Sujet : Re: 80dB now but still needs improvement at 1KHz
De : JL (at) *nospam* gct.com (john larkin)
Groupes : sci.electronics.designDate : 06. Nov 2024, 17:37:11
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <786nijda6cr5evih1a2fn0n43ui6f12psk@4ax.com>
References : 1 2 3 4 5 6 7 8 9 10
User-Agent : ForteAgent/8.00.32.1272
On Wed, 6 Nov 2024 14:50:47 +0000,
liz@poppyrecords.invalid.invalid(Liz Tuddenham) wrote:
Bill Sloman <bill.sloman@ieee.org> wrote:
>
[...]
LTSpice - in the right hands -
can help you understand what's going on on the bench quite a lot faster
than bench work on it's own.
>
It can help you understand what *should* be going on but benchwork shows
you what is really going on and it is up to you to understand why.
learning by benchwork is slower because it is complicated by having to
deal with reality.
I'm now simulating a rugged transimpedance amp for capacitive oil
level measurement. That would be a mess to evaluate by breadboarding,
partly because the real thing needs ADCs and FPGAs and such, and it's
hard to breadboard those.
Once the sim is tuned, we'll go directly to product PCB layout without
breadboarding.
One nice thing about design by simulation is that we can do a
schematic and PCB layout and set up to build some rev A first
articles, and parallel all those time delays with code development, so
we can start testing as soon as the first units are built.
So the best breadboard is the first production unit.