vakilia
Junior Member
Offline
Posts: 15
Atlanta, GA
|
Hello;
It is mostly recommended to use differential input-output configuration to avoid common mode noises and that's why in the simulation time if we look at each output (eg. in a Diff. I/O OTA), it might have much more noise including switching noise and clock feedthrough (if the circuit is sampled data one) rather than the time that we look at the difference of the outputs. However we should be able to test a real circuit in the same manner. I mean we should be able to connect the spectrum analyzer (S.A.) differentially to the outputs of the circuit (if we consider the circuit can drive two 50ohm load and the S.A. has only one input channel) and it is like connecting one of the output to the ground of the S.A. with a 50ohm resistance and the other one to the input channel of the S.A., the process that I did and I couldn't get good results because it is not a good idea to connect a wide spread ground of the S.A. to the output of a circuit generally, as I guess. Maybe there is a S.A. that has two channels but as I know they can not show the difference of the two channel to be able to simulate the differential output mathematically in the S.A.
Do you have any idea that how the people test the real circuits differentially?
Thanks Babal
|