Liste des Groupes | Revenir à se design |
On a sunny day (Fri, 18 Oct 2024 16:33:22 -0000 (UTC)) it happened Cursitor
Doom <cd999666@notformail.com> wrote in <veu2ki$3cmo3$1@dont-email.me>:
>Gentlemen,>
>
I've never spiced an oscillator AFAICR. Do they self-start in spice
simulators (LT in particular)? IOW, did Mike Engleheart build something
into the engine which generates wideband 'background' noise, particularly
at 'power up' as it were? I'm assuming there must be some such mechanism
and if there is, it must be present by default for all schematics one
attempts to simulate and not just oscillators?
Often, if not always, it is simpler to just solder a circuit together and measure it, than that ElTeaSpites jive.
Tried it, have it on a Linux PC, it did not give the right waveforms
lacking accurate models, and environment awareness. peeseebee, housing, cables, what not.
Used it for some LF filters as that saves math, but there are much simpler filter design programs
that do the same and better, just display the curves.
>
ElTeaSpites has become a religion...
Just use a soldering iron and a good scope, maybe also a spectrum analyzer and you have certainty.
I have done LF and GHz stuff that all works that way,
And you will need the right model for ElTeaSpites (always incomplete) and need to measure anyways.
ElTeaSpites sucks anyways when you add microcontrollers and those are everywhere,
You should be OK with all your boat anchors doing good measurements.
Modern stuff uses chips that give very limited info about what's in those,
you will need the right model for ElTeaSpites (those are always incomplete) and need to measure anyways.
Just test small sub-circuits and put it all together.
An alternative would perhaps be to as AI to design stuff.. save a lot of the time to get familiar with the ElTeaSpites.
>
Is that where it goes for electronics design? AI?
Les messages affichés proviennent d'usenet.