AoE x-Chapters, 4x.26, MOSFET current source, nodal analysis

[Jan Panteltje]

On a sunny day (Sat, 10 Aug 2019 08:44:21 -0500) it happened "Tim Williams"
<tiwill@seventransistorlabs.com> wrote in <qimhnb$i5d$1@dont-email.me>:

"Jan Panteltje" <pNaOnStPeAlMtje@yahoo.com> wrote in message
news:qim2na$vu$1@dont-email.me...

You silly widows users still do not understand the world is MUCH greater
than your latest auto update security hole.


Windows? No, I mean I don't even hear my Linux friends complaining of shit
that archaic. They used to. I assume it's been fixed. Are you using 90s
Redhat or something? Update that security hole!

What security hole?
It '&' is just part of the bash syntax and means run program in the background.

Or they still mumble about it but have resigned themselves to hopelessness
over the years...

Try reading a book on Unix, almost everything runs on Linux these days
except for some silly widows computers in homes of people
who have not bought / borrowed / stolen / rented / learned / what else have you a clue.

Why do it to yerselves?

And no this system is lemme see
~ # uname -a
Linux panteltje12 2.6.37.6-smp #1 SMP Sat Aug 3 19:23:48 CEST 2013 i686 AMD Sempron(tm) 145 Processor AuthenticAMD GNU/Linux

Slackware, probably the most sane distro there is.

Why bother updating? 'tworks right no?

Sure YOUR system may need updating because yet an other processor security disaster was discovered last week.
https://thenextweb.com/security/2019/08/06/researchers-discover-troubling-new-security-flaw-in-all-modern-intel-processors/
Not a day goes bye

https://www.reuters.com/article/us-intel-cyber/new-intel-security-flaws-could-slow-some-chips-by-nearly-20-idUSKCN1SK2OD

I KNOW intel procesors are part of the NSA spy network.
That is why I keep the invasion plans carved in stone^H^H^H^H^HMDISC hidden so they do not know the Orange House will be taken at noon.

That is why I run AMD so that sensitive data is not made public.

U Use Intel I Presume?


Huwei is releasing their own replacement for android shortly, thank Agent Orange's tariffs
I may just get one, I do not like android, this can only get better,
Dunno if it is Linux based...

<press power button to delete this text>

UUGH

 

[John Larkin]

On Sat, 10 Aug 2019 15:38:39 +0100, Tom Gardner
<spamjunk@blueyonder.co.uk> wrote:

On 10/08/19 14:47, John Larkin wrote:
On Sat, 10 Aug 2019 08:11:49 +0100, Tom Gardner
spamjunk@blueyonder.co.uk> wrote:

On 10/08/19 01:55, Winfield Hill wrote:
Plus, when you have an analytical
solution to your circuit, you can more easily see
what the trade offs are, and optimize the circuit.

That should be writ large, and engraved in all
university courses.

Unfortunately it is becoming a lost art

Learning classic circuit theory and analysis is critical to doing
original circuit design, but it's just the starting point.

Design is the opposite of analysis. And most all the interesting stuff
is seriously nonlinear.

What the theory can do is provide insight, guide creative fiddling.
Ultimately most of us solder parts to boards to make stuff that works,
not publish papers.

Completely understood and accepted. Anything significantly
non-linear virtually requires number crunching rather than
standard analysis.

Nonetheless, an analytical solution to a /simplified/ model
can yield valuable insights. The classic simplified model
in physics is exemplified by "...assume a spherical cow...".

There are many similar things in electronics, e.g. simple
model are used to estimate EMI/EMC between one comms system
and another. Imperfect? Of course; it never matches reality.
Useful? Yes.

Circuit design starts with the topology problem: what is the schematic
that we want to analyze? Where does it come from?

In my EE school, when I talked about designing things I was told
"Undergrads don't design; that starts in graduate school" so I didn't
apply for grad school.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Tim Williams]

"Tom Gardner" <spamjunk@blueyonder.co.uk> wrote in message
newsFA3F.982900$j33.789547@fx26.am4...

Nonetheless, an analytical solution to a /simplified/ model
can yield valuable insights. The classic simplified model
in physics is exemplified by "...assume a spherical cow...".

There are many similar things in electronics, e.g. simple
model are used to estimate EMI/EMC between one comms system
and another. Imperfect? Of course; it never matches reality.
Useful? Yes.

Don't read into it too deeply. JL is not so talented at analysis, and
forgets not to project his capability onto others.

The better lesson would be to work to your strengths, whatever they are; if
they are analytic, then work on that. If not, work on faster methods of
failing.

Note this traditional phrasing is as much derogatory as it is celebratory.
Indeed, some problems provably do not have closed analytical solutions, so
one of your better bets is simply cranking out attempts and picking the
better results. Thus, failing often.

Which, again, is no excuse to check ones' brain at the computer. Problems
can always be done smarter. Compare, say: the Leibniz formula for pi (the
alternating harmonic series, which sums to pi/4), which converges extremely
slowly; versus, the observation that the partial sums of that series
alternate around the final value, so, wouldn't it be great if we could stick
together pairs of terms with a properly tuned weighing factor and get much
closer to the true result in far fewer steps? Applying the Newton
convergence transformation gives a stupendous efficiency improvement
(thousands of terms for a few digits vs. about as many digits as terms
evaluated).

Is it better to run LTSpice for several hours, or sit with pad and pen for
several hours then check the result in ten minutes of simulation? If the
result is equivalent, it doesn't much matter for engineering purposes which
one did it. Limiting yourself to just one approach is the real problem.

Tim

--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Design
Website: https://www.seventransistorlabs.com/

 

[Tim Williams]

"Jan Panteltje" <pNaOnStPeAlMtje@yahoo.com> wrote in message
news:qimm3k$c6b$1@dont-email.me...

What security hole?
It '&' is just part of the bash syntax and means run program in the
background.

I made the not-unreasonable assumption that your text formatting woes are
due to a backwards, outdated system.

Linux has had hundreds of vulnerabilities over the years; you seem to take
offense at that, which is strange?

I mean, the first step towards security is acknowledging you have
vulnerabilities. The next step is securing them. You can't secure that
which you are in denial about.

Which in turn implies that you aren't actually interested in the security of
your machine, hence the strangeness.

Try reading a book on Unix, almost everything runs on Linux these days
except for some silly widows computers in homes of people
who have not bought / borrowed / stolen / rented / learned / what else
have you a clue.

Windows, excuse me? Where did that come in?

Oh, you must think I'm some sort of "fanboy" attacking your "camp" and so
you must rush to its defense, spouting off every concievable deflection
while continuing to ignore faults in your own system like being able to use
punctuation in a random filename?

Strange. Well, this part isn't strange at all, it's the same tribalism
that's tearing the world apart right now, really. More sad than strange.
:-\

Why do it to yerselves?

And no this system is lemme see
~ # uname -a
Linux panteltje12 2.6.37.6-smp #1 SMP Sat Aug 3 19:23:48 CEST 2013 i686
AMD Sempron(tm) 145 Processor AuthenticAMD GNU/Linux

Slackware, probably the most sane distro there is.

Why bother updating? 'tworks right no?

Ah, okay. So hopefully you've got most of these patched:
https://www.cvedetails.com/vulnerability-list/vendor_id-27/product_id-41/Slackware-Slackware-Linux.html

Waaait... 2.6? From 2011? But current version is 5.2.8 (stable)?! Then
more specific I guess would be this,
https://www.cvedetails.com/vulnerability-list.php?vendor_id=33&product_id=47&version_id=123871&page=1&hasexp=0&opdos=0&opec=0&opov=0&opcsrf=0&opgpriv=0&opsqli=0&opxss=0&opdirt=0&opmemc=0&ophttprs=0&opbyp=0&opfileinc=0&opginf=0&cvssscoremin=0&cvssscoremax=0&year=0&month=0&cweid=0&order=1&trc=214&sha=36e2264f1293f8fe45ef0e2c8ca35018397c6208
214 records.

Not that all of those appear to be critical RCE exploits, or are necessarily
in modules that you use regularly. I have no idea. But I mean, it's not
zero, which is all the point I'm after.

I must be misunderstanding something about Linux versioning or updating;
surely it would be strange for someone to be questioning my (so far
unstated) security practices, when they use a platform that hasn't been
updated in the better part of a decade?

And, like, it's entirely plausible that you've been watching those, and
patching your own kernel (and applications) locally. That would be cool.
But a big burden, and a big distraction from getting any work done.

"Recompiling the kernel" seems to have always been a frequent refrain from
the Linux community, which seems...bewildering to me? Do they expect
average users to know how to do that?

Sure YOUR system may need updating because yet an other processor security
disaster was discovered last week.
https://thenextweb.com/security/2019/08/06/researchers-discover-troubling-new-security-flaw-in-all-modern-intel-processors/
Not a day goes bye

https://www.reuters.com/article/us-intel-cyber/new-intel-security-flaws-could-slow-some-chips-by-nearly-20-idUSKCN1SK2OD

Yeah, some crazy stuff going on these days. Not to mention the still
present, and eminently hackable, Intel Management Engine hypervisor.

Good thing AMD doesn't have a Platform Security Proces--oh.

I KNOW intel procesors are part of the NSA spy network.
That is why I keep the invasion plans carved in stone^H^H^H^H^HMDISC
hidden so they do not know the Orange House will be taken at noon.

Uh?...

That is why I run AMD so that sensitive data is not made public.

U Use Intel I Presume?

What?

You again seem to be willing me into a side of a conflict I have no care
for, or stake in. I don't appreciate it.

Huwei is releasing their own replacement for android shortly, thank Agent
Orange's tariffs
I may just get one, I do not like android, this can only get better,
Dunno if it is Linux based...

press power button to delete this text

UUGH

Presumably Huwei will have "instrumentation" suitable for the Communist
Party of China's purposes. Probably not active monitoring, but a fairly
secure backdoor seems likely; the sort of thing our dear western leaders
like to keep pushing. They may have /some/ sense about it, like a unique
serial-numbered key per device, stored across isolated servers. I would
expect typical services to be general unlock, pushing hidden apps / updates
allowing targetted monitoring, etc.

This would be, I think, rather plausible and rather less
conspiracy-theoretic.

Better them than us; but things would be so much nicer if We Could All Just
Get Along(R). I guess we'll have to wait until Russia mass-pwns China's
citizens to say "toldyaso" to them, and to our own leaders who keep pushing
the same.

An interesting thing, as security goes: for all their problems, Apple has
been quite staunch in their rejection of such softening measures. Indeed
the gov't was so frustrated they even sued about it (and lost). It's not
often you see such a visible and explicit confirmation like that.

The NSA is certainly getting their tentacles on everything they can. All
the more reason to better review code, patch vulnerabilities as soon as they
are found, and keep systems updated. Use crypto early and often, and
responsibly.



For the record -- my desktop reports "AMD Phenom(tm) II X6 1055T Processor"
2.8GHz, and runs a 64-bit OS that is supported (for now..) and up to date.
So, not particularly new (about as old as yours, actually, and I think both
predating AMD's PSP?). I don't care for specsmanship, I don't run anything
that would benefit from more, and my graphics card mostly handles whatever
is left to do. And for that matter, the newest games I've played, haven't
even been 3D, what a waste it seems. Ah well.

Incidentally, *this* client (XP, OE6) is LAN attached only, and basically
just kept around for news only. Attackable definitely, but a limited
surface.

Tim

--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Design
Website: https://www.seventransistorlabs.com/

 

[Tom Gardner]

On 10/08/19 17:50, Tim Williams wrote:

"Tom Gardner" <spamjunk@blueyonder.co.uk> wrote in message
newsFA3F.982900$j33.789547@fx26.am4...
Nonetheless, an analytical solution to a /simplified/ model
can yield valuable insights. The classic simplified model
in physics is exemplified by "...assume a spherical cow...".

There are many similar things in electronics, e.g. simple
model are used to estimate EMI/EMC between one comms system
and another. Imperfect? Of course; it never matches reality.
Useful? Yes.

Don't read into it too deeply.

I don't

Your other points with which I wholeheartedly agree...

The better lesson would be to work to your strengths, whatever they are; if they
are analytic, then work on that.  If not, work on faster methods of failing.

Which, again, is no excuse to check ones' brain at the computer.  Problems can
always be done smarter.

Limiting yourself to just one approach is the real problem.

 

[Jan Panteltje]

On a sunny day (Sat, 10 Aug 2019 12:55:55 -0500) it happened "Tim Williams"
<tiwill@seventransistorlabs.com> wrote in <qin0f2$bk2$1@dont-email.me>:

For the record -- my desktop reports "AMD Phenom(tm) II X6 1055T Processor"
2.8GHz, and runs a 64-bit OS that is supported (for now..) and up to date.
So, not particularly new (about as old as yours, actually, and I think both
predating AMD's PSP?). I don't care for specsmanship, I don't run anything
that would benefit from more, and my graphics card mostly handles whatever
is left to do. And for that matter, the newest games I've played, haven't
even been 3D, what a waste it seems. Ah well.

Incidentally, *this* client (XP, OE6) is LAN attached only, and basically
just kept around for news only. Attackable definitely, but a limited
surface.

Tim

Na, I was not looking for a fight, thought you were.
I have been running Linux since 1998 or so (Soft Landing Systems Linux it was called, found it on some CD
that came with a magazine, and it replaced my win3.1 running on DRDOS with trumpet winsock.
Before that internet we had Viditel here, more bulletin board oriented
been 'online' since the eighties.


As I ran Free agent newsreader on that win 3.1 and there was no look alike Usenet newsreader for Linux,
I wrote NewsFleX (this one:
http://panteltje.com/panteltje/newsflex/index.html
still using it.

So and also wrote most other apps that run here, except the web browser, that is a moving target and a life long work
I have other interests.
On top of that I am and always have been : root
panteltje12: ~ # whoami
root

Ran my own servers back then until that became to much work
now website is at godaddy.
And godaddy has leased servers close to here in the Netherlands, so I am basically local
with that website.
Anyways tell me about the web, asked Hacktick to hack my servers, they tried, I watched and they failed.

So, those were wilder years and lots of things were hacked..
Having Linux and gcc was and still is cool.

But I have no illusions about security, WiF was hacked here so all is wired now.

I tried (because of somebody advertising it here) win what was it? Xp ? once
and it was removed really quick, new laptop came with some win version too,
replaced it with Ubuntu the same day, now 4 different versions of Linux in different partitions on it.
But this is my every day PC and very very hard to get into from the outside.
Of course you should always use brain when using the browser...
A very large part of the embedded world is running flavors of Linux, there are even satellites running it.
Talk about reliability.
~ # uptime
20:48:39 up 79 days, 8:47, 15 users, load average: 2.35, 3.27, 3.78
Last reboot was 79 days ago, had to change some power cables? do not remember,

For the rest I do not care who runs what, I wrote several OSes myself,
some multitasker too.
TCP stack for PIC, I can write what I want anytime in almost any language, except snake language 'python',
NOTHING in all those years that I wrote has been hacked,.
The hacked WiFi was a cheap Chinese security camera,,
So, to make it easier for NSA I always explain my plans here, like the invasion of the US tomorrow (? or was it later)
etc etc, and where I buy the plutonium and all that.
You gotta help those guys, else they die of boredom.

As you probably noticed I have not much to say about the subject, so I will leave it at this,
Reply at your own risk, that goes for using closed source software too, also from ratmond.

 

[whit3rd]

On Saturday, August 10, 2019 at 12:54:06 PM UTC-7, tabb...@gmail.com wrote:

On Saturday, 10 August 2019 18:55:51 UTC+1, Tim Williams wrote:

An interesting thing, as security goes: for all their problems, Apple has
been quite staunch in their rejection of such softening measures.

Was Apple's public stand for real, or was it showmanship when they complied behind closed doors? We'll probably never know.

Oh, we'll know. This kind of thing is a very temporary secret, at best.

Cisco sales (and stock) plummeted when their shipped products were found to have
(official US government) tampering, and Apple's iPhone sales are too important
to risk, so the public stand is probably also the private one.

Cisco's recent packaging is tamper-evident and more secure against
counterfeiting than most currency. It's a SERIOUS concern that Apple
and even Huawei must recognize, publicly and privately.

 

[John Larkin]

On Sat, 10 Aug 2019 11:50:08 -0500, "Tim Williams"
<tiwill@seventransistorlabs.com> wrote:

"Tom Gardner" <spamjunk@blueyonder.co.uk> wrote in message
newsFA3F.982900$j33.789547@fx26.am4...
Nonetheless, an analytical solution to a /simplified/ model
can yield valuable insights. The classic simplified model
in physics is exemplified by "...assume a spherical cow...".

There are many similar things in electronics, e.g. simple
model are used to estimate EMI/EMC between one comms system
and another. Imperfect? Of course; it never matches reality.
Useful? Yes.

Don't read into it too deeply. JL is not so talented at analysis, and
forgets not to project his capability onto others.

Not many people are talented at producing closed-form solutions to
nonlinear system response. Most interesting electronic systems are
nonlinear. What's a boy to do?

One big field where instinct+simulation usually (but not always) fails
is filter design. Things can get crazy and diverge fast.

I have managed up to 5th order, mostly luck. I've done some nice
one-side-absorptive lowpass filters by fiddling in Spice. The
criterion for absorbing cable reflections was fuzzy, what looks good.

"Modern filter design" is computer based anyhow. That's where the
tables in the books come from.


--

John Larkin Highland Technology, Inc trk

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[whit3rd]

On Friday, August 9, 2019 at 4:48:32 PM UTC-7, John Larkin wrote:

> Besides, I've forgotten most of that college math.

Math is universal. College is just a friendly place to learn some of it
Forgotten or not, math is omnipresent.

 

[John Larkin]

On Sat, 10 Aug 2019 15:59:51 +0100, Tom Gardner
<spamjunk@blueyonder.co.uk> wrote:

On 10/08/19 15:56, John Larkin wrote:
On Sat, 10 Aug 2019 15:38:39 +0100, Tom Gardner
spamjunk@blueyonder.co.uk> wrote:

On 10/08/19 14:47, John Larkin wrote:
On Sat, 10 Aug 2019 08:11:49 +0100, Tom Gardner
spamjunk@blueyonder.co.uk> wrote:

On 10/08/19 01:55, Winfield Hill wrote:
Plus, when you have an analytical
solution to your circuit, you can more easily see
what the trade offs are, and optimize the circuit.

That should be writ large, and engraved in all
university courses.

Unfortunately it is becoming a lost art

Learning classic circuit theory and analysis is critical to doing
original circuit design, but it's just the starting point.

Design is the opposite of analysis. And most all the interesting stuff
is seriously nonlinear.

What the theory can do is provide insight, guide creative fiddling.
Ultimately most of us solder parts to boards to make stuff that works,
not publish papers.

Completely understood and accepted. Anything significantly
non-linear virtually requires number crunching rather than
standard analysis.

Nonetheless, an analytical solution to a /simplified/ model
can yield valuable insights. The classic simplified model
in physics is exemplified by "...assume a spherical cow...".

There are many similar things in electronics, e.g. simple
model are used to estimate EMI/EMC between one comms system
and another. Imperfect? Of course; it never matches reality.
Useful? Yes.

Circuit design starts with the topology problem: what is the schematic
that we want to analyze? Where does it come from?

In my EE school, when I talked about designing things I was told
"Undergrads don't design; that starts in graduate school" so I didn't
apply for grad school.

Bizarre. Objectionable. My alma mater continues to have a
completely different tradition, 40 years later. They have
extremely well equipped labs, and the undergrads use them
at any time for any project, including personal projects.

Well, the place was pretty fuddy-duddy. Most of the EEs went into
power.


--

John Larkin Highland Technology, Inc trk

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

On Saturday, 10 August 2019 18:55:51 UTC+1, Tim Williams wrote:

An interesting thing, as security goes: for all their problems, Apple has
been quite staunch in their rejection of such softening measures. Indeed
the gov't was so frustrated they even sued about it (and lost). It's not
often you see such a visible and explicit confirmation like that.

The NSA is certainly getting their tentacles on everything they can. All
the more reason to better review code, patch vulnerabilities as soon as they
are found, and keep systems updated. Use crypto early and often, and
responsibly.

Was Apple's public stand for real, or was it showmanship when they complied behind closed doors? We'll probably never know.


NT

 

On Saturday, 10 August 2019 15:57:03 UTC+1, John Larkin wrote:

Circuit design starts with the topology problem: what is the schematic
that we want to analyze? Where does it come from?

In my EE school, when I talked about designing things I was told
"Undergrads don't design; that starts in graduate school" so I didn't
apply for grad school.

In fairness most undergrads don't design things. But if anyone wants to understand electronics at that age, they shoulda been designing things for years. Badly sometimes, how else does one learn.


NT

 

On Saturday, 10 August 2019 21:14:09 UTC+1, whit3rd wrote:

On Saturday, August 10, 2019 at 12:54:06 PM UTC-7, tabby wrote:
On Saturday, 10 August 2019 18:55:51 UTC+1, Tim Williams wrote:

An interesting thing, as security goes: for all their problems, Apple has
been quite staunch in their rejection of such softening measures.

Was Apple's public stand for real, or was it showmanship when they complied behind closed doors? We'll probably never know.

Oh, we'll know. This kind of thing is a very temporary secret, at best.

Cisco sales (and stock) plummeted when their shipped products were found to have
(official US government) tampering, and Apple's iPhone sales are too important
to risk, so the public stand is probably also the private one.

Cisco's recent packaging is tamper-evident and more secure against
counterfeiting than most currency. It's a SERIOUS concern that Apple
and even Huawei must recognize, publicly and privately.

point taken


NT

 

[Phil Hobbs]

On 8/10/19 4:01 PM, whit3rd wrote:

On Friday, August 9, 2019 at 4:48:32 PM UTC-7, John Larkin wrote:

Besides, I've forgotten most of that college math.

Math is universal. College is just a friendly place to learn some of it
Forgotten or not, math is omnipresent.

"Very deep. You should send that in to the Readers' Digest--they have a
page for people like you." --Ford Prefect




Cheers

Phil Hobbs

 

[Chris Jones]

On 10/08/2019 17:05, Phil Hobbs wrote:

On 8/9/19 4:50 PM, Winfield Hill wrote:
Here's a new section I'm hoping to complete, so it can be added to the
x-Chapter book before it goes to the printer in a few weeks. Please
look it over, but don't be too harsh, about its lack of mathematical
vigor.  It's closer to our usual back-of-the envelope approach to
calculations. Fixes for errors, suggestions for clarification,
improved accuracy, and comments welcome.

https://www.dropbox.com/s/7zl3yi789idg3s8/4x.26_Loop%20%26%20Nodal%20Analysis.pdf?dl=1



Nice.  I like your making a virtue out of a necessity (hand-drawn
figures).

One point that might be worth a footnote is that Kirchhoff's laws are a
low-frequency approximation, applicable only when radiation and
self-capacitance are negligible.

If your schematic is really complete, then I think that the laws apply
usefully, at least until the point where radiation is efficient. I am
assuming here that current through parasitic capacitances is counted
just as much as if it were a current flowing through a terminal of an
intentional capacitor. If, in your schematic and arithmetic, you leave
out things like the inductance and self-capacitance of wires, (and in
difficult cases, even the distributed capacitance at different points
along the inductance of wires), then of course the result of applying
Kirchoff's laws to the (incomplete) schematic won't predict the
behaviour of the actual construction. I suspect that radiation could
also be modelled in a way that allows Kirchoff's laws to be applied but
that the resulting schematic would be too complicated.

In making spice models for leadframes and bondwires, it makes a
significant difference whether or not one includes the self-capacitance
of the conductors (in addition to the mutual capacitances). Fastcap can
extract all of these. You can apportion the capacitances to different
points along the inductances of the conductors (extracted with
FastHenry). The resulting model is very useful when simulating packaged
chips, and of course the simulator at least attempts to satisfy
Kirchoff's laws.

 

[Phil Hobbs]

On 8/12/19 9:11 AM, Chris Jones wrote:

On 10/08/2019 17:05, Phil Hobbs wrote:
On 8/9/19 4:50 PM, Winfield Hill wrote:
Here's a new section I'm hoping to complete, so it can be added to
the x-Chapter book before it goes to the printer in a few weeks.
Please look it over, but don't be too harsh, about its lack of
mathematical vigor.  It's closer to our usual back-of-the envelope
approach to calculations. Fixes for errors, suggestions for
clarification, improved accuracy, and comments welcome.

https://www.dropbox.com/s/7zl3yi789idg3s8/4x.26_Loop%20%26%20Nodal%20Analysis.pdf?dl=1



Nice.  I like your making a virtue out of a necessity (hand-drawn
figures).

One point that might be worth a footnote is that Kirchhoff's laws are a
low-frequency approximation, applicable only when radiation and
self-capacitance are negligible.

If your schematic is really complete, then I think that the laws apply
usefully, at least until the point where radiation is efficient. I am
assuming here that current through parasitic capacitances is counted
just as much as if it were a current flowing through a terminal of an
intentional capacitor. If, in your schematic and arithmetic, you leave
out things like the inductance and self-capacitance of wires, (and in
difficult cases, even the distributed capacitance at different points
along the inductance of wires), then of course the result of applying
Kirchoff's laws to the (incomplete) schematic won't predict the
behaviour of the actual construction. I suspect that radiation could
also be modelled in a way that allows Kirchoff's laws to be applied but
that the resulting schematic would be too complicated.

Nope. Transmission lines at the schematic level are non-local, i.e. you
can't write a system of ODEs to describe a circuit with transmission
lines or significant radiation. Kirchhoff's laws are derived from
Maxwell's equations in the limit of low frequency (or alternatively, of
small size for a fixed frequency).

And if you have to model the circuit "in a way that allows Kirchhoff's
laws to be applied", you've implicitly admitted that they don't apply to
the actual circuit.

Don't get me wrong--K's equations are useful and all, but they have
limits. Being a physicist, I fully recognize the usefulness of sleazy
approximations, but you have to remember that that's what they are, or
you'll get snookered.

In making spice models for leadframes and bondwires, it makes a
significant difference whether or not one includes the self-capacitance
of the conductors (in addition to the mutual capacitances). Fastcap can
extract all of these. You can apportion the capacitances to different
points along the inductances of the conductors (extracted with
FastHenry). The resulting model is very useful when simulating packaged
chips, and of course the simulator at least attempts to satisfy
Kirchoff's laws.

Yeah, a lot of times you can patch up approximations by hand like
that--in my business one of the major examples is scalar optics, where
you treat the EM field as though it had only one component.
Polarization gets put in by hand once the calculation is done. Not the
absolute cleanest procedure conceptually, but it works great for almost
everything.

Another example is geometric optics, where you keep going back and forth
between plane waves and rays, i.e. between infinitely broad wavefronts
and infinitely narrow ones. When a ray encounters a curved dielectric
surface, you assume that it bounced off the tangent plane at the point
of incidence. That gives you the surface normal. Then you switch to
the plane wave picture and apply Snell's law and the Fresnel formulae to
get the direction and amplitude of the refracted ray. Both Snell and
Fresnel depend on the wave picture, specifically phase matching at the
boundary (Snell) and continuity of tangential E and perpendicular D
(Fresnel). The derivation of Snell's law requires translational
invariance, i.e. it only works on a flat boundary.

So to have well-founded confidence in our tools, we have to know where
they break down and how to patch them up.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
https://hobbs-eo.com

 

[George Herold]

On Monday, August 12, 2019 at 10:02:48 AM UTC-4, Phil Hobbs wrote:

On 8/12/19 9:11 AM, Chris Jones wrote:
On 10/08/2019 17:05, Phil Hobbs wrote:
On 8/9/19 4:50 PM, Winfield Hill wrote:
Here's a new section I'm hoping to complete, so it can be added to
the x-Chapter book before it goes to the printer in a few weeks.
Please look it over, but don't be too harsh, about its lack of
mathematical vigor.  It's closer to our usual back-of-the envelope
approach to calculations. Fixes for errors, suggestions for
clarification, improved accuracy, and comments welcome.

https://www.dropbox.com/s/7zl3yi789idg3s8/4x.26_Loop%20%26%20Nodal%20Analysis.pdf?dl=1



Nice.  I like your making a virtue out of a necessity (hand-drawn
figures).

One point that might be worth a footnote is that Kirchhoff's laws are a
low-frequency approximation, applicable only when radiation and
self-capacitance are negligible.

If your schematic is really complete, then I think that the laws apply
usefully, at least until the point where radiation is efficient. I am
assuming here that current through parasitic capacitances is counted
just as much as if it were a current flowing through a terminal of an
intentional capacitor. If, in your schematic and arithmetic, you leave
out things like the inductance and self-capacitance of wires, (and in
difficult cases, even the distributed capacitance at different points
along the inductance of wires), then of course the result of applying
Kirchoff's laws to the (incomplete) schematic won't predict the
behaviour of the actual construction. I suspect that radiation could
also be modelled in a way that allows Kirchoff's laws to be applied but
that the resulting schematic would be too complicated.

Nope. Transmission lines at the schematic level are non-local, i.e. you
can't write a system of ODEs to describe a circuit with transmission
lines or significant radiation. Kirchhoff's laws are derived from
Maxwell's equations in the limit of low frequency (or alternatively, of
small size for a fixed frequency).

Hmm Phil, to put this in my own words... and please correct me if I'm wrong...
or I'm only part right. I think a limitation of K's laws is that they
treat V and I as instantaneously the same everywhere.

George H.
(every theory is an approximation at some level)

And if you have to model the circuit "in a way that allows Kirchhoff's
laws to be applied", you've implicitly admitted that they don't apply to
the actual circuit.

Don't get me wrong--K's equations are useful and all, but they have
limits. Being a physicist, I fully recognize the usefulness of sleazy
approximations, but you have to remember that that's what they are, or
you'll get snookered.


In making spice models for leadframes and bondwires, it makes a
significant difference whether or not one includes the self-capacitance
of the conductors (in addition to the mutual capacitances). Fastcap can
extract all of these. You can apportion the capacitances to different
points along the inductances of the conductors (extracted with
FastHenry). The resulting model is very useful when simulating packaged
chips, and of course the simulator at least attempts to satisfy
Kirchoff's laws.

Yeah, a lot of times you can patch up approximations by hand like
that--in my business one of the major examples is scalar optics, where
you treat the EM field as though it had only one component.
Polarization gets put in by hand once the calculation is done. Not the
absolute cleanest procedure conceptually, but it works great for almost
everything.

Another example is geometric optics, where you keep going back and forth
between plane waves and rays, i.e. between infinitely broad wavefronts
and infinitely narrow ones. When a ray encounters a curved dielectric
surface, you assume that it bounced off the tangent plane at the point
of incidence. That gives you the surface normal. Then you switch to
the plane wave picture and apply Snell's law and the Fresnel formulae to
get the direction and amplitude of the refracted ray. Both Snell and
Fresnel depend on the wave picture, specifically phase matching at the
boundary (Snell) and continuity of tangential E and perpendicular D
(Fresnel). The derivation of Snell's law requires translational
invariance, i.e. it only works on a flat boundary.

So to have well-founded confidence in our tools, we have to know where
they break down and how to patch them up.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
https://hobbs-eo.com

 

[Phil Hobbs]

On 8/12/19 12:05 PM, George Herold wrote:

On Monday, August 12, 2019 at 10:02:48 AM UTC-4, Phil Hobbs wrote:
On 8/12/19 9:11 AM, Chris Jones wrote:
On 10/08/2019 17:05, Phil Hobbs wrote:
On 8/9/19 4:50 PM, Winfield Hill wrote:
Here's a new section I'm hoping to complete, so it can be added to
the x-Chapter book before it goes to the printer in a few weeks.
Please look it over, but don't be too harsh, about its lack of
mathematical vigor.  It's closer to our usual back-of-the envelope
approach to calculations. Fixes for errors, suggestions for
clarification, improved accuracy, and comments welcome.

https://www.dropbox.com/s/7zl3yi789idg3s8/4x.26_Loop%20%26%20Nodal%20Analysis.pdf?dl=1



Nice.  I like your making a virtue out of a necessity (hand-drawn
figures).

One point that might be worth a footnote is that Kirchhoff's laws are a
low-frequency approximation, applicable only when radiation and
self-capacitance are negligible.

If your schematic is really complete, then I think that the laws apply
usefully, at least until the point where radiation is efficient. I am
assuming here that current through parasitic capacitances is counted
just as much as if it were a current flowing through a terminal of an
intentional capacitor. If, in your schematic and arithmetic, you leave
out things like the inductance and self-capacitance of wires, (and in
difficult cases, even the distributed capacitance at different points
along the inductance of wires), then of course the result of applying
Kirchoff's laws to the (incomplete) schematic won't predict the
behaviour of the actual construction. I suspect that radiation could
also be modelled in a way that allows Kirchoff's laws to be applied but
that the resulting schematic would be too complicated.

Nope. Transmission lines at the schematic level are non-local, i.e. you
can't write a system of ODEs to describe a circuit with transmission
lines or significant radiation. Kirchhoff's laws are derived from
Maxwell's equations in the limit of low frequency (or alternatively, of
small size for a fixed frequency).

Hmm Phil, to put this in my own words... and please correct me if I'm wrong..
or I'm only part right. I think a limitation of K's laws is that they
treat V and I as instantaneously the same everywhere.

George H.
(every theory is an approximation at some level)

Everywhere on a given circuit node or loop, right. Anything with
transmission-line behaviour can't be modelled as an ODE--the fields
inside the T-line can be modelled with PDEs (Maxwell), but circuits are
all ODEs. The T-line has invisible internal state, so its circuit
behaviour is nonlocal.

Antennas have voltage nodes and antinodes on the same wire, so that K's
voltage law doesn't hold, and they have current nodes and antinodes as
well, so neither does the current law.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
https://hobbs-eo.com

 

[Chris Jones]

On 13/08/2019 00:02, Phil Hobbs wrote:

On 8/12/19 9:11 AM, Chris Jones wrote:
On 10/08/2019 17:05, Phil Hobbs wrote:
On 8/9/19 4:50 PM, Winfield Hill wrote:
Here's a new section I'm hoping to complete, so it can be added to
the x-Chapter book before it goes to the printer in a few weeks.
Please look it over, but don't be too harsh, about its lack of
mathematical vigor.  It's closer to our usual back-of-the envelope
approach to calculations. Fixes for errors, suggestions for
clarification, improved accuracy, and comments welcome.

https://www.dropbox.com/s/7zl3yi789idg3s8/4x.26_Loop%20%26%20Nodal%20Analysis.pdf?dl=1



Nice.  I like your making a virtue out of a necessity (hand-drawn
figures).

One point that might be worth a footnote is that Kirchhoff's laws are a
low-frequency approximation, applicable only when radiation and
self-capacitance are negligible.

If your schematic is really complete, then I think that the laws apply
usefully, at least until the point where radiation is efficient. I am
assuming here that current through parasitic capacitances is counted
just as much as if it were a current flowing through a terminal of an
intentional capacitor. If, in your schematic and arithmetic, you leave
out things like the inductance and self-capacitance of wires, (and in
difficult cases, even the distributed capacitance at different points
along the inductance of wires), then of course the result of applying
Kirchoff's laws to the (incomplete) schematic won't predict the
behaviour of the actual construction. I suspect that radiation could
also be modelled in a way that allows Kirchoff's laws to be applied but
that the resulting schematic would be too complicated.

Nope. Transmission lines at the schematic level are non-local, i.e. you
can't write a system of ODEs to describe a circuit with transmission
lines or significant radiation. Kirchhoff's laws are derived from
Maxwell's equations in the limit of low frequency (or alternatively, of
small size for a fixed frequency).

I would say that that a wire on a schematic is not a valid
representation of a transmission line, and if necessary I would
approximate a transmission line as a ladder of (ideally infinitely) many
series inductors and shunt capacitors. Of course very many components
are required for this to be reasonably accurate.

At frequencies where the number of required components is excessive, I
would then say that a schematic is not a good way to describe the
physical system.

And if you have to model the circuit "in a way that allows Kirchhoff's
laws to be applied", you've implicitly admitted that they don't apply to
the actual circuit.

If by actual circuit we mean the physical object, then really I only
expect Maxwell's equations to describe it, and I'm not very good at
solving those. In a completely general sense I'm not even sure how one
would try to apply Kirchoff's laws to an arbitrary three dimensional
piece of electronics.

Don't get me wrong--K's equations are useful and all, but they have
limits. Being a physicist, I fully recognize the usefulness of sleazy
approximations, but you have to remember that that's what they are, or
you'll get snookered.
Agreed.

I guess I might have a rather unusual idea of what a schematic is, and
this might be what causes me to take issue with what you said. To me, a
schematic ought to be something that, when simulated (by some ideal
simulator!), applying Kirchoff's laws, Ohm's law, i=C.dv/dt and so on,
would sufficiently accurately predict the behavoiur of the real system.

To me, if the predictions are wrong, then I blame the schematic as being
an inaccurate representation of the system, rather than blaming the
equations used to simulate the behaviour of the schematic. Perhaps my
philosophy on this topic comes from having had the job of making a
schematic (sometimes pulling in netlists from field solvers) in order to
simulate my design as implemented in a physical product. There was an
expectation that I would use a circuit simulator provided to me, that
did try to apply Kirchoff's laws (though not perfectly in the case of
KCL). I did at least have the luxury that the physical dimensions of the
system were a tiny fraction of a wavelength.

 

On Wednesday, 14 August 2019 13:00:54 UTC+1, Chris Jones wrote:

I would say that that a wire on a schematic is not a valid
representation of a transmission line, and if necessary I would
approximate a transmission line as a ladder of (ideally infinitely) many
series inductors and shunt capacitors. Of course very many components
are required for this to be reasonably accurate.

At frequencies where the number of required components is excessive, I
would then say that a schematic is not a good way to describe the
physical system.

draw the schematic more realistically, with the line as a long inductor with C all along it

I guess I might have a rather unusual idea of what a schematic is, and
this might be what causes me to take issue with what you said. To me, a
schematic ought to be something that, when simulated (by some ideal
simulator!), applying Kirchoff's laws, Ohm's law, i=C.dv/dt and so on,
would sufficiently accurately predict the behavoiur of the real system.

To me, if the predictions are wrong, then I blame the schematic as being
an inaccurate representation of the system, rather than blaming the
equations used to simulate the behaviour of the schematic. Perhaps my
philosophy on this topic comes from having had the job of making a
schematic (sometimes pulling in netlists from field solvers) in order to
simulate my design as implemented in a physical product. There was an
expectation that I would use a circuit simulator provided to me, that
did try to apply Kirchoff's laws (though not perfectly in the case of
KCL). I did at least have the luxury that the physical dimensions of the
system were a tiny fraction of a wavelength.

Really it's the sim that's wrong, but you can tackle its shortcomings by adding in parts to the sim's input so it sims more feaures of the real circuit.


NT