Why do I have this sneaking suspicion that Amir isn't responding to this little thread diversion because he's waiting for us to catch up? : )
It was reasonably proposed that we simulate a MM output impedance from the analyzer output when measuring SINAD and possibly other phono preamp transfer function characteristics.
The first thing that came to mind was LTSpice and it appears there's no upside and a fair amount of downside to adding the reactive behavior of a MM cartridge to the generator output.
(Edit 28 Sept 2021: I'm not so sure there's no upside anymore. I've ordered an appropriate inductor to do some experimenting. TBA.)
Per the simulation plot below, the response for the lowest and highest unbalanced output impedances selectable in the APx analyzer (20Ω and 600Ω) make no difference. The typical input impedance of a phono preamp (the vertical parallel resistor and capacitor) is a non-player here.
OTOH, the simulation of the typical MM output impedance (the horizontal series resistor and inductor) makes a big difference. Now I know what MM cartridge designers are dealing with. For the cartridge to have a flat response to 20kHz or so, the magnetomechanical (!) output of the cartridge will have to approximate the inverse of that curve. No doubt the electrical curve was designed on purpose to deal with the realities of MM cartridge technology decades ago. Nevertheless, it's all baked into the design now just as moving mass, mechanical and electrical compliances are baked into loudspeaker driver design.
LTSpice Simulation
1. SINAD is measured by injecting a 1kHz tone (at 5mV for MM) into the DUT, then comparing that to the sum of all the other stuff that shows up in the bandwidth of interest (20-20k). With phono preamps, most of the offense will be noise to the left of 1kHz, while the 5mV signal is unlikely to generate much distortion to the right of 1kHz. Thus the HF rolloff you see in the MM cartridge output simulation is unlikely to materially move the SINAD needle.
2. Frequency response data using the MM cartridge simulation circuit would earn derision worldwide. That's fine if everyone else is wrong and we discovered the truth, but is this really the truth? Seems to me that the reactive interplay between cartridge and preamp is a design constant that can be ignored unless you get a preamp that has reactive adjustability. Then you might add the MM simulation circuit in addition to the standard measurement method.
3. SIY's method of performing noise measurements with a real cartridge attached to the input makes a lot of sense. It's possible that it won't move the needle enough to make it worth the hassle, such as shielding to prevent it from becoming an antenna in a busy lab environment.
4. All the published phono preamp specs use standard analyzer I/O's, so adopting a reactive method for signal based measurement is going to need justification. Even then you'd probably have to also provide industry standard measurements. The goal around here to help people make informed purchase decisions (or have fun pretending to), and that's tough if your measurements can't be compared to the mfg. data.
5. My interest in home audio began in 1983 when Sony released the CDP-101 for $900 with no possibility of discount. I still have the CD that came with it. In 2021 I'm learning about record players! Before long someone's gonna discover fire! : )
God bless you and your precious family - Langston
Edit: I screwed up the circuit! Something bothered me about the series RC phono preamp load. I looked at some schematics and sure enough it's supposed to be parallel. Thus I lied for about an hour and a half. Better than usual. : ) The simulation has been corrected, oddly the conclusions remain the same.
