Author Topic: Adding a differential dc bias to a fully differential amplifier  (Read 4468 times)

0 Members and 1 Guest are viewing this topic.

Offline electronxTopic starter

  • Regular Contributor
  • *
  • Posts: 144
  • Country: 00
Adding a differential dc bias to a fully differential amplifier
« on: February 07, 2024, 08:03:18 am »
How do I add dc bias to a fully differential amplifier? I cannot add it with Vocm or Vref pins because the receiver system also has a differential input and the offset common mode signal I add to the Vocm or Vref pins will be accepted and deleted.

In this case, should I add (DC bias /(2 x G)) to the non inverting terminal and -(DC bias /(2 x G)) to the inverting pin for the dc bias I want to add?
in god we trust
 

Offline jbb

  • Super Contributor
  • ***
  • Posts: 1223
  • Country: nz
Re: Adding a differential dc bias to a fully differential amplifier
« Reply #1 on: February 08, 2024, 12:35:38 am »
I guess so… you should be able to simulate the situation which helps.

One wrinkle which comes to mind is that the DC injection you want is differential mode, but feeding in a reference from a single-ended DAC (and other wire tied to ground) you’ll inject differential mode and undesired common mode. I think this would result in the input signal source seeing a (slightly?) changing common mode input equivalent circuit.

If you want to keep really stable common mode input performance you’d need a proper differential DC injection source. Which could be a normal single-ended DAC and an opamp configured for G=-1.
 

Offline electronxTopic starter

  • Regular Contributor
  • *
  • Posts: 144
  • Country: 00
Re: Adding a differential dc bias to a fully differential amplifier
« Reply #2 on: February 09, 2024, 11:14:40 am »

Actually, I can add dc bias this way, but I don't have a vref pin. In this case, this added offset will be added to the differential signal, but when we examine it on its own, it may create common mode noise and cause differential noise for the opamp.And also, this added DC reference will be constant and will not be multiplied by the gain. Maybe I can add a reference signal that is multiplied by the gain by adding gain to the Vref signal with an external opamp.
« Last Edit: February 09, 2024, 11:17:14 am by electronx »
in god we trust
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7158
  • Country: pl
Re: Adding a differential dc bias to a fully differential amplifier
« Reply #3 on: February 09, 2024, 02:28:38 pm »
If you have a pair of well matched resistors, you could connect them in parallel with R1,R3 to any differential DC voltage and it will simply be added to the signal.
Particularly so if R1~R4 are already discrete resistors, so not extraordinarily matched, and adding another pair of discretes won't compromise CMRR too badly.
« Last Edit: February 09, 2024, 02:31:15 pm by magic »
 

Offline avandalen

  • Contributor
  • Posts: 16
  • Country: nl
Re: Adding a differential dc bias to a fully differential amplifier
« Reply #4 on: February 12, 2024, 12:09:58 pm »
Maybe my article about this will help you. https://avdweb.nl/tech-tips/electronics/general-summing-amp
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf