Skip to main content

Timeline for Differential Pairs and GND

Current License: CC BY-SA 4.0

9 events
when toggle format what by license comment
Jan 6, 2020 at 19:43 vote accept Kinka-Byo
Jan 6, 2020 at 19:26 answer added Circuit fantasist timeline score: 2
Jan 6, 2020 at 16:48 comment added Circuit fantasist @ Kinka-Byo, Very interesting question giving rise to interesting thoughts about the philosophy of differential pair biased by the side of sources. There is always a ground but it is the common point between the sources. At a differential mode it is really a "stiff" ground (constant voltage)... a kind of a virtual ground... but at a common-mode it is a "soft"... "movable" ground... i.e. there is not a ground:) And this is the clever trick in this configuration - there is a ground for the useful differential signal and there is no ground for the undesired common-mode signal... "dynamic ground".
Jan 6, 2020 at 16:38 comment added Kinka-Byo But how can the Mosfet read the input signal, if it is not referred to GND?
Jan 6, 2020 at 16:19 comment added Circuit fantasist Indeed, the biasing networks for the gates will solve the first problem (missing resistances between the gates and ground)...
Jan 6, 2020 at 16:15 answer added bobflux timeline score: 1
Jan 6, 2020 at 16:11 comment added Circuit fantasist A whole bunch of interesting questions that are difficult to answer in a single comment... Most importantly, circuits 2 and 4 will not work. They can work if you connect resistors (with high resistances) between gates and ground (try to imagine where currents flow and what voltages are in this configuration). To figure out why a common current source is connected between the sources and ground, change simultaneously both input voltages (the so-called "common mode") and observe how the collector voltages change towards the ground. Try to explain what the impact of this is in both cases above...
Jan 6, 2020 at 15:55 answer added DKNguyen timeline score: 0
Jan 6, 2020 at 15:42 history asked Kinka-Byo CC BY-SA 4.0