville, Marks&Spencer.

But most of the rest were complete bastards who built large factories
and employed the equivalent of bonded labour stuck very high density
slum housing. It was not surprising that unions were formed in some
cases the manager really did hold the whip hand - literally.

As John pointed out, that was a transient effect, an unusual, historic
dislocation.  Machines meant that few could farm what had previously
required the toil of many.  So there were lots of workers looking for
work.

Short term, that's painful.  Long term, that's creative destruction,
society re-allocating resources from something no longer needed, to
something people do want and need.

The pattern repeats:

http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2010/05/15/BUPJ1DEGG....

Manufacturing seeks cheap labor and makes lots of stuff. Pretty soon,
that labor isn't cheap any more. Eventually the world may run out of
places with cheap labor.

Intelligent manufacturing seeks to automate the duller and more
repetitive parts of the job, but moving to China does require less
investment. Effectively, investment in manufacturing automation has
been on hold since since the US started exporting its low-paid jobs to
China, and the rest of the world joined the rush.

China's costs are increasing, as Japan's and Korea's did. Capital
moves in to poor countries for the cheap labor. After a generation or
two, the locals have learned the technology and bootstrapped their
education and skills; they are no longer cheap labor but competitors.
So the capital (including the new capital from the formerly cheap
countries!) moves on, looking for more cheap labor.

It's sort of foreign aid that actually works. It's a natural
propagation mode of technology. It's wonderful.

But your "investment in manufacturing automation has
been on hold " statement is entirely bogus.

And your evidence to support this claim is? I didn't say that
investment had stopped - merely that there wasn't all that much of it
because it was cheaper to ship low-skilled work to low wage countries.
My authority is Alan Greenspan's autobiography.

<snip>

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 30, 9:26 pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplier would be an easy start and would be interesting to a
number of posters here.

What c-multiplier? I've been away for a week.

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 31, 2:20 am, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Sun, 30 May 2010 17:06:35 -0700 (PDT), Bill Sloman



bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

Many years ago I was worried about the temperature coefficient of the
Early effect, but sanity set in before we got worried enough to start
measuring it - we finally solved our problem in a slightly more
expensive way that wasn't sensitive to the Early effect.

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 31, 2:20 am, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Sun, 30 May 2010 17:06:35 -0700 (PDT), Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too. But we're wrong John. Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

Many years ago I was worried about the temperature coefficient of the
Early effect, but sanity set in before we got worried enough to start
measuring it - we finally solved our problem in a slightly more
expensive way that wasn't sensitive to the Early effect.

--
Bill Sloman, Nijmegen

 

[John Larkin]

On Sun, 30 May 2010 19:17:14 -0700 (PDT), Bill Sloman
<bill.sloman@ieee.org> wrote:

On May 31, 2:20 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 17:06:35 -0700 (PDT), Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too. But we're wrong John. Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

Models, models! Can't anybody solder any more?

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

In theory?

Semiconductors are generally pretty consistant. And when you don't
know if the Early voltage is 50 or 5000, and you don't trust the
models, measurement is the thing to do.

John

 

[Bill Sloman]

On May 31, 6:36 am, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Sun, 30 May 2010 19:17:14 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 31, 2:20 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 17:06:35 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

Models, models! Can't anybody solder any more?

Simulation is quicker, and easier to probe.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

In theory?

Semiconductors are generally pretty consistant. And when you don't
know if the Early voltage is 50 or 5000, and you don't trust the
models, measurement is the thing to do.

The Gummel-Poon model isn't too good at low collector voltages, but
manufacturer's Gummel-Poon data isn't going to give you a 10:1
uncertainty - or anything like it - in the Early contribution to Vbe
at collector voltages above a few volts.

You want your c-muliplier to work with a Vce around 200mV, so you
probably do need a VBIC simulation, or a well-instrumented real
circuit = measuring 140dB of rejection is demanding.

--
Bill Sloman, Nijmegen

 

[John Larkin]

On Mon, 31 May 2010 02:05:38 -0700 (PDT), Bill Sloman
<bill.sloman@ieee.org> wrote:

On May 31, 6:36 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 19:17:14 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 31, 2:20 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 17:06:35 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

Models, models! Can't anybody solder any more?

Simulation is quicker, and easier to probe.

And often useless in predicting how real electronics will work. Oops,
did I say "work"? Sorry.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

In theory?

Semiconductors are generally pretty consistant. And when you don't
know if the Early voltage is 50 or 5000, and you don't trust the
models, measurement is the thing to do.

The Gummel-Poon model isn't too good at low collector voltages, but
manufacturer's Gummel-Poon data isn't going to give you a 10:1
uncertainty - or anything like it - in the Early contribution to Vbe
at collector voltages above a few volts.

You want your c-muliplier to work with a Vce around 200mV, so you
probably do need a VBIC simulation, or a well-instrumented real
circuit = measuring 140dB of rejection is demanding.

No, it's not all that hard, although I doubt I'll get anything like
140 dB at low frequencies. I suppose I'll have to do it myself.

John

 

[John Larkin]

On Mon, 31 May 2010 10:10:11 -0700 (PDT), Bill Sloman
<bill.sloman@ieee.org> wrote:

On May 31, 7:00 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Mon, 31 May 2010 02:05:38 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 31, 6:36 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 19:17:14 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 31, 2:20 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 17:06:35 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

Models, models! Can't anybody solder any more?

Simulation is quicker, and easier to probe.

And often useless in predicting how real electronics will work. Oops,
did I say "work"? Sorry.

And occcasionally misleading about real circuits, but then bread-
boards can be misleading too, but you wouldn't admit that any of your
bread-boards had lead you to the wrong conclusion, would you.

Only in cases where they actually did.

You don't have that problem.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

In theory?

Semiconductors are generally pretty consistant. And when you don't
know if the Early voltage is 50 or 5000, and you don't trust the
models, measurement is the thing to do.

The Gummel-Poon model isn't too good at low collector voltages, but
manufacturer's Gummel-Poon data isn't going to give you a 10:1
uncertainty - or anything like it - in the Early contribution to Vbe
at collector voltages above a few volts.

You want your c-muliplier to work with a Vce around 200mV, so you
probably do need a VBIC simulation, or a well-instrumented real
circuit - measuring 140dB of rejection is demanding.

No, it's not all that hard, although I doubt I'll get anything like
140 dB at low frequencies. I suppose I'll have to do it myself.

Your enthusiasm for work is an example to us all, though you probably
pay yourself better than any other employer would - who else would
believe your claims about the quality of the work you do (let alone
think you can do)?

Your enthusiasm for not-work is a different sort of inspiration. I pay
myself about what I'd make as an employee, and I'm not the
highest-paid person in my company.

I have tons of things to do, like finish this stupid program I'm
working on instead of hiking. I was hoping someone would volunteer to
measure the c-multiplier lf rejection, to cross-check the sims and/or
my own eventual testing. Right now Rob has got the AM502 amplifier I'd
like to use. (I should buy another one, but they're *expensive* these
days. That's another story.) That amp, and some signal averaging on a
scope, should resolve 140 dB, if that's the number.

Obviously you're not going to do it.

John

 

[Bill Sloman]

On May 31, 7:00 pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Mon, 31 May 2010 02:05:38 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 31, 6:36 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 19:17:14 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 31, 2:20 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 17:06:35 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

Models, models! Can't anybody solder any more?

Simulation is quicker, and easier to probe.

And often useless in predicting how real electronics will work. Oops,
did I say "work"? Sorry.

And occcasionally misleading about real circuits, but then bread-
boards can be misleading too, but you wouldn't admit that any of your
bread-boards had lead you to the wrong conclusion, would you.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

In theory?

Semiconductors are generally pretty consistant. And when you don't
know if the Early voltage is 50 or 5000, and you don't trust the
models, measurement is the thing to do.

The Gummel-Poon model isn't too good at low collector voltages, but
manufacturer's Gummel-Poon data isn't going to give you a 10:1
uncertainty - or anything like it - in the Early contribution to Vbe
at collector voltages above a few volts.

You want your c-muliplier to work with a Vce around 200mV, so you
probably do need a VBIC simulation, or a well-instrumented real
circuit - measuring 140dB of rejection is demanding.

No, it's not all that hard, although I doubt I'll get anything like
140 dB at low frequencies. I suppose I'll have to do it myself.

Your enthusiasm for work is an example to us all, though you probably
pay yourself better than any other employer would - who else would
believe your claims about the quality of the work you do (let alone
think you can do)?

--
Bill Sloman, Nijmegen

 

[Bill Sloman]

On May 31, 7:40 pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:

On Mon, 31 May 2010 10:10:11 -0700 (PDT),Bill Sloman



bill.slo...@ieee.org> wrote:
On May 31, 7:00 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Mon, 31 May 2010 02:05:38 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 31, 6:36 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 19:17:14 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 31, 2:20 am, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 17:06:35 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 30, 9:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sun, 30 May 2010 12:06:42 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 24, 11:26 pm, John Larkin
jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Sat, 22 May 2010 04:45:06 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:
On May 22, 1:41 am, dagmargoodb...@yahoo.com wrote:
On May 21, 5:06 pm, John Larkin

jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Fri, 21 May 2010 14:34:17 -0700 (PDT),Bill Sloman

bill.slo...@ieee.org> wrote:

The facts of the case are that you don't like developing complete
systems, bcause it takes too long and ties up too much capital and
engineering effort, and you've found yourself a niche where you can
develop useful sub-systems, some of which you can sell to several
customers.

Yes. Engineering is too valuable to sell once. Production can sell
copies of engineering for decades.

Your customers would probably be happier if you took on turn-key
development contracts, but that kind of big chunk of development takes
skills that you don't seem to have - perhaps wisely.
Big projects that go wrong regularly destroy the businesses that took
them on.

I have been in the systems business, and now that I have my own
company I never want to do it again.

Me too.  But we're wrong John.  Bill says we should do systems,

Actually I said that John might be wise to keep out of a risky area.
Since he doesn't like developing large complex systems, perhaps
because he isn't good at it, this is merely endorsing his preference.

I've done systems work. Lots of it. The economics stink, and the time
pressures interfere with sleeping and skiing.

They never worried me, but I don't ski - there are cheaper ways to
break a leg.

You are equally skilled at not doing simple electronics and not doing
complex systems.

True, but irrelevant. Not doing something isn't actually a skill.
Getting a Dutch employer to hire a 67-year-old would seem to require a
skill that I haven't got, but there's not much evidence that anybody
in the Netherlands has ever mastered it, so maybe you could interest
yourself in some more achievable goal?

Do some real electronics and tell us about it.

Measuring (not simulating) the Early effect feedthrough on the
c-multiplierwould be an easy start and would be interesting to a
number of posters here.

Whatc-multiplier? I've been away for a week.

Broad hint: look for a thread whose title includes the word
"C-multiplier"

Interesting thread. The Gummel-Poon model - as used in Spice and
freely
available from most manufacturers - doesn't model the Early effect all
that accurately.

LTSpice will also use the rather better VBIC model if you can get hold
of the parameters. The manufacturers don't make them available, and
when I wanted them I couldn't get any via any of the - very limited -
personal contacts I've got.

Models, models! Can't anybody solder any more?

Simulation is quicker, and easier to probe.

And often useless in predicting how real electronics will work. Oops,
did I say "work"? Sorry.

And occcasionally misleading about real circuits, but then bread-
boards can be misleading too, but you wouldn't admit that any of your
bread-boards had lead you to the wrong conclusion, would you.

Only in cases where they actually did.

You don't have that problem.

Not at this instant, but then neither do you - at this instant.

You can - in theory - measure them yourself, but you need to do it on
enough transistors to get some idea of the production spread (which
you won't necessarily get by measuring a succession of transistors
from the same batch).

In theory?

Semiconductors are generally pretty consistant. And when you don't
know if the Early voltage is 50 or 5000, and you don't trust the
models, measurement is the thing to do.

The Gummel-Poon model isn't too good at low collector voltages, but
manufacturer's Gummel-Poon data isn't going to give you a 10:1
uncertainty - or anything like it - in the Early contribution to Vbe
at collector voltages above a few volts.

You want your c-muliplier to work with a Vce around 200mV, so you
probably do need a VBIC simulation, or a well-instrumented real
circuit - measuring 140dB of rejection is demanding.

No, it's not all that hard, although I doubt I'll get anything like
140 dB at low frequencies. I suppose I'll have to do it myself.

Your enthusiasm for work is an example to us all, though you probably
pay yourself better than any other employer would - who else would
believe your claims about the quality of the work you do (let alone
think you can do)?

Your enthusiasm for not-work is a different sort of inspiration.

I'm not in the least enthusiastic about it. You seem to think
otherwise, for the - for you - excellent reason that it suits you to
think that way, and you don't have a problem about parading your
unrealistic fantasies.

I pay
myself about what I'd make as an employee, and I'm not the
highest-paid person in my company.

Really?

I have tons of things to do, like finish this stupid program I'm
working on instead of hiking. I was hoping someone would volunteer to
measure the c-multiplier lf rejection, to cross-check the sims and/or
my own eventual testing. Right now Rob has got the AM502 amplifier I'd
like to use. (I should buy another one, but they're *expensive* these
days. That's another story.) That amp, and some signal averaging on a
scope, should resolve 140 dB, if that's the number.

Obviously you're not going to do it.

Obviously. I've not got the equipment, and you haven't mentioned
paying for the help, which might justify getting the appropriate
equipment together. There is a Pico Technology USB scope sitting on
the other side of the room, but it isn't hooked up yet.

--
Bill Sloman, Nijmegen

 

On May 30, 7:47 pm, Bill Sloman <bill.slo...@ieee.org> wrote:

On May 22, 6:35 am, Bill Bowden <wrongaddr...@att.net> wrote:



On May 21, 3:24 am,Bill Sloman<bill.slo...@ieee.org> wrote:

After that you try to say you're
not trying to ""sell"" socialism??

Not really. Americans ignore the way the rest of the world does
things, despite the fact that some ways of running a country are
better managed outside the USA. Health care is the the classic example
- US health care cost half as much again per head as the best foreign
systems (in France and Germany) while providing no better health care
for prosperous employed Americans than the French and German systems
provide for everybody, while providng much worse health care for the
less well-off part of the US population.

Actually, health care costs in the US are inflated due to the
additional R&D costs other countries don't pay.

http://www.studentnewsdaily.com/commentary/the-cost-of-free-governmen...

"Countries with government-run health care save money by relying on
the United States to pay the research and development costs for new
medical technology and medications. If we adopt the cost-control
policies that have limited innovation in other countries, everyone
will suffer."

The bulk of the R&D costs of drug development is paid for by the
companies developing the drugs, and recovered from the people who buy
the drugs,

Yes.

many of them outside the US.

Are you saying Europeans use the latest drugs?

The US healthcare cost per
head is directly comparable with the healthcare costs per head in
Frnce and Germany, which are only two-thirds as high.

The French and
German doctors use the same drugs as their US equivalents,

Equivalents = generics, proving Mr. Bowden's point.

and since
all of their patients are fully insured - as compared with 65% of US
patients (another 20% of US patients are under-insured)

they probably
cover a larger proportion of the drug development costs than their US
equivalents.

Nope.

Yahoo around a bit for "cancer drugs" and your fave EU country. When
I do I see loads of horror stories--desperate patients on old
therapies, unable to get the latest (American) drugs ideal for their
specific conditions. Too expensive, one assumes.

--
Cheers,
James Arthur

 

[Bill Sloman]

On Jun 1, 4:04 pm, dagmargoodb...@yahoo.com wrote:

On May 30, 7:47 pm,Bill Sloman<bill.slo...@ieee.org> wrote:



On May 22, 6:35 am, Bill Bowden <wrongaddr...@att.net> wrote:

On May 21, 3:24 am,Bill Sloman<bill.slo...@ieee.org> wrote:

After that you try to say you're
not trying to ""sell"" socialism??

Not really. Americans ignore the way the rest of the world does
things, despite the fact that some ways of running a country are
better managed outside the USA. Health care is the the classic example
- US health care cost half as much again per head as the best foreign
systems (in France and Germany) while providing no better health care
for prosperous employed Americans than the French and German systems
provide for everybody, while providng much worse health care for the
less well-off part of the US population.

Actually, health care costs in the US are inflated due to the
additional R&D costs other countries don't pay.

http://www.studentnewsdaily.com/commentary/the-cost-of-free-governmen....

"Countries with government-run health care save money by relying on
the United States to pay the research and development costs for new
medical technology and medications. If we adopt the cost-control
policies that have limited innovation in other countries, everyone
will suffer."

The bulk of the R&D costs of drug development is paid for by the
companies developing the drugs, and recovered from the people who buy
the drugs,

Yes.

many of them outside the US.

Are you saying Europeans use the latest drugs?

Yes. They also develop quite a few of them.

http://en.wikipedia.org/wiki/Pharmaceutical_drug

Note that of the ten "block-buster" drugs listed - more that $1B
revenue per year - seven belong to European=based drug companies. The
three drugs developed by US-based companies are good for twice the
revenue of the seven developed by European-based companies -
Atorvastatin (Lipitor) on its own matches their combined revenue. I
was recently switched from Lipitor to the european-developed
Simvastatin, probably because Simvastatin is now cheaper (patent
expired) and equally effective, though Pfizer would like you to think
otherwise

http://blogs.wsj.com/health/2007/09/05/the-statin-switcheroo-lipitor-and-simvastatin/

The US healthcare cost per
head is directly comparable with the healthcare costs per head in
Frnce and Germany, which are only two-thirds as high.
The French and
German doctors use the same drugs as their US equivalents,

Equivalents = generics, proving Mr. Bowden's point.

Patented drugs - which is what Mr. Bowden was talking about - aren't
available as generics. If they are produced by anyone who doesn't own
the patent, the producer has to pay royalties to the patent-holder.
You claim to understand business and high technology and should
understand this. The European pharmacutical companies make much too
much money out of their patented drugs to countenance the sort of
piracy that goes on in parts of the less developed world.

and since
all of their patients are fully insured - as compared with 65% of US
patients (another 20% of US patients are under-insured)
they probably
cover a larger proportion of the drug development costs than their US
equivalents.

Nope.

Yahoo around a bit for "cancer drugs" and your fave EU country.  When
I do I see loads of horror stories--desperate patients on old
therapies, unable to get the latest (American) drugs ideal for their
specific conditions.  Too expensive, one assumes.

More likely, not cost-effective enough. Drug companies are good at
tweaking drug formulations and claiming that their latest (and most
expensive) recipe is what the patient really needs. Evidence-based
medicine rarely confirms their claims. Patients are less inclined to
wait until some touted variation has made it through clinical trials
and been certified by the Cochrane collaboration.

http://www.cochrane.org/

Desperate patients form the basis of a lot of right-wing ancedotes
about European health care. Any time now you are going to tell me that
Stephen Hawking would be dead by now (which he isn't) if he'd been
looked after by the British National Health Service (which has looked
after him ever since he was diagnosed with motor neurone disease back
in 1963) as right-wing critics of "socialised medicine" have been
known to do.

The British National Health Service is appreciably more careful with
its money than US health insurers or the French and German
equivalents, but the British Health Service is also appreciably
cheaper than its French and German equivalents. It doesn't deliver the
same public health statistics as the French and German systems (which
do as well as the best of US healthcare, available only to the fully
insured 65%) though better than the US system can manage fpr the
population as a whole.

--
Bill Sloman, Nijmegen

 

On May 27, 8:56 am, Joerg <inva...@invalid.invalid> wrote:

dagmargoodb...@yahoo.com wrote:
On May 26, 8:30 pm, Joerg <inva...@invalid.invalid> wrote:

[snip]

Fair enough. But why not try to fix the current system first?

You mean of 1099s, Schedule C, D, E, and payroll and employer matching
taxes, of Social Security tax, Medicare Tax, Alternative Minimum Tax,
of Earned Income credits, and capital gains tax, each with their own
schedules and computations, each affecting the others?

Yes. We need to look at the tax code now and that from, say, 1960. Then
figure out why the heck it has gotten so complicated. Start rolling
back. One reason is easy and we already know it, it is evidenced in the
most recent 1099 reporting requirements.

I'm just really kind of at a loss. We're talking about replacing 26
op-amps, a few a/d's, a mess of power supplies, a handful of uC's and
a bag of trimpots with ... a resistor.

You'd rather re-tweak the trimpots on the kludge? Re-program the
break-points? That just ... well, I scarcely know what to say.

I guess the thing is this: ignoring the many other taxes eliminated
under the Fair Tax, our current income tax system all by itself is as
complicated as it is because everyone wants it that way. Each wants
their special part of it. To reform that, you'd have to win battles
with each and every one of them, which is impossible.

Things like AMT would have never become an issue at all if some
politicians wouldn't be so greedy.

That's why a visible tax is good. It discourages shenanigans. It's
not perfect, but far more accountable than a thousand hidden taxes.

[snip]

In contrast, the announcement of a "fair tax" could be like
hitting the ignition button. Kablouie.

[snip]

What's your concern?
That all retirees would pull their money out and rush into more durable
investments such as real estate. You might see nothing but a plume of
dust where your local bank used to be ;-)

That makes no sense to me.  Running out and buying stuff does not
improve your tax treatment, whether income-taxed savings are Fair Tax
exempted or not.

Sure it does. You buy that retirement place now -> You pay with taxed
money and the home is tax free. You buy the retirement place too late
when "fair tax" is in effect -> You pay with taxed money and you pay
another tax on the home.

Oh, now I see your point. But wouldn't that be great, a real, honest
"stimulus package"? That seems a darn sight better than bribing them
with taxpayer dollars to buy houses.

If someone did want to buy a bunch of stuff, that stimulates
manufacturing and other business, so it's fine for the country.  It
might be kind of dumb to blow all your dough, however.

The smart thing would be to take yourself, your family, your dough and
move to a place where there will be no new tax on stuff. Guadalajara
province is said to be nice ...

No joke, a lot of serious discretionary income would be gone.

Not income--that's the same. The overall tax rate is the same, it's
just superbly easier to pay.

One great advantage of the Fair Tax is that it removes moral hazard.
That is, it puts the tax you pay right out in front, for all to see.
And, unlike today, everyone pays it, so everyone has a vested interest
in keeping it low.

And some people will pay it twice. It's right in front of them, for all
to see.

We already discussed that. They'd compromise, and exempt Roth IRAs,
etc.

That, my friend, will do more to rein in spending and save the country
than just about anything else could.

You honestly believe that?

Yes.

It takes a wee political shift in a certain
direction and you have willy-nilly spending. Then ... "oh s..t!" ... we
have to raise the "fair tax" rate from 23% to 26%.

Fine, but that's when the people get engaged and say "No." Like with
property tax in California.

Right now the changes are so complicated, we get hornswoggled and
hardly even know who bit us. (Who can read the 76,000 pages of current
IRS code and regulations that are amended, on average, twice a day?)

And what do you do with states that have dug themselves into a hole?
Like California with its reckless spending for super fat bureaucrat
pensions. Do you give them a bigger chunk of the "fair tax" than states
that knew how to do a budget well? And how are you going to muffle the
public backlash from doing that?

That issue is unrelated--it's Fair Tax neutral.

(Sorry for the late post--I'm swamped.)

--
Cheers,
James Arthur

 

On May 27, 9:57 am, Joerg <inva...@invalid.invalid> wrote:

dagmargoodb...@yahoo.com wrote:
On May 26, 8:18 pm, Joerg <inva...@invalid.invalid> wrote:
dagmargoodb...@yahoo.com wrote:
On May 26, 9:27 am, Joerg <inva...@invalid.invalid> wrote:
JosephKK wrote:
On Sun, 23 May 2010 07:05:38 -0700, Joerg <inva...@invalid.invalid
wrote:
JosephKK wrote:
On Sat, 22 May 2010 08:39:13 -0700, Joerg <inva...@invalid.invalid
wrote:
So let's see, since we can't have an assessor then John Q.Public must
self-file into some computer system. "Hmm, so what do we enter here for
the materials? One box of nails, a pack of drywall screws, the hot dog I
had outside Home Depot. Don't remember the rest ..."
That is all recorded in the tax receipts.
What receipts? Case in point, and I was right behind the guy: Dude had a
huge cart in tow at the cash register. A toilet, two sinks, tile, pipe,
mortar, the works. He could barely pull it. Ka-ching ... "That'll be
eighthundred Dollars and .." He whipped out a huge wallet and paid the
whole chebang in cash. Dollar bills. No check, no credit card, no name
given. Now how exactly is this going to be recorded?
At a bare minimum, in the tax receipts that the store reports (by sale).
They may not know just who paid, but they do know it _got paid_ on those
items.

Do you honestly believe this guy would dutifully file and remit the 23%
"fair tax" from the amount he collects from the homeowner?

Businesses that sell to the public have to collect Fair Tax, just like
they collect slaes tax today. Can they cheat? Sure, I guess,
especially if they're small. Not if they're large--they'll get
caught.

Because of all the other record keeping involved with a real property
sale, yes.

What record keeping? I seriously doubt that uncle Leroy will remember
who rebuilt the deck 15 years ago. Or wants to remember.

Okay, I've got a short break so I'll try to catch up with you guys...
Uncle Leroy paid Fair Tax when he bought the building materials.  If
he hired a contractor, Leroy pays Fair Tax when he pays the
contractor.  No recordkeeping.  End of story.

Not at all end of story. Read what I wrote above about the guy who
bought the complete remodel materials with hard cash. He's going to make
uncle Leroy a special deal, provided that no paperwork crosses the table
and there will be a cash payment. Meaning the guy who did the work puts
all it this cash in his pocket. IOW, nobody ever paid tax for the work,
only for the materials.

But it's possible and common to pay people "under the table right"
now, so I don't see how that's different.  In your scenario at least
he would've paid Fair Tax on the materials, so you'd collect part of
the tax due.  Under an income tax the cheater gets off totally scot-
free, labor and materials.

Big difference: The guys who work in the underground economy are
typically at the low end of the income scale already. Their regular job
went away et cetera. They would hardly pay an income tax anyhow, maybe
10%. What was the number of Americans that paid nothing this year? Over
40% AFAIR. Now if there would be a 23% flat tax the motivation to go
under the table will increase big time because that's an even bigger
chunk than the few bucks he'd have paid in income tax. And he'd have to
collect and pay that right now (talking about bureaucracy ...).

Let's not forget, there's lots of people out there who wouldn't even
know how to file and pay that 23%. Because some have never even filed a
1040 (although of course they should have).

[...]

I guess (cringe) one advantage of a VAT is that it forces everyone to
become tax-collectors--no one wants to be stuck paying it, so they're
eager to collect their rebate from the next guy in the chain.

What rebate? If you aren't a business there will not be any rebate. "A
fond perdue" is how they often call that in Europe.

Rebate: for all the middlemen in the supply chain.

In countries with high VAT the underground economy is rampant. I know
it, I lived in such countries. This so-called "fair tax" will have the
same effect as a VAT because to the people it works the same way. They
do not care how it's called, a tax is a tax is a tax.

Yes, but ditto for income tax.

It'll increase, because 23% is more than 10%.

But things people buy in store _will_ be taxed, offsetting that.

I met a crew doing some work the other day. Of taxes, the chief said
"I don't pay those mthr'f'r's one f'n dime." He pays his crew, um,
discretely, too.

--
Cheers,
James Arthur