PSU Design...

[Anthony William Sloman]

On Monday, May 8, 2023 at 10:35:20 PM UTC+10, John Larkin wrote:

On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

<snip>

I just find it fascinating how different people excell at certain things, but are useless at others. It\'s a source of infinite wonder to me.

Humans sure specialize. I think that, in a tribe, some people grow food, some make arrows, and some shoot arrows. Cows and oysters don\'t seem to specialize as much.

Humans are social animals. We haven\'t been that way as long as the social insects so the specialistion is in a pretty primitive stage at the moment

> What fascinates me is how humans, hunter-gatherers, evolved to be able to do calculus and design electronics.

Since you don\'t seem to be able to manage either, I do imagine that you do find both skills fascinating. There are other skills. Priests and administrators have been around for longer. Carpenters and masons go back even further, but metal working is more recent. Scribes came in with priests and administrators. but since you don\'t seem to be able to write about what you do that particular skill may not register with you.

Cursitor Doom seems to be some kind of bard - he doesn\'t understand anything but he does like an absurd story.

--
Bill Sloman, Sydney

 

[John Larkin]

On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <cd@notformail.com>
wrote:

On Mon, 08 May 2023 05:35:04 -0700, John Larkin
jlarkin@highlandSNIPMEtechnology.com> wrote:

On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <cd@notformail.com
wrote:

On Mon, 08 May 2023 04:43:24 -0700, John Larkin
jlarkin@highlandSNIPMEtechnology.com> wrote:

On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <cd@notformail.com
wrote:

On Sun, 7 May 2023 22:38:18 +0100, piglet <erichpwagner@hotmail.com
wrote:

On 07/05/2023 13:39, Cursitor Doom wrote:

Well, I\'ve never seen anything like this before in a linear PSU from a
respected company. It\'s almost as if they were forced to use
transformers with not enough suitable secondary windings. There\'s a
revision date on this board: 1971.

Not sure I understand why you think there aren\'t enough secondary
windings? There is one for -/+50, one for -/+15 one for +5 and one for
80-90 which added to +50 made 130V. Seems like plenty, how would more
windings make it simpler? The 130V regulation occurring at 50V is
economical, other HV supplies of that era used that trick too.

piglet

Well it\'s simple for you because you\'re an electronics wizzard, Erich.
I\'m just a humble hobbyist and \"not particularly good at electronics\"
as I describe myself. I certainly did the right thing by not taking it
up as a career, I\'m perfectly certain of that.

Well, that Tek schamatic is kind of a tangle. Not at all obvious on
first sight.

I\'ve been chatting to some chap on another forum who\'s repaired
several of these boards and he agrees. They\'re a pain in the arse
having those interdependencies and I don\'t care what the venerable
piglet says to the contrary.


Maybe you\'re not autistic enough to be very good at electronics
design.

I just find it fascinating how different people excell at certain
things, but are useless at others. It\'s a source of infinite wonder to
me.

Humans sure specialize. I think that, in a tribe, some people grow
food, some make arrows, and some shoot arrows. Cows and oysters don\'t
seem to specialize as much.

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants.

Cows and oysters don\'t do that.

We\'re talking mostly
small discoveries and inventions over huge periods of time, with the
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.

AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

 

[Anthony William Sloman]

On Monday, May 8, 2023 at 11:03:24 PM UTC+10, Cursitor Doom wrote:

On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

<snip>

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly small discoveries and inventions over huge periods of time, with the odd quantum leap thrown in every couple of hundred years. Now we stand on the brink of AI and that should be the new quantum leap.

There aren\'t any quantum leaps. There are lots of small discoveries and inventions - more in recent years as we have developed social tools like writing and peer-reviewed publication to let us pick out the more useful ones and pass them on to other people who do find them useful.

Historians of science do like to pick out culture heroes - like Newton, even though the secretive creep gets a lot of credit for stuff that other people invented at much the same time and passed on to other people while he was reluctant to publish take the credit until he wrung every last bit of advantage out getting the idea early.

Sadly cooperative progress makes for less dramatic narratives.

We\'ll know that artificial intelligence has got here when Google puts in a filter that stops a a and Cursitor Doom from posting irrelevant nonsense here. I\'m not holding my breath.

--
Bill Sloman, Sydney

 

[Anthony William Sloman]

On Tuesday, May 9, 2023 at 1:17:08 AM UTC+10, John Larkin wrote:

On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

<snip>

> AI will probably turn out to be nonsense. It\'s just some more code.

It solved the protein folding problem.

> Somebody please ask AI for a circuit to parallel 7812\'s.

Real intelligence would tell you not to bother. If you need a regulated +12V output of more than 1.5A , get a bigger linear regulator or move to a switcher so you don\'t have dissipate as much heat. You\'ll need to filter the output of switching regular fairly carefully, but there are lots of options.. Not having to dissipate a lot of heat helps a lot.

--
Bill Sloman, Sydney

 

[Cursitor Doom]

On Mon, 8 May 2023 08:35:03 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

On Tuesday, May 9, 2023 at 1:17:08?AM UTC+10, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

AI will probably turn out to be nonsense. It\'s just some more code.

It solved the protein folding problem.

Somebody please ask AI for a circuit to parallel 7812\'s.

Real intelligence would tell you not to bother. If you need a regulated +12V output of more than 1.5A , get a bigger linear regulator or move to a switcher so you don\'t have dissipate as much heat. You\'ll need to filter the output of switching regular fairly carefully, but there are lots of options. Not having to dissipate a lot of heat helps a lot.

Is that the best you can come up with, Bill? Seriously, I could have
come up with something more useful than that succession of platitudes.

 

[Cursitor Doom]

On Mon, 8 May 2023 08:20:04 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

On Monday, May 8, 2023 at 11:03:24?PM UTC+10, Cursitor Doom wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly small discoveries and inventions over huge periods of time, with the odd quantum leap thrown in every couple of hundred years. Now we stand on the brink of AI and that should be the new quantum leap.

There aren\'t any quantum leaps. There are lots of small discoveries and inventions - more in recent years as we have developed social tools like writing and peer-reviewed publication to let us pick out the more useful ones and pass them on to other people who do find them useful.

No quantum leaps?? What about Pythagoras? Galileo? Newton/Leibnitz?
Einstein? Gutenberg? Marconi? Nobel? Oppenheimer? Teller? Berners-Lee?

>We\'ll know that artificial intelligence has got here when Google puts in a filter that stops a a and Cursitor Doom from posting irrelevant nonsense here. I\'m not holding my breath.

Yes, that would be a wonderfully useful tool for Socialists who can\'t
take their crackpot theories being deconstructed by people they\'d much
rather execute if the possibility existed.

 

[whit3rd]

On Monday, May 8, 2023 at 6:03:24 AM UTC-7, Cursitor Doom wrote:

On Mon, 08 May 2023 05:35:04 -0700, John Larkin
jla...@highlandSNIPMEtechnology.com> wrote:

On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com
wrote:

On Mon, 08 May 2023 04:43:24 -0700, John Larkin
jla...@highlandSNIPMEtechnology.com> wrote:

On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com
wrote:

On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com
wrote:

On 07/05/2023 13:39, Cursitor Doom wrote:

Well, I\'ve never seen anything like this before in a linear PSU from a
respected company. It\'s almost as if they were forced to use
transformers with not enough suitable secondary windings. There\'s a
revision date on this board: 1971.

Not sure I understand why you think there aren\'t enough secondary
windings? There is one for -/+50, one for -/+15 one for +5 and one for
80-90 which added to +50 made 130V. Seems like plenty, how would more
windings make it simpler? The 130V regulation occurring at 50V is
economical, other HV supplies of that era used that trick too.

piglet

Well it\'s simple for you because you\'re an electronics wizzard, Erich..
I\'m just a humble hobbyist and \"not particularly good at electronics\"
as I describe myself. I certainly did the right thing by not taking it
up as a career, I\'m perfectly certain of that.

Well, that Tek schamatic is kind of a tangle. Not at all obvious on
first sight.

I\'ve been chatting to some chap on another forum who\'s repaired
several of these boards and he agrees. They\'re a pain in the arse
having those interdependencies and I don\'t care what the venerable
piglet says to the contrary.


Maybe you\'re not autistic enough to be very good at electronics
design.

I just find it fascinating how different people excell at certain
things, but are useless at others. It\'s a source of infinite wonder to
me.
....
What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly
small discoveries and inventions over huge periods of time, with the
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.

The first leap was complex speech; that was long enough ago that we
have \"Broca\'s region\", a brain section for the purpose.

The second leap, going from lore in poems and tales to print, was... paper.
Near-universal literacy and inexpensive reams of writing material is the
basis of modern thought. My Kindle isn\'t as durable a reader as my old books.

We aren\'t finished incorporating literacy, there\'s too many symbol systems (and
even pictograph writing, very inefficient) and that divides modern societies.
It was a great leap forward when Latin, the language of Rome, was spread over
the ancient world; that dominated for a millenium AFTER the fall of the empire.
I\'m still using the symbol set, a kind of Latin alphabet.

AI will be good for translation, I suppose, but failing to make a survival-like test
for inner parts of AI construct will mean that AI is not going to get its faults
pruned out; it\'ll grow, like redundant DNA codons, into several kinds of monstrosity.

 

[John Larkin]

On Mon, 08 May 2023 18:29:20 +0100, Cursitor Doom <cd@notformail.com>
wrote:

On Mon, 8 May 2023 08:35:03 -0700 (PDT), Anthony William Sloman
bill.sloman@ieee.org> wrote:

On Tuesday, May 9, 2023 at 1:17:08?AM UTC+10, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

AI will probably turn out to be nonsense. It\'s just some more code.

It solved the protein folding problem.

Somebody please ask AI for a circuit to parallel 7812\'s.

Real intelligence would tell you not to bother. If you need a regulated +12V output of more than 1.5A , get a bigger linear regulator or move to a switcher so you don\'t have dissipate as much heat. You\'ll need to filter the output of switching regular fairly carefully, but there are lots of options. Not having to dissipate a lot of heat helps a lot.

Is that the best you can come up with, Bill? Seriously, I could have
come up with something more useful than that succession of platitudes.

https://www.breitbart.com/tech/2023/05/08/ai-systems-like-chatgpt-rely-on-an-army-of-trainers-making-15-an-hour/

So, behind AI, are some low-paid people answering the questions!

https://www.youtube.com/watch?v=YWyCCJ6B2WE

 

[Cursitor Doom]

On Mon, 8 May 2023 11:47:22 -0700 (PDT), whit3rd <whit3rd@gmail.com>
wrote:

On Monday, May 8, 2023 at 6:03:24?AM UTC-7, Cursitor Doom wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin
jla...@highlandSNIPMEtechnology.com> wrote:

On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com
wrote:

On Mon, 08 May 2023 04:43:24 -0700, John Larkin
jla...@highlandSNIPMEtechnology.com> wrote:

On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com
wrote:

On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com
wrote:

On 07/05/2023 13:39, Cursitor Doom wrote:

Well, I\'ve never seen anything like this before in a linear PSU from a
respected company. It\'s almost as if they were forced to use
transformers with not enough suitable secondary windings. There\'s a
revision date on this board: 1971.

Not sure I understand why you think there aren\'t enough secondary
windings? There is one for -/+50, one for -/+15 one for +5 and one for
80-90 which added to +50 made 130V. Seems like plenty, how would more
windings make it simpler? The 130V regulation occurring at 50V is
economical, other HV supplies of that era used that trick too.

piglet

Well it\'s simple for you because you\'re an electronics wizzard, Erich.
I\'m just a humble hobbyist and \"not particularly good at electronics\"
as I describe myself. I certainly did the right thing by not taking it
up as a career, I\'m perfectly certain of that.

Well, that Tek schamatic is kind of a tangle. Not at all obvious on
first sight.

I\'ve been chatting to some chap on another forum who\'s repaired
several of these boards and he agrees. They\'re a pain in the arse
having those interdependencies and I don\'t care what the venerable
piglet says to the contrary.


Maybe you\'re not autistic enough to be very good at electronics
design.

I just find it fascinating how different people excell at certain
things, but are useless at others. It\'s a source of infinite wonder to
me.
...
What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly
small discoveries and inventions over huge periods of time, with the
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.

The first leap was complex speech; that was long enough ago that we
have \"Broca\'s region\", a brain section for the purpose.

The second leap, going from lore in poems and tales to print, was... paper.
Near-universal literacy and inexpensive reams of writing material is the
basis of modern thought. My Kindle isn\'t as durable a reader as my old books.

We aren\'t finished incorporating literacy, there\'s too many symbol systems (and
even pictograph writing, very inefficient) and that divides modern societies.
It was a great leap forward when Latin, the language of Rome, was spread over
the ancient world; that dominated for a millenium AFTER the fall of the empire.
I\'m still using the symbol set, a kind of Latin alphabet.

AI will be good for translation, I suppose, but failing to make a survival-like test
for inner parts of AI construct will mean that AI is not going to get its faults
pruned out; it\'ll grow, like redundant DNA codons, into several kinds of monstrosity.

All these scientists who\'ve come out demanding a pause in AI
development aren\'t concerned that AI will turn rogue and start trying
to wipe us out. They\'re worried *humans* will take a valuable tool
which has much capacity to do good, and weaponise it against the rest
of humanity. I think we should listen to them and address their
concerns with legislation if appropriate.

 

[Anthony William Sloman]

On Tuesday, May 9, 2023 at 3:30:50 AM UTC+10, Cursitor Doom wrote:

On Mon, 8 May 2023 08:35:03 -0700 (PDT), Anthony William Sloman
bill....@ieee.org> wrote:
On Tuesday, May 9, 2023 at 1:17:08?AM UTC+10, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail..com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

AI will probably turn out to be nonsense. It\'s just some more code.

It solved the protein folding problem.

Somebody please ask AI for a circuit to parallel 7812\'s.

Real intelligence would tell you not to bother. If you need a regulated +12V output of more than 1.5A , get a bigger linear regulator or move to a switcher so you don\'t have dissipate as much heat. You\'ll need to filter the output of switching regular fairly carefully, but there are lots of options. Not having to dissipate a lot of heat helps a lot.

Is that the best you can come up with, Bill? Seriously, I could have come up with something more useful than that succession of platitudes.

It recognised that John Larkin was making a joke, and responded in a similar vein. Your vaunted sense of humour has let you down, or maybe you don\'t know enough to recognise a comic exchange when you see one.

--
Bill Sloman, Stdney

 

[Anthony William Sloman]

On Tuesday, May 9, 2023 at 3:44:04 AM UTC+10, Cursitor Doom wrote:

On Mon, 8 May 2023 08:20:04 -0700 (PDT), Anthony William Sloman
bill....@ieee.org> wrote:
On Monday, May 8, 2023 at 11:03:24?PM UTC+10, Cursitor Doom wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly small discoveries and inventions over huge periods of time, with the odd quantum leap thrown in every couple of hundred years. Now we stand on the brink of AI and that should be the new quantum leap.

There aren\'t any quantum leaps. There are lots of small discoveries and inventions - more in recent years as we have developed social tools like writing and peer-reviewed publication to let us pick out the more useful ones and pass them on to other people who do find them useful.

No quantum leaps?? What about Pythagoras?

The \"Pythagoras\" theorem was known to the Babylonians

> Galileo?

Galileo exploited - but didn\'t invent - the telescope. Scarcely a quantum leap.

> Newton/Leibnitz?

Leibnitz and Newton both \"invented\" calculus at much the same time, which does suggest that some earlier work had given them both much the same idea.. The notation we use is the one Leibnitz talked about to other people. Newton called his version \"fluxions\" and was much slower to tell other people about it.

> Einstein?

Was a very clever guy, but his insights were based on earlier work. He may have invented Special Relativity, but one of it\'s features is called Lorenz contraction;

https://en.wikipedia.org/wiki/Length_contraction

> Gutenberg?

Gutenberg introduced the printing press in Europe, long after it has been popular in China. The Chinese writing system uses a lot more characters than European alphabets, so moveable type was rather simpler and more useful in Europe. The process by which European writing systems became alphabetic doesn\'t seem to have involved any quantum leaps.

> Marconi?

Marconi wasn\'t an inventer, but rather an exploiter of existing technology.

> Nobel?

Nobel made a lot of money out of the simple idea stabilising nitorgylcerin by soaking it up into kieselguhr

https://en.wikipedia.org/wiki/Diatomaceous_earth

No kind of quantum leap.

> Oppenheimer? Teller? Berners-Lee?

All three took existing technologies a little further.

We\'ll know that artificial intelligence has got here when Google puts in a filter that stops a a and Cursitor Doom from posting irrelevant nonsense here. I\'m not holding my breath.

Yes, that would be a wonderfully useful tool for Socialists who can\'t take their crackpot theories being deconstructed by people they\'d much rather execute if the possibility existed.

I\'m sure you like to think that, but then again you think that you parading your fatuous delusions here is some kind of exercise in \"deconstruction\".

Executing you would solve the problem, but it would be the kind of over-kill that people like you do go in for. Sensible people would prefer to see your talents exploited where they might be useful - if you actually have any kind of useful talent. You do seem to be literate, but you don\'t seem to be able to understand what you read - Alfred Nobel made some kind of quantum leap in technology?

--
Bill Sloman, Sydney

 

[Anthony William Sloman]

On Tuesday, May 9, 2023 at 7:24:05 AM UTC+10, Cursitor Doom wrote:

On Mon, 8 May 2023 11:47:22 -0700 (PDT), whit3rd <whi...@gmail.com> wrote:
On Monday, May 8, 2023 at 6:03:24?AM UTC-7, Cursitor Doom wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

<snip>

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly small discoveries and inventions over huge periods of time, with the odd quantum leap thrown in every couple of hundred years. Now we stand on the brink of AI and that should be the new quantum leap.

The first leap was complex speech; that was long enough ago that we have \"Broca\'s region\", a brain section for the purpose.

https://www.sciencedirect.com/topics/medicine-and-dentistry/wernicke-area

Broca\'s and Wernicke\'s areas of the brain are important in speech processing, and your speech is impaired if either is damaged. Both are present in non-human primates, so they weren\'t evolved to serve that purpose, but rather co-opted to manage the work when we started evolving speech and seem to have got bigger as we got better at speech and made more use of it.

> >The second leap, going from lore in poems and tales to print, was... paper. Near-universal literacy and inexpensive reams of writing material is the basis of modern thought. My Kindle isn\'t as durable a reader as my old books.

The Babylonians went in for clay tablets, and northern Europeans wrote on parchment a lot, and carved runs into sticks. Paper is just a particular technology.

> >We aren\'t finished incorporating literacy, there\'s too many symbol systems (and even pictograph writing, very inefficient) and that divides modern societies.

Not so that you would notice, China\'s crap writing system works a whole lot better with a computer interface and they are now publishing a lot of scientific papers that seem to integrate seamlessly with the rest of the literature.

It was a great leap forward when Latin, the language of Rome, was spread over the ancient world; that dominated for a millenium AFTER the fall of the empire.
I\'m still using the symbol set, a kind of Latin alphabet.

A common language is always handy. Across the world most people speak several languages. Quite a few English speakers don\'t bother. I got some minimal competence in German French and Russian as a undergraduate, and ended up getting fluent (if not al that correct) in Dutch in my early fifities

AI will be good for translation, I suppose, but failing to make a survival-like test for inner parts of AI construct will mean that AI is not going to get its faults
pruned out; it\'ll grow, like redundant DNA codons, into several kinds of monstrosity.

Redundant DNA codons are often adapted to take on rather different jobs than the ones their ancestral codon had evolved to fill.

AI should acquire the capacity to review its own code - as we seem to be doing with our DNA. It\'s an obvious enough idea.

> All these scientists who\'ve come out demanding a pause in AI development aren\'t concerned that AI will turn rogue and start trying to wipe us out. They\'re worried *humans* will take a valuable tool which has much capacity to do good, and weaponise it against the rest of humanity. I think we should listen to them and address their concerns with legislation if appropriate.

As if Cursitor Doom has read any of the stuff \"all these scientists\" had written. He\'s presumably echoing Russia Today or Breitbart.

Good luck with drafting that kind of legislation. Even with the AI we\'ve got now, drafting laws that lawyers can\'t find a loophole in is impossible, and when the lawyers involved have gotten promoted to the US Supreme Court, rationality is a lost cause.

--
Bill Sloman, Sydney

 

[Cursitor Doom]

On Mon, 8 May 2023 22:27:01 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

On Tuesday, May 9, 2023 at 3:44:04?AM UTC+10, Cursitor Doom wrote:
On Mon, 8 May 2023 08:20:04 -0700 (PDT), Anthony William Sloman
bill....@ieee.org> wrote:
On Monday, May 8, 2023 at 11:03:24?PM UTC+10, Cursitor Doom wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly small discoveries and inventions over huge periods of time, with the odd quantum leap thrown in every couple of hundred years. Now we stand on the brink of AI and that should be the new quantum leap.

There aren\'t any quantum leaps. There are lots of small discoveries and inventions - more in recent years as we have developed social tools like writing and peer-reviewed publication to let us pick out the more useful ones and pass them on to other people who do find them useful.

No quantum leaps?? What about Pythagoras?

The \"Pythagoras\" theorem was known to the Babylonians

Galileo?

Galileo exploited - but didn\'t invent - the telescope. Scarcely a quantum leap.

Newton/Leibnitz?

Leibnitz and Newton both \"invented\" calculus at much the same time, which does suggest that some earlier work had given them both much the same idea. The notation we use is the one Leibnitz talked about to other people. Newton called his version \"fluxions\" and was much slower to tell other people about it.

Einstein?

Was a very clever guy, but his insights were based on earlier work. He may have invented Special Relativity, but one of it\'s features is called Lorenz contraction;

https://en.wikipedia.org/wiki/Length_contraction

Gutenberg?

Gutenberg introduced the printing press in Europe, long after it has been popular in China. The Chinese writing system uses a lot more characters than European alphabets, so moveable type was rather simpler and more useful in Europe. The process by which European writing systems became alphabetic doesn\'t seem to have involved any quantum leaps.

Marconi?

Marconi wasn\'t an inventer, but rather an exploiter of existing technology.

Nobel?

Nobel made a lot of money out of the simple idea stabilising nitorgylcerin by soaking it up into kieselguhr

https://en.wikipedia.org/wiki/Diatomaceous_earth

No kind of quantum leap.

Oppenheimer? Teller? Berners-Lee?

All three took existing technologies a little further.

We\'ll know that artificial intelligence has got here when Google puts in a filter that stops a a and Cursitor Doom from posting irrelevant nonsense here. I\'m not holding my breath.

Yes, that would be a wonderfully useful tool for Socialists who can\'t take their crackpot theories being deconstructed by people they\'d much rather execute if the possibility existed.

I\'m sure you like to think that, but then again you think that you parading your fatuous delusions here is some kind of exercise in \"deconstruction\".

Executing you would solve the problem, but it would be the kind of over-kill that people like you do go in for. Sensible people would prefer to see your talents exploited where they might be useful - if you actually have any kind of useful talent. You do seem to be literate, but you don\'t seem to be able to understand what you read - Alfred Nobel made some kind of quantum leap in technology?

In Nobel\'s case, it wasn\'t the invention of dynamite that was the
quantum leap. The leap was what it enabled mankind to achieve in
undertaking the kind of vast construction projects that before had
been way too labour intensive to be economically viable.
And Teller/Oppenheimer? The progression from fission to fusion in the
space of a few short years wasn\'t a quantum leap in your view??
Jeezus....
As usual, you fail to grasp the huge consequences of all these
advances, preferring instead to focus on inflating your own ego (it\'s
big enough already, Bill!)

 

[Rocky]

On Monday, May 8, 2023 at 5:17:08 PM UTC+2, John Larkin wrote:

On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com
wrote:
snip
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.
AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

ChatGPT gave the following:
To design a circuit that puts 7812 regulators in parallel to provide 10 amps at 12 volts, we can follow the steps below:

Step 1: Calculate the power dissipation of a single 7812 regulator
The maximum output current of a 7812 regulator is 1A. Therefore, the power dissipation of a single 7812 regulator can be calculated as follows:

P = (Vin - Vout) * Iout
P = (Vin - 12V) * 1A
P = (Vin - 12V)

For example, if the input voltage is 18V, the power dissipation of a single 7812 regulator would be 6W.

Step 2: Calculate the number of 7812 regulators needed
To provide 10A at 12V, we need 10/1 = 10 7812 regulators in parallel.

Step 3: Select a suitable power supply
The input voltage to the parallel regulators must be high enough to provide sufficient headroom for the voltage drop across the regulators. We can use a 24V power supply for this purpose.

Step 4: Connect the 7812 regulators in parallel
To connect the 7812 regulators in parallel, we can connect their input pins to the 24V power supply and their output pins to a common output node. We should also add a small value resistor (e.g. 0.1 ohms) in series with the output of each regulator to balance the current sharing between the regulators.

Step 5: Add heat sinks
Since each regulator will dissipate a significant amount of power, we need to add suitable heat sinks to ensure they do not overheat.

Step 6: Test the circuit
We can test the circuit by measuring the output voltage and current using a multimeter. If the output voltage and current are within the desired range, the circuit is working correctly.

Note: Designing high-power circuits like this can be challenging and should only be attempted by experienced engineers with the appropriate skills and tools.

 

[Cursitor Doom]

On Tue, 9 May 2023 05:38:03 -0700 (PDT), Rocky <robertgush@gmail.com>
wrote:

On Monday, May 8, 2023 at 5:17:08?PM UTC+2, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com
wrote:
snip
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.
AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

ChatGPT gave the following:

How much of the verbiage below is a verbatim quote from Chat GPT and
how much is your own commenting?

To design a circuit that puts 7812 regulators in parallel to provide 10 amps at 12 volts, we can follow the steps below:

Step 1: Calculate the power dissipation of a single 7812 regulator
The maximum output current of a 7812 regulator is 1A. Therefore, the power dissipation of a single 7812 regulator can be calculated as follows:

P = (Vin - Vout) * Iout
P = (Vin - 12V) * 1A
P = (Vin - 12V)

For example, if the input voltage is 18V, the power dissipation of a single 7812 regulator would be 6W.

Step 2: Calculate the number of 7812 regulators needed
To provide 10A at 12V, we need 10/1 = 10 7812 regulators in parallel.

Step 3: Select a suitable power supply
The input voltage to the parallel regulators must be high enough to provide sufficient headroom for the voltage drop across the regulators. We can use a 24V power supply for this purpose.

Step 4: Connect the 7812 regulators in parallel
To connect the 7812 regulators in parallel, we can connect their input pins to the 24V power supply and their output pins to a common output node. We should also add a small value resistor (e.g. 0.1 ohms) in series with the output of each regulator to balance the current sharing between the regulators.

Step 5: Add heat sinks
Since each regulator will dissipate a significant amount of power, we need to add suitable heat sinks to ensure they do not overheat.

Step 6: Test the circuit
We can test the circuit by measuring the output voltage and current using a multimeter. If the output voltage and current are within the desired range, the circuit is working correctly.

Note: Designing high-power circuits like this can be challenging and should only be attempted by experienced engineers with the appropriate skills and tools.

 

[Rocky]

On Tuesday, May 9, 2023 at 2:59:16 PM UTC+2, Cursitor Doom wrote:

On Tue, 9 May 2023 05:38:03 -0700 (PDT), Rocky <rober...@gmail.com
wrote:
On Monday, May 8, 2023 at 5:17:08?PM UTC+2, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com
wrote:
snip
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.
AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

ChatGPT gave the following:
How much of the verbiage below is a verbatim quote from Chat GPT and
how much is your own commenting?
To design a circuit that puts 7812 regulators in parallel to provide 10 amps at 12 volts, we can follow the steps below:

Step 1: Calculate the power dissipation of a single 7812 regulator
The maximum output current of a 7812 regulator is 1A. Therefore, the power dissipation of a single 7812 regulator can be calculated as follows:

P = (Vin - Vout) * Iout
P = (Vin - 12V) * 1A
P = (Vin - 12V)

For example, if the input voltage is 18V, the power dissipation of a single 7812 regulator would be 6W.

Step 2: Calculate the number of 7812 regulators needed
To provide 10A at 12V, we need 10/1 = 10 7812 regulators in parallel.

Step 3: Select a suitable power supply
The input voltage to the parallel regulators must be high enough to provide sufficient headroom for the voltage drop across the regulators. We can use a 24V power supply for this purpose.

Step 4: Connect the 7812 regulators in parallel
To connect the 7812 regulators in parallel, we can connect their input pins to the 24V power supply and their output pins to a common output node. We should also add a small value resistor (e.g. 0.1 ohms) in series with the output of each regulator to balance the current sharing between the regulators.

Step 5: Add heat sinks
Since each regulator will dissipate a significant amount of power, we need to add suitable heat sinks to ensure they do not overheat.

Step 6: Test the circuit
We can test the circuit by measuring the output voltage and current using a multimeter. If the output voltage and current are within the desired range, the circuit is working correctly.

Note: Designing high-power circuits like this can be challenging and should only be attempted by experienced engineers with the appropriate skills and tools.

It is simple cut & paste. the prompt was:
\"design a circuit that puts 7812 regulators in parallel to provide 10 amp at 12 volts\"

 

[Anthony William Sloman]

On Tuesday, May 9, 2023 at 6:47:53 PM UTC+10, Cursitor Doom wrote:

On Mon, 8 May 2023 22:27:01 -0700 (PDT), Anthony William Sloman
bill....@ieee.org> wrote:
On Tuesday, May 9, 2023 at 3:44:04?AM UTC+10, Cursitor Doom wrote:
On Mon, 8 May 2023 08:20:04 -0700 (PDT), Anthony William Sloman
bill....@ieee.org> wrote:
On Monday, May 8, 2023 at 11:03:24?PM UTC+10, Cursitor Doom wrote:
On Mon, 08 May 2023 05:35:04 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 13:04:10 +0100, Cursitor Doom <c...@notformail..com> wrote:
On Mon, 08 May 2023 04:43:24 -0700, John Larkin <jla...@highlandSNIPMEtechnology.com> wrote:
On Mon, 08 May 2023 00:23:58 +0100, Cursitor Doom <c...@notformail.com> wrote:
On Sun, 7 May 2023 22:38:18 +0100, piglet <erichp...@hotmail.com> wrote:
On 07/05/2023 13:39, Cursitor Doom wrote:

snip

What fascinates me is how humans, hunter-gatherers, evolved to be able
to do calculus and design electronics.

Simply by standing on the shoulders of giants. We\'re talking mostly small discoveries and inventions over huge periods of time, with the odd quantum leap thrown in every couple of hundred years. Now we stand on the brink of AI and that should be the new quantum leap.

There aren\'t any quantum leaps. There are lots of small discoveries and inventions - more in recent years as we have developed social tools like writing and peer-reviewed publication to let us pick out the more useful ones and pass them on to other people who do find them useful.

No quantum leaps?? What about Pythagoras?

The \"Pythagoras\" theorem was known to the Babylonians

Galileo?

Galileo exploited - but didn\'t invent - the telescope. Scarcely a quantum leap.

Newton/Leibnitz?

Leibnitz and Newton both \"invented\" calculus at much the same time, which does suggest that some earlier work had given them both much the same idea. The notation we use is the one Leibnitz talked about to other people. Newton called his version \"fluxions\" and was much slower to tell other people about it.

Einstein?

Was a very clever guy, but his insights were based on earlier work. He may have invented Special Relativity, but one of it\'s features is called Lorenz contraction;

https://en.wikipedia.org/wiki/Length_contraction

Gutenberg?

Gutenberg introduced the printing press in Europe, long after it has been popular in China. The Chinese writing system uses a lot more characters than European alphabets, so moveable type was rather simpler and more useful in Europe. The process by which European writing systems became alphabetic doesn\'t seem to have involved any quantum leaps.

Marconi?

Marconi wasn\'t an inventer, but rather an exploiter of existing technology.

Nobel?

Nobel made a lot of money out of the simple idea of stabilising nitorgylcerin by soaking it up into kieselguhr

https://en.wikipedia.org/wiki/Diatomaceous_earth

No kind of quantum leap.

Oppenheimer? Teller? Berners-Lee?

All three took existing technologies a little further.

We\'ll know that artificial intelligence has got here when Google puts in a filter that stops a a and Cursitor Doom from posting irrelevant nonsense here. I\'m not holding my breath.

Yes, that would be a wonderfully useful tool for Socialists who can\'t take their crackpot theories being deconstructed by people they\'d much rather execute if the possibility existed.

I\'m sure you like to think that, but then again you think that you parading your fatuous delusions here is some kind of exercise in \"deconstruction\".

Executing you would solve the problem, but it would be the kind of over-kill that people like you do go in for. Sensible people would prefer to see your talents exploited where they might be useful - if you actually have any kind of useful talent. You do seem to be literate, but you don\'t seem to be able to understand what you read - Alfred Nobel made some kind of quantum leap in technology?

In Nobel\'s case, it wasn\'t the invention of dynamite that was the quantum leap. The leap was what it enabled mankind to achieve in
undertaking the kind of vast construction projects that before had been way too labour intensive to be economically viable.

Don\'t be silly. The progressive mechanisation of construction was one of those tedious incremental programs, and Nobel was one of the many people who kept it moving on.

> And Teller/Oppenheimer? The progression from fission to fusion in the space of a few short years wasn\'t a quantum leap in your view??

Clearly not. Neither Teller or Oppenheimer \"invented\" fission.

https://en.wikipedia.org/wiki/Lise_Meitner

probably deserves the credit for that. The fact that Otto Hahn got the Nobel prize for it on his own is one of the more disgraceful episodes in the history of the Nobel prize.
And it was one more bit of incremental development, though nuclear bombs do represent a fairly dramatic increment in our capacity to make a mess of the planet.

Teller merely realised that a fission bomb could get a chunk of tritium, deuterium or lithium deuteride hot and dense enough to undergo nuclear fusion.. Once you had the fission bomb it was a fairly obvious way of getting an even bigger blast - and even more neutrons - which the spectacularly dirty fission-fusion-fission bomb exploited,

Jeezus....
As usual, you fail to grasp the huge consequences of all these advances, preferring instead to focus on inflating your own ego (it\'s big enough already, Bill!)

The huge potential consequences of these developments - the fact that we haven\'t used a fission bomb in anger since 1945 makes it fairly obvious that they aren\'t actually \"advances\" - means that it\'s a technology that we should have had enough sense not to spend time and money on developing.

You don\'t seem to have any sense at all. Quite how pointing this out is supposed to inflate my ego escapes me. Wasting time drawing your attention to the fact that you don\'t know what you are talking about is a depressing reminder that I don\'t have enough constructive activities to fill my time, which in in fact tends more to deflate my ego.

If I had more self-respect I\'d ignore clowns like you, but I\'m reduced to shooting fish in a barrel.

--
Bill Sloman, Sydney

 

[John Larkin]

On Tue, 9 May 2023 05:38:03 -0700 (PDT), Rocky <robertgush@gmail.com>
wrote:

On Monday, May 8, 2023 at 5:17:08?PM UTC+2, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com
wrote:
snip
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.
AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

ChatGPT gave the following:
To design a circuit that puts 7812 regulators in parallel to provide 10 amps at 12 volts, we can follow the steps below:

Step 1: Calculate the power dissipation of a single 7812 regulator
The maximum output current of a 7812 regulator is 1A. Therefore, the power dissipation of a single 7812 regulator can be calculated as follows:

P = (Vin - Vout) * Iout
P = (Vin - 12V) * 1A
P = (Vin - 12V)

For example, if the input voltage is 18V, the power dissipation of a single 7812 regulator would be 6W.

Step 2: Calculate the number of 7812 regulators needed
To provide 10A at 12V, we need 10/1 = 10 7812 regulators in parallel.

Step 3: Select a suitable power supply
The input voltage to the parallel regulators must be high enough to provide sufficient headroom for the voltage drop across the regulators. We can use a 24V power supply for this purpose.

Step 4: Connect the 7812 regulators in parallel
To connect the 7812 regulators in parallel, we can connect their input pins to the 24V power supply and their output pins to a common output node. We should also add a small value resistor (e.g. 0.1 ohms) in series with the output of each regulator to balance the current sharing between the regulators.

Step 5: Add heat sinks
Since each regulator will dissipate a significant amount of power, we need to add suitable heat sinks to ensure they do not overheat.

Step 6: Test the circuit
We can test the circuit by measuring the output voltage and current using a multimeter. If the output voltage and current are within the desired range, the circuit is working correctly.

Note: Designing high-power circuits like this can be challenging and should only be attempted by experienced engineers with the appropriate skills and tools.

What question did you ask?

Whoever did that, they got it wrong. What do you expect for $15 per
hour?

 

[Anthony William Sloman]

On Tuesday, May 9, 2023 at 10:38:07 PM UTC+10, Rocky wrote:

On Monday, May 8, 2023 at 5:17:08 PM UTC+2, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com
wrote:
snip
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.
AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

ChatGPT gave the following:
To design a circuit that puts 7812 regulators in parallel to provide 10 amps at 12 volts, we can follow the steps below:

Step 1: Calculate the power dissipation of a single 7812 regulator
The maximum output current of a 7812 regulator is 1A. Therefore, the power dissipation of a single 7812 regulator can be calculated as follows:

P = (Vin - Vout) * Iout
P = (Vin - 12V) * 1A
P = (Vin - 12V)

For example, if the input voltage is 18V, the power dissipation of a single 7812 regulator would be 6W.

Step 2: Calculate the number of 7812 regulators needed
To provide 10A at 12V, we need 10/1 = 10 7812 regulators in parallel.

Step 3: Select a suitable power supply
The input voltage to the parallel regulators must be high enough to provide sufficient headroom for the voltage drop across the regulators. We can use a 24V power supply for this purpose.

Step 4: Connect the 7812 regulators in parallel
To connect the 7812 regulators in parallel, we can connect their input pins to the 24V power supply and their output pins to a common output node. We should also add a small value resistor (e.g. 0.1 ohms) in series with the output of each regulator to balance the current sharing between the regulators.

Step 5: Add heat sinks
Since each regulator will dissipate a significant amount of power, we need to add suitable heat sinks to ensure they do not overheat.

Step 6: Test the circuit
We can test the circuit by measuring the output voltage and current using a multimeter. If the output voltage and current are within the desired range, the circuit is working correctly.

Note: Designing high-power circuits like this can be challenging and should only be attempted by experienced engineers with the appropriate skills and tools.

As you\'d expect, ChatGPT misses the point that 7812 parts don\'t produce an exactly 12V output, so the chip that happens to have the highest output voltage is going to try and source most of the current. The design problem getting the parts to share the load - and they weren\'t designed to let you do that easily.

Presumably John Larkin had that in mind when he posed the problem - I\'m not impressed by his design skills, but it is a pretty obvious point.

--
Bill Sloman, Sydney

 

[Cursitor Doom]

On Tue, 9 May 2023 07:13:19 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

On Tuesday, May 9, 2023 at 10:38:07?PM UTC+10, Rocky wrote:
On Monday, May 8, 2023 at 5:17:08?PM UTC+2, John Larkin wrote:
On Mon, 08 May 2023 14:01:43 +0100, Cursitor Doom <c...@notformail.com
wrote:
snip
odd quantum leap thrown in every couple of hundred years. Now we stand
on the brink of AI and that should be the new quantum leap.
AI will probably turn out to be nonsense. It\'s just some more code.

Somebody please ask AI for a circuit to parallel 7812\'s.

ChatGPT gave the following:
To design a circuit that puts 7812 regulators in parallel to provide 10 amps at 12 volts, we can follow the steps below:

Step 1: Calculate the power dissipation of a single 7812 regulator
The maximum output current of a 7812 regulator is 1A. Therefore, the power dissipation of a single 7812 regulator can be calculated as follows:

P = (Vin - Vout) * Iout
P = (Vin - 12V) * 1A
P = (Vin - 12V)

For example, if the input voltage is 18V, the power dissipation of a single 7812 regulator would be 6W.

Step 2: Calculate the number of 7812 regulators needed
To provide 10A at 12V, we need 10/1 = 10 7812 regulators in parallel.

Step 3: Select a suitable power supply
The input voltage to the parallel regulators must be high enough to provide sufficient headroom for the voltage drop across the regulators. We can use a 24V power supply for this purpose.

Step 4: Connect the 7812 regulators in parallel
To connect the 7812 regulators in parallel, we can connect their input pins to the 24V power supply and their output pins to a common output node. We should also add a small value resistor (e.g. 0.1 ohms) in series with the output of each regulator to balance the current sharing between the regulators.

Step 5: Add heat sinks
Since each regulator will dissipate a significant amount of power, we need to add suitable heat sinks to ensure they do not overheat.

Step 6: Test the circuit
We can test the circuit by measuring the output voltage and current using a multimeter. If the output voltage and current are within the desired range, the circuit is working correctly.

Note: Designing high-power circuits like this can be challenging and should only be attempted by experienced engineers with the appropriate skills and tools.

As you\'d expect, ChatGPT misses the point that 7812 parts don\'t produce an exactly 12V output, so the chip that happens to have the highest output voltage is going to try and source most of the current. The design problem getting the parts to share the load - and they weren\'t designed to let you do that easily.

Presumably John Larkin had that in mind when he posed the problem - I\'m not impressed by his design skills, but it is a pretty obvious point.

Yeah, so it\'s using an ideal model. Bit of a problem there. But
they\'ll perfect it in time, I\'ve no doubt.