Understanding Record Repro

Fatmangolf wrote: This is very helpful and clear on the pitfalls of MC stages, as well as showing the expertise that makes the GSP/HFC ones sound so good! How dependent is the peak on the cartridge's characteristics and does increasing the load capacitor to say 100pF change the graph much please?

Adding capacitance brings the peak lower in frequency. Reducing the resistance damps the peak.

The two plots show with 100pf the resonant frequency is 2MHz, and with 2 x 100pf it is 1.6MHz. You won't get halving because the LC equation uses a square root.

By reducing the resistance the peak is damped which reduces its selectivity (a radio term).

Further increases in capacitance reduce the resonant frequency, and interestingly reduce the sharpness of the resonant frequency, but to smooth out the peak the resistance still has to be reduced.

You will note the output with 4n7 (4700pf) is linear below 100kHz with all values of resistance shown. This is getting on for 8 times the audible range in adults. However, there are many who swear they hear a tonal difference. There should be little phase difference and without working it out I guess it will be around 10 degrees.

You will also notice a difference in output level between 500 Ohms and 100 Ohms, and switching between these will give around 2-3dB difference in amplitude. People often say on switching from 100 Ohms to 500 Ohms that it sounds better, but it is only because the volume has increased.

Because the volume has increased there will be a tonal difference but not from the cartridge or amplifier. It is because of the way our ears work. See Fletcher–Munson curves.

What we really need is to switch the gain at the same time as we switch the resistance and then do AB-blind testing to find if there is any perceived difference.

This is a very interesting discussion - thanks for the eye (rather ear) opening details. A basic questionthat caught my mind with this discussion:

The Goldring Eroica LX has a similar specification and is modelled below into 100 Ohms and a 4,700pf capacitor complete with an approximation of coil interwinding capacitance.



By the way, the left hand scale is relative - I did not set the model to give the exact output which won't be +50dB at 1kHz as it shows - more like -64db, but it illustrates the rate of change.

I also need to state that there is no audio in the 20kHz plus region of this plot (because some idiots on another forum will think I'm saying that). Well, there will be harmonics whose outputs are decaying in level stretching some octave(s) up until the record output or stylus movement gives up the ghost.

The rest of the curve is simply stimulus - what the wiring picks up and is "allowed" to span to, because of the cartridge inductance (no matter how small - think radio inductors).

As the slope comes down in previous examples you'd think any higher frequencies would be severely attenuated so as not to interfere, but parasitics will always exist and each winding connects directly to the next by capacitive coupling, and here I have modelled it using the value stated for a metal film resistor (0.35pf) simply because there is no data for the actual cartridge - or any of the cartridges - let's face it, they're made for audio frequencies so no need... The same up-flick applies to all pevious models and you'll just have to imagine they're there (I'm not re-doing them).

And so we can refer back to the scope images and now realise why we could see interference up in this capacitive region, because there wasn't any short to prevent it. By properly shielding the arm tube (something the manufacturers seem ignorant of) the interference was reduced, and that kinda proves the lot, don't it?

Extract the integrator function out of the 'RIAA' phono stage EQ by making your mind take out the kinks in there for the response from the actual record, and you can see it takes the up-slope shown above and lifts the lower end so as to make the up-slope level.

When it meets the resonant peak at around 800kHz (in the above plot that is) it then follows the plot...

OK I'll do it for you. Take a look here:



See?



Edited by Graham Slee - 26 Sep 2017 at 5:57am

shashirao wrote: 1. For shielding the MC cartridge - my P******t Xtension setup does not have a headshell, how would the wiring update work? currently my cables inside my 'carbon' tonearm terminates into a 5pin DIN.