Phil Allison <pallison49@gmail.com> wrote in news:b7756c49-689e-4ad7- a010-cc8c8f46ae77@googlegroups.com:> John Larkin wrote: > > ------------------ > >>> >> >> ** JL leaves you wondering if he is in agreement - or not. >> > >> >He was agreeing and elaborating, adding a current example. >> > >> >> Guess he just likes to "have an edge" as Mr Eastwood remarked. >> >> I am under no obligation to make sense, > > > > ** ROTFL - John, you have certainly succeeded in that direction. > > > Wot a bullshitting ass. > > > > ..... PhilIt was all that surplus lead on his solder joints.
"Mike Engelhardt has parted ways with Analog Devices"
Started by ●March 4, 2020
Reply by ●March 11, 20202020-03-11
Reply by ●March 11, 20202020-03-11
DecadentLinuxUserNumeroUno@decadence.org wrote in news:r4bq1u$11mv$1 @gioia.aioe.org:> Phil Allison <pallison49@gmail.com> wrote in news:b7756c49-689e-4ad7-> a010-cc8c8f46ae77@googlegroups.com: > >> John Larkin wrote: >> >> ------------------ >> >>>> >>> >> ** JL leaves you wondering if he is in agreement - or not. >>> > >>> >He was agreeing and elaborating, adding a current example. >>> > >>> >> Guess he just likes to "have an edge" as Mr Eastwoodremarked.>>> >>> I am under no obligation to make sense, >> >> >> >> ** ROTFL - John, you have certainly succeeded in that direction. >> >> >> Wot a bullshitting ass. >> >> >> >> ..... Phil > > It was all that surplus lead on his solder joints. >He even gets his stencils a little thick...
Reply by ●March 11, 20202020-03-11
On 8/3/20 3:24 pm, jlarkin@highlandsniptechnology.com wrote:> On Sun, 8 Mar 2020 04:01:13 -0000 (UTC), Przemek Klosowski > <przemek@tux.dot.org> wrote: >> Mike's 2018 interview says that his main job at AD is creating AD part >> models in LTSpice >> >> https://www.youtube.com/watch?v=x6TrbD7-IwU > > "Intuition is the most important part of engineering." > That's great.I think it's true in computer programming too. However I'd like to point out that a large part of this "intuition" actually uses our aesthetic sense. By experience, we learn to prefer solutions that have a certain simplicity, and to avoid things we've come to recognise as kluges. This is *learned* behaviour, not "natural intuition". The aesthetic preference applies in mathematics too, of course. CH
Reply by ●March 11, 20202020-03-11
On Wed, 11 Mar 2020 15:08:06 -0700 (PDT), Phil Allison <pallison49@gmail.com> wrote:>John Larkin is Nuts wrote: > >----------------- > >> > >> >>> >> >> >> ** JL leaves you wondering if he is in agreement - or not. >> >> > >> >> >He was agreeing and elaborating, adding a current example. >> >> > >> >> >> Guess he just likes to "have an edge" as Mr Eastwood remarked. >> >> >> >> I am under no obligation to make sense, >> > >> > >> > >> >** ROTFL - John, you have certainly succeeded in that direction. >> >> Yes. The circuits almost always work first try. >> > >** He did it again !!!Yes. You should see the 70 volt pulses that I'm making.> > Wot a hoot.Yup, great fun. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by ●March 11, 20202020-03-11
On Thursday, March 12, 2020 at 1:39:52 AM UTC+11, jla...@highlandsniptechnology.com wrote:> On 11 Mar 2020 06:28:32 -0700, Winfield Hill <winfieldhill@yahoo.com> > wrote: > > >Bill Sloman wrote... > >> > >> There are other ways of predicting what a circuit will do, > >> and you can - in principle - create you own Spice models of > >> parts which better reflect what they do in the circumstances > >> in which you wish to use them. > > > > I like to make small SPICE models of parts, derived from > > analytical expressions of a few critical things that're > > going on in the aspect of the circuit I'm evaluating. > > > > For example, consider an op-amp driving a power MOSFET to > > create a controlled current source. The FET's high gate > > capacitance, along with the bootstrapped source resistor, > > creates a confusing control loop. My RIS-796, a 250-amp > > LED pulser project, uses this. If you want, sets of files: > > > >https://www.dropbox.com/sh/tcmiahzzughadfk/AABtgFDy01cuTDWDRjujP6jva?dl=1 > > > > In AoE x-Chapter 4x.26, we struggled and derived a set of > > analytical equations for this circuit. See article here. > > > >https://www.dropbox.com/s/o4g4mhzl70rsi9t/4x.26_MOSFET_CS_nodal-analysis_final.pdf? > >dl=1 > > > > The circuit basis for the equations uses the MOSFET's gm, > > equation id = gm (vg-vs), and its gate capacitance, Ciss. > > The equations are hairy. But we also suggest you can make > > a simple SPICE circuit with the op-amp, the FET's id and > > Ciss, plus additional Rs and Cs, to evaluate the circuit. > > > > Such a scheme may only works well over a limited range of > > conditions, e.g., using the value for gm at the FET's 250A > > current, means that the reduced-current startup won't be > > accurately modeled. But it's still quick and useful. And > > you can repeat the SPICE run, with lower values of gm, to > > get an idea of what's happening during the pulse startup. > > Nearly all interesting systems are nonlinear, and analytic equations > are hard or impossible for nonlinear systems.The Gummel-Poon model of the transistor is non-linear - largely exponential - but it is extensively used. It might be hard, but it certainly isn't impossible.> So the math becomes guidance, suggestions or starting points for > simulation or experiment.Mathematical models are always over-simplifications of reality, but large chunks of reality are non-linear, and the guidance can be very useful. -- Bill Sloman, Sydney
Reply by ●March 11, 20202020-03-11
On Thursday, March 12, 2020 at 6:00:19 AM UTC+11, dagmarg...@yahoo.com wrote:> On Tuesday, March 10, 2020 at 7:30:24 PM UTC-4, Phil Allison wrote: > > John Larkin wrote:<snip>> > > >> but if intuition is all you've got, you might agree. > > > > > > > >** And if you have none you are stuffed. > > Yep. Math is for verifying an idea after the fact. Or sometimes > predicting that a better idea is possible. But without the idea > first, there's nothing to do the math on.John Larkin's claim was " "Intuition is the most important part of engineering." Engineering may include inventing new solutions, but that's only part of the business. Intuition is handy, but it's not under rational control, and it comes up with a lot of bad ideas. The most important part of engineering is transforming ideas into stuff that works reliably. Everybody has intuition. https://en.wikipedia.org/wiki/Thinking,_Fast_and_Slow points out that everybody has two modes of thought - the fast, quick and instinctive reactions that get labelled as intuitions - and the slower, more rational approach, that involves chains of reasoning. Engineering is all about the second mode. <snip> -- Bill Sloman, Sydney
Reply by ●March 11, 20202020-03-11
Bill Sloman is Lying wrote:> dagmarg...@yahoo.com wrote: > > Phil Allison wrote: > > > John Larkin wrote: > > > > > > >> but if intuition is all you've got, you might agree. > > > > > > > > > >** And if you have none you are stuffed. > > > > Yep. Math is for verifying an idea after the fact. Or sometimes > > predicting that a better idea is possible. But without the idea > > first, there's nothing to do the math on. > > John Larkin's claim was " "Intuition is the most important part of engineering."** Not JL's claim at all.> > Engineering may include inventing new solutions, but that's only part of the business. Intuition is handy, but it's not under rational control,** Nonsense. Because Bill lacks insight and intuative thinking he is relying on a dictionary definition that is not appropriate I predicted exactly this in an earlier post.> The most important part of engineering is transforming ideas into stuff that works reliably.** Major red herring.> > Everybody has intuition.** Wrong definition. Engineering "intuition" is a totally different thing. Already described by me as being informed by great insight born of experience and understanding. Bill will never get it, since he will never admit being a dull autistic and rote learning thinker. Plus giant asshole. ..... Phil
Reply by ●March 11, 20202020-03-11
On Thursday, March 12, 2020 at 6:21:36 AM UTC+11, dagmarg...@yahoo.com wrote:> On Tuesday, March 10, 2020 at 6:49:50 PM UTC-4, George Herold wrote: > > On Monday, March 9, 2020 at 10:56:08 AM UTC-4, jla...@highlandsniptechnology.com wrote: > > > On Mon, 9 Mar 2020 04:14:07 -0700 (PDT), bulegoge@columbus.rr.com > > > wrote:<snip>> And to John's point, much of that steel came to us by > Bessemer's fiddling not suggested to him by any equation > unavailable to everyone else, but by sheer dogged empirical > determination, then additions and refinements from later > minds.Except that James Nasmyth had much the same idea - to the extent that Bessemer offered him a one third share in the patent. https://en.wikipedia.org/wiki/James_Nasmyth Robert Forester Mushet did thousands of experiments at his Darkhill Ironworks to get the process to work right, and Bessemer ended up giving him an annual pension of £300, a very considerable sum, which he paid for over 20 years; possibly with a view to keeping the Mushets from legal action.> And why did Bessemer do it?He was an inventor from way back, with a long history of looking for better ways to provide a product for which there was a known market.> Wiki says he was inspired by a conversation with Napoleon III to solve the high price of artillery. I.e., the need was to facilitate war.It's not in this Wikipedia article - the reference is only to military ordnance https://en.wikipedia.org/wiki/Henry_Bessemer> Bessemer's motivation? Possibly profit.Obviously profit.> He made some astute business > maneuvers suggesting so. And the innovation and the cheap steel that > resulted certainly wouldn't have happened in a socialist country, > without that profit motive.James Arthur thinks that communist countries are the only socialist countries. Democratic socialist countries - as in Scandinavia - are perfectly comfortable with the profit motive.> (And think of the loss to humanity, > think of the world today without cheap steel...)Another one of James Arthur's concocted sound-bites.> But personally, like some of us, I think Bessemer did it because > he was having a blast. > > > I totally agree about intelligent fiddling. (intuition) > > which you only get by fiddling with stuff.If you fiddle intelligently, you can make sense of what you are doing, and it stops being intuitive. It's a slower process, but much less error-prone.> > I should do more ltspice. I mostly fiddle with solder.It's a lot quicker to exploit mathematical simulations - worked out by people who had thought about what they were doing - and they do produce more reliable results than intuition, but every mathematical model is a simplification of reality. You've still got to solder something together and make sure that it does what ltspice predicts it will. -- Bill Sloman, Sydney
Reply by ●March 11, 20202020-03-11
On Thursday, March 12, 2020 at 6:38:28 AM UTC+11, John Larkin wrote:> On Wed, 11 Mar 2020 00:45:47 -0700 (PDT), buecherk@gmail.com wrote: > > > > >> Engineering involves thinking about what you are doing. If you rely only on intuition you are an artist, not an engineer. > > > >No. No good efficient engineering without intuition. > > > >There are 37 possible solutions to my problem. I could try them one by one and finish in 2027. I could simulate for hours, do analysis and calculations, to determine where to start. > > There might be 1e7 solutions to your problem. You can't simulate them > all. You can sleep on it and let your brain evaluate the 1e7 solutions > and pick a few good ones to simulate.Or you can look at few of them and think about it, which gives your sleeping brain more to work with.> >Good intuition and experience 'might' make me choose the right one at the startoff. And then, combine that with good engineering and simulation and soldering and .... > > But don't commit too soon. It's best to stay confused for a few days. > > It's also very helpful to talk to someone else about a circuit > concept. I just did that, and the results were good, if hard to draw.Talking to somebody about a circuit concept is central to engineering. Lots of engineering thinking is non-verbal, and nailing the ideas down clearly enough to let you talk about them is a valuable exercise. You've got to do it anyway when you document the design, but it really does pay to do it early. -- Bill Sloman, Sydney
Reply by ●March 11, 20202020-03-11
On Thursday, March 12, 2020 at 9:01:59 AM UTC+11, John Larkin wrote:> On Wed, 11 Mar 2020 14:44:34 -0700 (PDT), Phil Allison > <pallison49@gmail.com> wrote: > > >John Larkin wrote: > > > >------------------ > > > >>> > >> >> ** JL leaves you wondering if he is in agreement - or not. > >> > > >> >He was agreeing and elaborating, adding a current example. > >> > > >> >> Guess he just likes to "have an edge" as Mr Eastwood remarked. > >> > >> I am under no obligation to make sense, > > > > > > > >** ROTFL - John, you have certainly succeeded in that direction. > > Yes. The circuits almost always work first try. > > Read this: > > https://www.amazon.com/What-Care-Other-People-Think/dp/0393355640/ref=sr_1_2?keywords=book+feinman+people+think&qid=1583964038&sr=8-2Sadly, John Larkin is no Feinman. What worked for a card-carrying physics genius isn't a effective for people who don't know enough about science to realise that observational sciences are as reliable as experimental sciences. -- Bill Sloman, Sydney
