0

So far I measured CMRR by first obtaining the open loop gain and exporting the wave, and then doing the same with the AC source in common mode, but this adds the need for additional software, and make the process difficult automatize.

I don't think there's any easy way, since the two test circuits are different (basically with the AC source connected as common or differential input), but I wondered if there is any HSPICE trick to make it work.

Ironil
  • 73
  • 8
  • 1
    A trick like this? http://www.electro-tech-online.com/threads/cmrr-in-lt-spice.141348/ – Voltage Spike Dec 28 '16 at 17:50
  • Yes, it's quite easier to do two copies of the circuit and operate than trying to set up complex algorithms. Maybe add it as an answer so I can accept and upvote? – Ironil Dec 28 '16 at 22:12
  • You can set the AC magnitudes to parameters and then rerun the simulation with .ALTER and .PARAM statements or the .STIM command. Two copies of the same circuit is undesirable for monte-carlo simulations, also it is more work and a source of human error. – HKOB Mar 23 '18 at 22:01

1 Answers1

1

I'm not very familiar with HSPICE (mainly LTSpice), but I think this can be done with .MEAS and .STEP commands, which I believe are common to both environments.

Basically, design a step command such that one run will configure the sources so as to measure differential performance, and the second run will configure them as common-mode.

Then, use a measure command to measure the output amplitudes of each run, and a second measure command to find the ratio of these two.

uint128_t
  • 8,664
  • 6
  • 26
  • 28
  • LTSpice has an example called LoopGain (and LoopGain2) which uses parametric sweep to enable AC sources and uses the @ operator to select outputs at various sweeps. Maybe this is also applicable in HSPICE. – Mike Dec 30 '17 at 16:12