EDAboard.com | EDAboard.eu | EDAboard.de | EDAboard.co.uk | RTV forum PL | NewsGroups PL

good book, slightly on-topic

Ask a question - edaboard.com

elektroda.net NewsGroups Forum Index - Electronics Design - good book, slightly on-topic

Goto page Previous  1, 2, 3  Next

Steve Wilson
Guest

Thu Dec 29, 2016 4:05 am   



krw_at_notreal.com wrote:

Quote:
On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the
market at half price?

It's hard to cut the cost of service in half.


As I understand, his cost of service is close to zero. The Chinese can
easily match that.

Quote:
Why do you feel they will not do so in the future?

His business isn't copying others.


Non sequitur.


Guest

Thu Dec 29, 2016 4:08 am   



On Thursday, December 29, 2016 at 7:45:33 AM UTC+11, k...@notreal.com wrote:
Quote:
On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the market at
half price?


Probably because John sells bespoke electronics for a niche market which isn't all that large.

Quote:
It's hard to cut the cost of service in half.

Why do you feel they will not do so in the future?

His business isn't copying others.


It's more copying himself with minor variations. He has said here that his average development time is about a fortnight.

--
Bill Sloman, Sydney


Guest

Thu Dec 29, 2016 4:14 am   



On Wed, 28 Dec 2016 21:05:26 GMT, Steve Wilson <no_at_spam.com> wrote:

Quote:
krw_at_notreal.com wrote:

On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the
market at half price?

It's hard to cut the cost of service in half.

As I understand, his cost of service is close to zero. The Chinese can
easily match that.


You obviously don't understand much.
Quote:

Why do you feel they will not do so in the future?

His business isn't copying others.

Non sequitur.


You obviously don't understand much.

John Larkin
Guest

Thu Dec 29, 2016 6:26 am   



On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

Quote:
John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the market at
half price?


Maybe because the market is too small. Maybe because they couldn't
market them. Maybe because they don't understand them.

Quote:

Why do you feel they will not do so in the future?


They haven't so far. We see a little competition for scientific and
aerospace instrumentation from Europe, a tiny bit from Japan,
basically none from anywhere else.

You would think that the Chinese (and Brazilian) space and aircraft
industries would need instrumentation. I don't know where they get it
from.





--

John Larkin Highland Technology, Inc

lunatic fringe electronics

John Larkin
Guest

Thu Dec 29, 2016 7:45 am   



On Tue, 27 Dec 2016 13:24:22 -0700, Don Y
<blockedofcourse_at_foo.invalid> wrote:

Quote:
On 12/27/2016 1:19 PM, Jim Thompson wrote:
On Tue, 27 Dec 2016 13:14:51 -0700, Don Y
blockedofcourse_at_foo.invalid> wrote:

On 12/27/2016 9:47 AM, John Larkin wrote:
The other is how the policy was born. This was in the heyday of cruel
Malthusian stupidity like Ehrlich's "The Population Bomb" book and
Mao's Cultural Revolution. Mao emptied the universities and industries
of all the smart people and sent them out to farm millet or something.
But he wanted ICBMs, so the only scientific institution untouched was
the rocket scientists. Some of them decided to do social engineering
and wrote some population simulation models to run on their primitive
computers. They decided that the only way to prevent population
catastrophe and widespread starvation was one-child; Mao was doing a
pretty good job on the starvation thing already. They ignored the few
social science people who objected, because computer models are
obviously perfect.

Yes, and anyone who would rely on models to, for example, predict
tomorrow's weather or design an electronic circuit would be a fool...

"Yes, and anyone who would rely on models to, for example, [...]
design an electronic circuit would be a fool..."

Really? Then how do you propose to design a multi-thousand device
chip?

My point was that the failure of one alleged model (population control) doesn't
mean that ALL uses of computer models are flawed. My choice of
examples was very deliberate. :


I never calimed that.

Quote:

(how do you think the folks designing the processes work, entirely on empirical
observation? or the materials used in those processes? etc. :> )


Have you ever written a dynamic system simulation yourself, real code
as opposed to running some canned product? I have, many times.

When coding such a system, it's reasonable to assume an error in every
10 or 20 lines of code. So one writes the code and tests it against
some known solution, and iterate/debug until they agree. The test case
might be an analytical solution, some classic example, or a physical
experiment. We debug until we think there are no bugs.

Lacking some test mechanism, the simulation code will probably be
wrong. For a large-scale energetic nonlinear chaotic system, like
population or climate, there may be no way to test the simulation code
or the input parameters. So it will almost always be wrong.
Hindcasting doesn't help much.

Even LT Spice sometimes does goofy things. What most people do, when
it is erratic, is fiddle with time steps or solver options or parts
values until we get something that we *think* looks closer to reality,
in other words tells us what we want to hear. But maybe we just made
it worse.



--

John Larkin Highland Technology, Inc

lunatic fringe electronics

Don Y
Guest

Thu Dec 29, 2016 8:30 am   



On 12/28/2016 5:45 PM, John Larkin wrote:
Quote:
On Tue, 27 Dec 2016 13:24:22 -0700, Don Y
blockedofcourse_at_foo.invalid> wrote:

On 12/27/2016 1:19 PM, Jim Thompson wrote:
On Tue, 27 Dec 2016 13:14:51 -0700, Don Y
blockedofcourse_at_foo.invalid> wrote:

On 12/27/2016 9:47 AM, John Larkin wrote:
The other is how the policy was born. This was in the heyday of cruel
Malthusian stupidity like Ehrlich's "The Population Bomb" book and
Mao's Cultural Revolution. Mao emptied the universities and industries
of all the smart people and sent them out to farm millet or something.
But he wanted ICBMs, so the only scientific institution untouched was
the rocket scientists. Some of them decided to do social engineering
and wrote some population simulation models to run on their primitive
computers. They decided that the only way to prevent population
catastrophe and widespread starvation was one-child; Mao was doing a
pretty good job on the starvation thing already. They ignored the few
social science people who objected, because computer models are
obviously perfect.

Yes, and anyone who would rely on models to, for example, predict
tomorrow's weather or design an electronic circuit would be a fool...

"Yes, and anyone who would rely on models to, for example, [...]
design an electronic circuit would be a fool..."

Really? Then how do you propose to design a multi-thousand device
chip?

My point was that the failure of one alleged model (population control) doesn't
mean that ALL uses of computer models are flawed. My choice of
examples was very deliberate. :

I never calimed that.


By the same token, did *I* claim you did?? (that's the trouble with
innuendo -- you never know *intent*! :> )

Quote:
(how do you think the folks designing the processes work, entirely on empirical
observation? or the materials used in those processes? etc. :> )

Have you ever written a dynamic system simulation yourself, real code
as opposed to running some canned product? I have, many times.


We build models every day. We model how we expect our users to interact
with our devices. We model how we expect our devices to behave. We model
how often they will need calibration/repair/replacement. etc.

Quote:
When coding such a system, it's reasonable to assume an error in every
10 or 20 lines of code. So one writes the code and tests it against
some known solution, and iterate/debug until they agree. The test case
might be an analytical solution, some classic example, or a physical
experiment. We debug until we think there are no bugs.


The *first* step in building a model is to have some understanding of
the process/phenomenon/etc. that you are trying to model. This can
either be intuitive or analytic. Can be something you derive or that
a heuristic algorithm SUGGESTS to you (based on *its* observations
of the process/phenomenon/etc.).

When I was in high school, I designed and built a predictive "setback
thermostat": tell it WHEN you want the house at a particular
temperature instead of when to CHANGE the setpoint and it sorts out
ow and when to alter the setpoint. Simple: house's thermal mass
is a capacitor; furnace is a voltage source; radiators have some
thermal resistance to them limiting the rate the house can heat up;
house itself is shunted by a "loss element". I felt I understood
the way a house reacted to its heating system and environment.

Worked great (hot-water-heat has a lot of lag so getting the setpoint
adjusted early is much more comfortable than having to wait for the
house to catch up to your wishes).

Except at "certain times". Didn't occur to me in my initial model that
the outside temperature would have such a big impact on the loss
characteristics. And, once I realized that, didn't realize how much
wind speed (and direction!) would influence the mess. Or, cloud
cover curtailing radiant losses overnight. Or...

"Gee, this is a lot harder than it seemed! No wonder why
thermostat vendors just offer devices that change the setpoint
*AT* a particular time and don't bother trying to optimize for
these OBVIOUS issues!"

OTOH, my current "HVAC model" takes into consideration all of these
things -- and more. Do I expect it to "learn" overnight? No.
OTOH, it's sitting there 24/7/365 with nothing else to do so it
costs very little for it to watch continuously and keep tweeking its
model.

Recently discovered a gaping hole in said model: it has no way of knowing
when I've opened doors/windows to benefit from exterior conditions
(air the house out, get a cool/warm breeze flowing through, etc.).
So, my reliance on the model to determine when the HVAC plant has failed
(furnace hasn't shut off evern when commanded to do so) doesn't work;
the house can continue to heat up even with the furnace verified as OFF
(if the air blowing into the house is warmer than the house interior!)

There are no bugs in the *code* (I am "sufficiently skilled in the art")
but, rather, in the model itself.

Quote:
Lacking some test mechanism, the simulation code will probably be
wrong. For a large-scale energetic nonlinear chaotic system, like
population or climate, there may be no way to test the simulation code
or the input parameters. So it will almost always be wrong.
Hindcasting doesn't help much.


Really? The weather forecast here is almost always "spot on".
Even to the extent that it predicts wind speeds ON AN HOURLY BASIS
(useful if you are painting /en plein air/. I find this level of
accuracy amusing as the weather at our end of the street is
significantly different than the other end -- yet the "citizen
weather stations" allow the forecast to be tweeked to address
those variations (of course, it never SNOWS on one end of the street
while baking under a hot sun on the other!)

We rely on it to determine when to protect our citrus trees:
freezing (32F) requires no extra measures; OTOH, 28F means
we will incur losses. And, 30 minutes at 28F is considerably
different from 4 hours at 28F (the forecast gives us that
level of detail with a high degree of confidence).

Quote:
Even LT Spice sometimes does goofy things. What most people do, when
it is erratic, is fiddle with time steps or solver options or parts
values until we get something that we *think* looks closer to reality,
in other words tells us what we want to hear. But maybe we just made
it worse.


Steve Wilson
Guest

Thu Dec 29, 2016 8:30 am   



John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Quote:
Even LT Spice sometimes does goofy things. What most people do, when
it is erratic, is fiddle with time steps or solver options or parts
values until we get something that we *think* looks closer to reality,
in other words tells us what we want to hear. But maybe we just made
it worse.


I have been using SPICE since the DOS days. It is excellent at displaying
ideal situations without circuit parasitics such as crosstalk, ground
bounce, circuit noise, EMI and RFI problems, thermal effects, degraded
components, variations in components, and all the other ills that plague
circuit design. It is excellent at showing you how a circuit actually
works, so you can understand what you are really seeing when viewing a
noisy trace on an oscilloscope.

It is extremely good at developing new circuit ideas since you don't have
to waste time soldering and unsoldering various components to test a new
idea. You can view any parameter in a circuit that you wish, such as the
base current in a Colpitts oscillator. You could never do this in
hardware.

You cannot destroy a circuit by operating it beyond component limits. An
actual circuit may be difficult to troublehoot if it only lasts a
microsecond past power on.

But you have to be careful about your models. The simulation is only as
good as the model you supply. I notice you have made a number of false
assumptions about the capabilities of SPICE, based on models that fail
basic mathematical analysis. SPICE is much better than you think, and you
have used the failure of your models to claim that much larger
simulations are also wrong.

Witness the success of designing airplanes. These are obviously multi-
billion dollar projects, and simulating the airflow and dynamic
performance is essential to the success of the project. When the
prototype takes off on its first flight, it is highly instrumented and
they can measure just about every important flight parameter. If you
watch the Youtube videos of the flight analysis, they show the result of
dynamic perturbations match the simulations almost exactly.

So the basic approach of fiddling until you "get something that we
*think* looks closer to reality" is wrong. I have been surprised many
times by finding that a circuit performs much different from what I
expected, and this has led to new appreciations of circuit design and
performance.

Just understand your models and their limitations, and you can build on
the new knowledge of a good simulation.


Guest

Thu Dec 29, 2016 8:30 am   



On Thu, 29 Dec 2016 03:22:16 GMT, Steve Wilson <no_at_spam.com> wrote:

Quote:
krw_at_notreal.com wrote:

On Wed, 28 Dec 2016 21:05:26 GMT, Steve Wilson <no_at_spam.com> wrote:

krw_at_notreal.com wrote:

On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the
market at half price?

It's hard to cut the cost of service in half.

As I understand, his cost of service is close to zero. The Chinese can
easily match that.

You obviously don't understand much.

Why do you feel they will not do so in the future?

His business isn't copying others.

Non sequitur.

You obviously don't understand much.

You waste my time. PLONK


You obviously don't understand much. You've proven it.

Steve Wilson
Guest

Thu Dec 29, 2016 8:30 am   



John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Quote:
On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the
market at half price?


Maybe because the market is too small. Maybe because they couldn't
market them. Maybe because they don't understand them.


Why do you feel they will not do so in the future?

They haven't so far. We see a little competition for scientific and
aerospace instrumentation from Europe, a tiny bit from Japan,
basically none from anywhere else.

You would think that the Chinese (and Brazilian) space and aircraft
industries would need instrumentation. I don't know where they get it
from.


Very good answers. Thank you.

John Larkin
Guest

Thu Dec 29, 2016 8:30 am   



On Wed, 28 Dec 2016 19:54:52 -0700, Don Y
<blockedofcourse_at_foo.invalid> wrote:

Quote:
On 12/28/2016 5:45 PM, John Larkin wrote:
On Tue, 27 Dec 2016 13:24:22 -0700, Don Y
blockedofcourse_at_foo.invalid> wrote:

On 12/27/2016 1:19 PM, Jim Thompson wrote:
On Tue, 27 Dec 2016 13:14:51 -0700, Don Y
blockedofcourse_at_foo.invalid> wrote:

On 12/27/2016 9:47 AM, John Larkin wrote:
The other is how the policy was born. This was in the heyday of cruel
Malthusian stupidity like Ehrlich's "The Population Bomb" book and
Mao's Cultural Revolution. Mao emptied the universities and industries
of all the smart people and sent them out to farm millet or something.
But he wanted ICBMs, so the only scientific institution untouched was
the rocket scientists. Some of them decided to do social engineering
and wrote some population simulation models to run on their primitive
computers. They decided that the only way to prevent population
catastrophe and widespread starvation was one-child; Mao was doing a
pretty good job on the starvation thing already. They ignored the few
social science people who objected, because computer models are
obviously perfect.

Yes, and anyone who would rely on models to, for example, predict
tomorrow's weather or design an electronic circuit would be a fool...

"Yes, and anyone who would rely on models to, for example, [...]
design an electronic circuit would be a fool..."

Really? Then how do you propose to design a multi-thousand device
chip?

My point was that the failure of one alleged model (population control) doesn't
mean that ALL uses of computer models are flawed. My choice of
examples was very deliberate. :

I never calimed that.

By the same token, did *I* claim you did?? (that's the trouble with
innuendo -- you never know *intent*! :> )

(how do you think the folks designing the processes work, entirely on empirical
observation? or the materials used in those processes? etc. :> )

Have you ever written a dynamic system simulation yourself, real code
as opposed to running some canned product? I have, many times.

We build models every day. We model how we expect our users to interact
with our devices. We model how we expect our devices to behave. We model
how often they will need calibration/repair/replacement. etc.

When coding such a system, it's reasonable to assume an error in every
10 or 20 lines of code. So one writes the code and tests it against
some known solution, and iterate/debug until they agree. The test case
might be an analytical solution, some classic example, or a physical
experiment. We debug until we think there are no bugs.

The *first* step in building a model is to have some understanding of
the process/phenomenon/etc. that you are trying to model. This can
either be intuitive or analytic. Can be something you derive or that
a heuristic algorithm SUGGESTS to you (based on *its* observations
of the process/phenomenon/etc.).

When I was in high school, I designed and built a predictive "setback
thermostat": tell it WHEN you want the house at a particular
temperature instead of when to CHANGE the setpoint and it sorts out
ow and when to alter the setpoint. Simple: house's thermal mass
is a capacitor; furnace is a voltage source; radiators have some
thermal resistance to them limiting the rate the house can heat up;
house itself is shunted by a "loss element". I felt I understood
the way a house reacted to its heating system and environment.

Worked great (hot-water-heat has a lot of lag so getting the setpoint
adjusted early is much more comfortable than having to wait for the
house to catch up to your wishes).

Except at "certain times". Didn't occur to me in my initial model that
the outside temperature would have such a big impact on the loss
characteristics. And, once I realized that, didn't realize how much
wind speed (and direction!) would influence the mess. Or, cloud
cover curtailing radiant losses overnight. Or...

"Gee, this is a lot harder than it seemed! No wonder why
thermostat vendors just offer devices that change the setpoint
*AT* a particular time and don't bother trying to optimize for
these OBVIOUS issues!"

OTOH, my current "HVAC model" takes into consideration all of these
things -- and more. Do I expect it to "learn" overnight? No.
OTOH, it's sitting there 24/7/365 with nothing else to do so it
costs very little for it to watch continuously and keep tweeking its
model.

Recently discovered a gaping hole in said model: it has no way of knowing
when I've opened doors/windows to benefit from exterior conditions
(air the house out, get a cool/warm breeze flowing through, etc.).
So, my reliance on the model to determine when the HVAC plant has failed
(furnace hasn't shut off evern when commanded to do so) doesn't work;
the house can continue to heat up even with the furnace verified as OFF
(if the air blowing into the house is warmer than the house interior!)

There are no bugs in the *code* (I am "sufficiently skilled in the art")
but, rather, in the model itself.

Lacking some test mechanism, the simulation code will probably be
wrong. For a large-scale energetic nonlinear chaotic system, like
population or climate, there may be no way to test the simulation code
or the input parameters. So it will almost always be wrong.
Hindcasting doesn't help much.

Really? The weather forecast here is almost always "spot on".
Even to the extent that it predicts wind speeds ON AN HOURLY BASIS
(useful if you are painting /en plein air/. I find this level of
accuracy amusing as the weather at our end of the street is
significantly different than the other end -- yet the "citizen
weather stations" allow the forecast to be tweeked to address
those variations (of course, it never SNOWS on one end of the street
while baking under a hot sun on the other!)

We rely on it to determine when to protect our citrus trees:
freezing (32F) requires no extra measures; OTOH, 28F means
we will incur losses. And, 30 minutes at 28F is considerably
different from 4 hours at 28F (the forecast gives us that
level of detail with a high degree of confidence).


The weater forcasts here on the west coast are terrible. Last week, in
the mountains, we had a "shelter in place" blizzard warning, really
dire stuff. Three hours after the warning was still active, it was
clear and sunny. The storm never showed up. Same thing happens a lot
in San Francisco. The wind patterns along the coast are erratic, so
yesterday's weekly forcast often looks nothing like today's weekly
forcast.

Long-term forcasts, like weeks ahead, are useless, except there is an
average seasonal trend. If I predict it won't rain on June 22, I'll
probably be right. Any year.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

Steve Wilson
Guest

Thu Dec 29, 2016 8:30 am   



krw_at_notreal.com wrote:

Quote:
On Wed, 28 Dec 2016 21:05:26 GMT, Steve Wilson <no_at_spam.com> wrote:

krw_at_notreal.com wrote:

On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the
market at half price?

It's hard to cut the cost of service in half.

As I understand, his cost of service is close to zero. The Chinese can
easily match that.

You obviously don't understand much.

Why do you feel they will not do so in the future?

His business isn't copying others.

Non sequitur.

You obviously don't understand much.


You waste my time. PLONK

Dimitrij Klingbeil
Guest

Thu Dec 29, 2016 8:30 am   



On 29.12.2016 00:26, John Larkin wrote:
Quote:

You would think that the Chinese (and Brazilian) space and aircraft
industries would need instrumentation. I don't know where they get it
from.


Well, it looks like they have their own local instrument developers.

The CETC (China Electronics Technology Corporation) has a subsidiary
that produces RF and microwave instruments ("The 41st Institute").

If one looks at their website (www.ei41.com), right on the main page
they show a quite nice 3 Hz to 50 GHz spectrum analyzer (AV4051H),
with lots of signal analysis and demodulation capabilities, that is
expandable with external transverters up to 325 GHz.

With these capabilities, they should be able to come up with decent
instruments of other types and for other uses too.

However I'm not sure if equipment of this sort is even exported. They
surely have their own version of ITAR and various national security
related export restrictions. Plus if you somehow manage to get one
anyway, and by some carelessness happen to fry its input mixer or
something, any attempt to send it back to the manufacturer for repair
will most likely put you in jail, thanks to your own country's rules.

Tom Gardner
Guest

Thu Dec 29, 2016 4:11 pm   



On 29/12/16 02:54, Don Y wrote:
> Really? The weather forecast here is almost always "spot on".

That's because you don't have "weather", you have
"climate with occasional exceptions" :)

Live near/under the jet stream and you will have
weather to the extent that sometimes the forecast
is inaccurate after <6 hours.

Tom Gardner
Guest

Thu Dec 29, 2016 4:14 pm   



On 29/12/16 03:22, Steve Wilson wrote:
Quote:
krw_at_notreal.com wrote:

On Wed, 28 Dec 2016 21:05:26 GMT, Steve Wilson <no_at_spam.com> wrote:

krw_at_notreal.com wrote:

On Wed, 28 Dec 2016 18:35:47 GMT, Steve Wilson <no_at_spam.com> wrote:

John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Risk-takers invent things and design circuits. I don't have any
Chinese competition, and don't expect any.

Why have the Chinese not copied your products and put them on the
market at half price?

It's hard to cut the cost of service in half.

As I understand, his cost of service is close to zero. The Chinese can
easily match that.

You obviously don't understand much.

Why do you feel they will not do so in the future?

His business isn't copying others.

Non sequitur.

You obviously don't understand much.

You waste my time. PLONK


You're not the first to reach that conclusion.

A tell-tale on this group is to spot those with
potty mouths and those that frequently use irrelevant
(and incorrect) political labels.

John Larkin
Guest

Fri Dec 30, 2016 12:38 am   



On Thu, 29 Dec 2016 03:59:19 GMT, Steve Wilson <no_at_spam.com> wrote:

Quote:
John Larkin <jjlarkin_at_highlandtechnology.com> wrote:

Even LT Spice sometimes does goofy things. What most people do, when
it is erratic, is fiddle with time steps or solver options or parts
values until we get something that we *think* looks closer to reality,
in other words tells us what we want to hear. But maybe we just made
it worse.

I have been using SPICE since the DOS days. It is excellent at displaying
ideal situations without circuit parasitics such as crosstalk, ground
bounce, circuit noise, EMI and RFI problems, thermal effects, degraded
components, variations in components, and all the other ills that plague
circuit design. It is excellent at showing you how a circuit actually
works, so you can understand what you are really seeing when viewing a
noisy trace on an oscilloscope.

It is extremely good at developing new circuit ideas since you don't have
to waste time soldering and unsoldering various components to test a new
idea. You can view any parameter in a circuit that you wish, such as the
base current in a Colpitts oscillator. You could never do this in
hardware.

You cannot destroy a circuit by operating it beyond component limits. An
actual circuit may be difficult to troublehoot if it only lasts a
microsecond past power on.

But you have to be careful about your models. The simulation is only as
good as the model you supply.


For things like opamps, I don't supply the models. Most IC models are
encrypted, so I have no idea what's inside. I do know that many IC
models are unrealistic. I used one LTC opamp model that could generate
kilovolts on one pin, probably from some internal ideal current
source.

LT Spice defaults to speed over accuracy. Measure the base current of
a 2N2222 at -100 kilovolts on the base.

Spice also tends to dynamically pick big dTs, which can have strange
effects. Connect an L in parallel with a C and goose it to ring.
Measure the period. Compare to the calculated period.


I notice you have made a number of false
Quote:
assumptions about the capabilities of SPICE, based on models that fail
basic mathematical analysis. SPICE is much better than you think, and you
have used the failure of your models to claim that much larger
simulations are also wrong.


My simulations aren't often wrong, but they are right because I can
spot when Spice is doing weird things, and I *do* fiddle until I think
it's modeling accurately.

One situation when LT Spice is wrong is when there are a wide range of
time constants in a multiple-feedback loop. If you set dT small enough
that the sim is accurate, it will take hours to simulate a millisecond
of real time. With the default dT, it can do crazy things.

Quote:

Witness the success of designing airplanes. These are obviously multi-
billion dollar projects, and simulating the airflow and dynamic
performance is essential to the success of the project. When the
prototype takes off on its first flight, it is highly instrumented and
they can measure just about every important flight parameter.


Why prototype and why instrument an aircraft if the sim is accurate?

Note that NASA and the big aircraft companies still use wind tunnels.
The turbulent air flow models aren't entirely to be trusted, even with
petaflop supercomputers. Things like wing flutter still surprise
people.


If you
Quote:
watch the Youtube videos of the flight analysis, they show the result of
dynamic perturbations match the simulations almost exactly.


On Youtube maybe.

Quote:

So the basic approach of fiddling until you "get something that we
*think* looks closer to reality" is wrong. I have been surprised many
times by finding that a circuit performs much different from what I
expected, and this has led to new appreciations of circuit design and
performance.


If you are surprised "many times" by unexpected circuit behavior, you
are modeling and designing badly.

Quote:

Just understand your models and their limitations, and you can build on
the new knowledge of a good simulation.


I can't understand an encrypted behavioral model, except by fiddling
with it and judging if its behavior makes sense to me. That's not very
scientific.

The fast circuits - nanosecond and picosecond - I don't usually bother
to model at all; I breadboard that stuff, or just design the product
and see how the first one works.

That's what's fun about electronics to me. It's not all analytical.
There is plenty of room for instinct and lots of cool surprises. I
know things about parts, and the use of parts, that the manufacturers
don't.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

Goto page Previous  1, 2, 3  Next

elektroda.net NewsGroups Forum Index - Electronics Design - good book, slightly on-topic

Ask a question - edaboard.com

Arabic versionBulgarian versionCatalan versionCzech versionDanish versionGerman versionGreek versionEnglish versionSpanish versionFinnish versionFrench versionHindi versionCroatian versionIndonesian versionItalian versionHebrew versionJapanese versionKorean versionLithuanian versionLatvian versionDutch versionNorwegian versionPolish versionPortuguese versionRomanian versionRussian versionSlovak versionSlovenian versionSerbian versionSwedish versionTagalog versionUkrainian versionVietnamese versionChinese version
RTV map EDAboard.com map News map EDAboard.eu map EDAboard.de map EDAboard.co.uk map