Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
There appears to be need for different ground references. One for input signals, another for output signals. The output stage has an important purpose. Its ground wires (OUT 1M, OUT 2P) are also marked FB GND. They are return paths for current to and from the device.
Therefore I think all wires marked FB GND are intended to be connected close together, probably by as short leads as possible.
Oddly, FB GND also appears (but is not highlighted) at
several chassis-ground symbols. These look like output
filter returns. Maybe the chassis is the prime signal ground.
Or maybe the schematic defeats itself, because "FB GND"
is also taking all the supply decoupling cap return currents
which might not be such a good thing if you wanted quiet.
Perhaps you might find some better insight on product
application notes pages for the component, if there are
any to be found.
Thank you.There is nothing about this case in datasheet .In the output stage GND is directly conected to FB GND;So maybe We can use GND as FB GND in DIY ;Don't you think so?
Throughout the schematic you have capacitors to ground (value 100 nF). This agrees with the usual guideline when working with IC's, to put a .1uF capacitor from sensitive pins to ground. The purpose is to reduce spikes, noise, oscillations, fluctuations, etc. Advice says to connect the capacitor leads as close physically to the IC as possible.
That is why I think all the FB GND lines should be connected together by short leads, while being kept separate from other ground networks (such as those for incoming signal and power supply). Furthermore it's probably a good idea to keep the power supply ground wires separate from incoming signal ground wires.
Decoupling always wants the current loop area minimized
(as this is L, which kills C at HF). But I think the question
is, is (or should be) the "FB GND" be "common" or have a
special treatment. To me the schematic seems untrustworthy
as far as communicating this intent. I do not know the part
at the heart of it all, to say whether any of the signal
path involved "FB GND" points ought to be segregated
from the power rail decoupling.
And I tend to doubt that decoupling as shown is not the
result of deep thought; making VDD-VSS decoupling
(where the shoot-through spikes should circulate) pinned
to GND in the middle, can push trash onto GND instead of
sending it back home neatly. Yes, you -can-. But is this
really handling the glitch energy impulses to best effect?
Would have to understand the guts of the part, to say.
For an output pulse or transition, where is the load return
point and the drive source supply? That's the loop that
wants shunted.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.