Load and cable capacitance will interact with the inductance of the cartridge's coils to produce a resonant peak in frequency response, at upper treble or ultrasonic frequencies. Increased capacitance will raise the amplitude and lower the frequency of that peak.
MM cartridges often count on that peak to compensate for what would otherwise be an upper treble rolloff, at frequencies that are within the audible (<20 kHz) part of the spectrum.
BTW, that is the exact opposite of the effect of capacitance in line-level interconnects, where too much capacitance will roll off the upper treble. In that situation, the cable capacitance is interacting with the output impedance of the component driving the cable, which is essentially resistive and not inductive.
Best regards,
-- Al |
From the things I have read, and the cart designer I have talked to on the subject of that resonant peak in frequency response, that peak is waaaaaaay outside the audible band. T_bone, That's true in the case of MC's, because of their much lower inductance (which results primarily from the smaller number of turns in their coils). But in the case of MM's, with more turns and much higher inductance, the peak occurs at much lower frequencies. See the following: http://www.hagtech.com/loading.htmlYour comments about the possible effects on phono stages of the peak that MC's can produce, at very high ultrasonic frequencies, are quite correct btw. Best regards, -- Al |
If the inductance estimates are correct, it means one wants to have capacitance as low as possible in order to preserve the harmonics 'air' which extend beyond the audible band. It would be difficult to imagine why one would want to bring that mountain closer to the audible frequency. No, not for MM's. As I indicated in my previous post, if the frequency of that mountain is brought down such that it occurs just a little above where the cartridge would otherwise be rolling off, it will compensate for that rolloff within the audible range, and have the effect of extending the overall treble response. Lower capacitance will move the peak to a point where it will no longer provide that compensation, and where the response will already have rolled off significantly, resulting in a less extended treble. As an aside, I am looking at the formula just above the MM FrR calculator and I don't understand how a high SQRT(LxC) in the denominator would lower any ratio with a 1 in the numerator. As with any fraction, if the denominator increases and the numerator stays the same, the resultant value goes down. The presence of the square root function simply slows down the rate of decrease as LC increases. Best regards, -- Al |
T-bone, if we use units of henries for L, and units of farads for C (which results in F being in units of hertz), then both L and C will be far less than 1. In fact C will be on the order of 0.0000000001 farads (or 100 pf). Therefore we will end up with a number in the denominator that is much smaller than 1, with the numerator being equal to 1. Therefore F, in units of Hertz, will be a number that is much larger than 1.
As for Farads x Henries resulting in frequency, after application of the square root and reciprocal functions and some constants, yes that is not intuitively obvious. But the derivation is as follows:
Resonance occurs at the frequency at which the magnitude of the impedance of L becomes equal to the magnitude of the impedance of C. Since the impedances of L and C have opposite polarities in the complex plane (don't ask!), the equal and opposite magnitudes will cancel each other out at that frequency, resulting in a net impedance of zero (apart from the resistance that is present).
The magnitude of the impedance of an inductor, aka its inductive reactance, is measured in ohms and is equal to 2 x pi x F x L.
The magnitude of the impedance of a capacitor, aka its capacitive reactance, is also measured in ohms and is equal to 1/(2 x pi x F x C).
So to find the resonant frequency we set those two formulas equal to each other, and solve for F.
2 x pi x F x L = 1/(2 x pi x F x C).
Re-arranging that equation:
1 = 2 x pi x F x L x 2 x pi x F x C
Therefore F x F = 1/(2 x pi x L x 2 x pi x C)
Therefore F = 1/(2 x pi x (sqrtLC))
Voila!
Best regards,
-- Al |
T_bone & Lew, thanks very much for the nice words.
JB, in my experience MM cartridge specs do usually indicate recommended load capacitance.
The 47K number, btw, is not capacitance, it is the recommended load RESISTANCE (47,000 ohms) for many MM's, and approximately corresponds to the input resistance of many or most phono stages or preamp phono inputs, that are intended for use with MM's.
Recommended load capacitance for MM's is typically in the area of 100 to 400 pf (picofarads). The cabling will usually be a more significant contributor to that than the input capacitance of the phono stage or preamp phono input, if the phono stage or preamp does not have a provision to switch in additional capacitance.
As you may realize, capacitance can be added by soldering an appropriate capacitor to an rca plug, and connecting it at the phono stage or preamp phono input with a y-adapter. I have a commercially made kit that I purchased in the 1980's which provides a selection of rca plugs with capacitors of various values soldered onto them, but I haven't seen such a kit offered in recent times.
Best regards,
-- Al
|
Thanks, Stan. I just checked and sure enough my "Phono Equalization Kit" was made by DB Systems.
It provides polystyrene capacitors, mounted on rca plugs, with values of 100, 150, 200, 300, and 400 pf; 100 ohm 1% metal film resistors mounted on rca plugs for use with MC's; a pair of spare plugs; and a pair of y-adapters. The part number of the kit was DBP-6.
The instructions also made reference to a DBP-11 "Capacitance Loading Switchbox," which could accept inputs from two turntables, provide independently selectable capacitive loading of each cartridge, and allow the user to select which cartridge's outputs are routed to the phono stage.
Best regards,
-- Al |
|
