EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Atomillo on April 04, 2021, 11:20:25 am
-
Hello to all:
I'm recently designing a CSA for a X-Ray fluorescence project. I decided to simulate a circuit that has already been employed for a similar result (https://physicsopenlab.org/2017/06/22/x-ray-spectroscopy-with-pin-photodiode/ (https://physicsopenlab.org/2017/06/22/x-ray-spectroscopy-with-pin-photodiode/)) using OrCAD (because it's the program I'm using at uni). I asked for a Noise simulation with the current source I1 as the I/V source, and the Out node as the output.
Because I'm not interested in the original noise spectrum, but instead in what results after passing the signal through a low-pass and high-pass filter with the same time constant (1us), I take the value provided by NTOTAL (ONOISE) (which is in V^2/Hz, right?) and multiply it by the module squared of the resulting transfer function (tau*w)^2/(1+((tau*w)^2))^2 using the following command:
PWR(2*pi*Frequency*PWR(10,-6),2)/PWR((1+PWR(2*pi*Frequency*PWR(10,-6),2)),2)
IfI understood this correctly, I now have the spectrum that results from the pulse shaping network in V^2/Hz units. I now take the square root of this to get a V/Hz spectrum, and integrate this to get the total RMS noise.
The question: where did I screw up? (Because of my inexperience with both noise design and OrCAD, it's fair to assume I *did* screw up even if I don't know where!)
Many thanks!