I am planning on using a difference amplifier IC much like the AD8206 to measure the differential voltage across a high side shunt in order to monitor the current.

I chose a differential amplifier over a current sense amplifier (CSA) IC because its more robust to large input differences, less sensitive to input filtering and I don't have a need for high speed current measurement so the lowered bandwidth is ok for me.

  1. The issue with differential amplifiers verse CSAs is that they attenuate the signal before feeding it into its internal amplifiers. Is this an issue? I don't fully understand the ramification for input signal attenuation. I get that its a reduction in amplitude so possible resolution?
  2. The CMMR is worse in differential amplifiers than CSAs. But why do I care? I am struggling to grasp the importance of CMMR in application. I know that in an ideal world an op-amp only applies a gain the differential between the 2 inputs and ignores the common mode. From my reading I've found [common mode voltage = (V1 - V2) / 2]. So High CMRR rejects the gain associated with that common mode voltage? Can someone illuminate the application importance for CMMR?

Thank you for your time.


You can null out the CMRR error, once.

But if your high-voltage line or whatever you are monitoring, is bouncing around, the AC_CMRR becomes important.

So you should quantify the trash/hum/ripple/spikes on the signal you are monitoring, versus frequency. Then compare that to the CMRR VERSUS FREQUENCY curve of your circuit.

| improve this answer | |

Why would you not just use the AD8206? That part is tailor made for high side current shunt voltage drop measurements. The specifications look quite good to me for such an application.

At some loss of bandwidth due to added capacitance you could protect this part with a pair of back to back parallel Schottly diodes across the inputs and a 65V zener or other tranzorb device at one of the inputs to GND of the specifications are not robust enough for you.

| improve this answer | |
  • \$\begingroup\$ I am planning on just using the AD8206. But that was a decision based on several factors. One of the draw backs of the decision to use the AD8206 and what that it attenuates the input signal verses a CSA like the AD8210. I'm just trying to understand the cost associated with the signal attenuation. @Michael Karas \$\endgroup\$ – Tim51 Apr 9 at 2:38
  • \$\begingroup\$ The on chip resistor divider which is the attenuation to which you allude is exactly what gives this part the ability to support the 65V common mode range for high side current sensing. Despite that the part still offers an overall gain of 20. At some loss of overall efficiency in your overall system design you still have the option to increase the value of the shunt resistor if the gain of 20 does not satisfy your downwind A/D dynamic range. In a power supply application you can still put the feedback sensing down wind from the sense resistor so that it's voltage drop (continued) \$\endgroup\$ – Michael Karas Apr 9 at 3:56
  • \$\begingroup\$ (continued from above) does not introduce an error in the output regulated voltage. \$\endgroup\$ – Michael Karas Apr 9 at 3:56
  • \$\begingroup\$ Right and I understand that the attenuation provided chip resistor divider is what allows the difference amplifier in the IC to be tolerant to 65V common mode. If I understand you correctly, the cost of attenuation is a reduction in overall output gain. I just need to make sure that gain of 20 is acceptable for my A/Ds dynamic range and if not I can use the shunt resistor to increase the gain. Am I correct? @Michael Karas \$\endgroup\$ – Tim51 Apr 9 at 16:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.