MC and Step-Up Cartridge Loading

Typical beginner's question, when you start to learn more than you wanted to know: I seem to have a fairly bad loading mismatch between my cartridge (Benz Micro MC20E2L) and transformer (MA Cotter Mk2-P). I had an Accuphase AC-2 which wore out, and I replaced it with the Benz about a month ago, before I had learned enough about matching and specs. To make a long story short, the Cotter has a gain of 16.8:1 or 24.5db. Working through the transformer math on Vinylengine, this translates to a "natural impedance" of 167, while the Benz wants a load of 400-720 (i.e. >400, and using the Vinylengine formula, an upper of 720). At this point my options seem to be either get a different step-up transformer (Denon has a couple of 10:1 transformers that translate to 470 ohm loading) or get a different cartridge. So my questions are: first, if I get a transformer with the correct loading, what change should I hear? And would you go the transformer route, which is actually less expensive than a new cartridge, or would you go for one of the less expensive high output MCs and eliminate the step-up problem completely? Please keep cost effectiveness in mind, since in the current economy I'm really trying to buy more LPs and fewer bits of hardware! Thanks in advance for your advice.
You have 2 issues, gain and impedance. If you lower the gain to 10:1 to raise the impedance you may not end up with enough gain.

The 167 ohm assumes that the MM input has an impedance 47,000 ohms, which would be typical. This is usually controlled by a single resistor on the input of the stage so the simplest thing would be to change the value of the resistor. A 120K ohm resistor would give you a reflected (natural) value of about 425.

It is proportional to the square of the turns ratio so 16.8 times 16.8 is about 283.

167 times 283 = 47,000
425 times 283 = 120,000
I agree with Herman's comments, but I'll add a couple of things:

1)The main effect of resistive loading of a moving coil cartridge is to smooth out what would otherwise be an ultrasonic peak in the frequency response:

With excessive loading (167 ohms instead of the recommended 400+), the frequency response may roll off within the audio range (perhaps somewhere between 10 and 20 kHz), instead of above the audio range. It's not possible to say in a more quantitative manner without knowing the cartridge inductance, which doesn't appear to be specified. Perhaps you could contact the manufacturer and ask them what the inductance is, and plug that value into the calculator at the link above.

2)If your preamp input impedance is fixed at 47K ohms (some preamps have selectable input impedance, sometimes including 100K), another approach might be to have someone prepare a small shielded enclosure with rca female connectors at each end, and containing a roughly 73K good quality resistor, in series with the hot lead of each channel. 73K in series with the 47K input impedance of the preamp = 120K.

That would avoid any internal modification of your preamp, as well as the possibility of any possible subtle sonic effects that might result from changing its internal input resistor (e.g., on leakage or bias currents into its input stage).

As Herman mentioned, however, you might possibly still face the problem of inadequate gain (resulting in excess hiss), depending on how good the signal-to-noise ratio of your preamp is.

-- Al
First, thanks to both Herman and Almarg. With regards to gain, the MC20E2L puts out .5mv, so a 1:10 step up would get me to 5mv, a good value. In fact, the 1:16.8 of the Cotter is giving me 8.4mv, which is close to the overload on my ARC SP-8, which specifies a max phono input of 9mv. One thing I wasn't clear about, if I make the resistor mod(s) you suggest, does that also have the effect of lowering the input voltage to the phono input circuit? ( I was a math major, not an electrical engineer, so my ignorance shows - my guess is that the resistor will turn the "extra" voltage into heat, and have the effect of lowering the voltage to the phono input, but you are teaching me, and I know that guesses will get me in trouble!) Thanks again.
Changing the internal 47K resistor in the preamp to 120K will not affect the gain significantly, but has the two potential disadvantages that I mentioned (internal preamp modification; possible subtle electrical effects).

Putting an external 73K or so resistor in series avoids those disadvantages, but will reduce the signal level into the preamp by 8.1 db, corresponding to 20 x log (47/120). That would cause you to turn up the volume control by 8.1 db to get the volume you would have gotten without the series resistor, which in turn would raise the background hiss level by 8.1 db.

That may or may not be significant, depending on the signal-to-noise ratio of the preamp, which you can roughly judge by seeing what the hiss levels are that you presently have, and what they become if you turn up the volume moderately. Keep in mind the rough rule of thumb that a 10 db increase in volume corresponds to a doubling of subjectively perceived loudness.

For a "beginner," you ask remarkably intelligent questions!

-- Al
According to the specs shown here:

the phono input overload point is 900 mv, not 9 mv. (Of course, it's safe to assume that output clipping would occur well before that large an input were reached, given the 60db gain from phono input to main output, which is a voltage gain of 1000!).

The reference also indicates that "Magnetic phono may have any value from 10 ohms to 100K ohms substituted." Not sure how that is done physically, but see if you can determine that via the manual or by opening the top cover. 100K would cause the cartridge to see 354 ohms through your present transformer, which I'd imagine is close enough to the recommended range to be satisfactory.

-- Al
I agree with Al's math but my gut (+ experience) tells me that the 2nd approach of a series R is the more evil of the 2 since by forcing this tiny signal to flow through a series R you are dividing the incoming voltage by 2:3 after stepping it up 17:1 = about 11:1.

At that point I think an xformer with the 10:1 ratio is a better option.

If your gain structure is good with the 17:1 then I would definitely try changing the internal R to a larger value. It says 100K max but I doubt 120K would affect anything other than the input impedance and (as Al pointed out) the bigger the input R the more potential for noise.

A good MM cartridge is also a good option or high output MC designed to work into 47K. Did you buy this preamp new? All of this assumes that somebody along the line didn't change the value of the input R to something other than 47K.

If your gain structure is good with the 17:1 then I would definitely try changing the internal R to a larger value.

I agree. I made the suggestion about an external resistor before finding (if I correctly interpret the statement in the specs I linked to) that the resistor value can be changed without doing soldering work on an internal pc board, and before seeing the statement that 100K is electrically acceptable.

One minor correction to your post: Inserting 73K in series with a 47K input impedance would divide the 16.8:1 ratio down to 6.6:1, not 11:1.

-- Al
sorry, did it in my head and flipped the ratio over

Once again, thanks to Al and Herman for your kind tutoring! To put this to bed for now, I'm going to do a few things. First, the SP-8 Rev2 pre-amp is supposed to have special "under the circuit board" soldering posts (i.e. accessible from the bottom cover) for both reducing output gain for amp matching, and for upping the phono input to 100K, so I'm going to see if my ARC dealer can do this (while I practice soldering on something less valuable...). I am the original owner, so I know nothing has been changed from factory specs to date. Secondly, it turns out that the Cotter transformer can be "restrapped" to a lower step-up ratio which will increase its loading, again a bit of soldering that I will ask my engineering friend to do (and count this as a soldering tutorial). Lastly, I confess to having been fooled by the overload specs for the ARC. The extensive tutorial on step up transformers on Vinylengine states that the usual input range for MM phono circuits in 2.5-10mv, with a more usual max around 7.5-8mv. So when I read the 900 mv overload in the ARC specs I thought there was a problem with decimal points or units, but now understanding that input voltage will vary with frequency, I see that the discrepancy was that while cartridge manufacturers specify output for a specific frequency, ARC gives a total overload number which covers all frequencies. All said, I will still be more comfortable with a "standardized" input of closer to 5mv (10:1 step up) than the higher number I'm getting from the higher step up with my current strapping. Since both of you have been so kind, I wanted to give some "closure."

Thanks for the good update and summary. Yes, cartridge output specs are usually defined for 5 cm/sec groove velocities at 1 kHz, and the RIAA equalization that is utilized on records provides a boost of around 20db at 20 kHz, relative to that 1 kHz reference frequency (and also a cut of about 20 db at 20 Hz). The preamp provides an inverse response curve during playback. 20 db is a factor of 10 in voltage terms, so a cartridge with a 5 mv output at 1 kHz would output 50 mv at 20 kHz, for the same groove velocity. Of course, real-world recordings are unlikely to have that kind of content, at least for any significant amount of time; if they did tweeter meltdowns would be a common occurrence!.

I would think, also, that worst case groove velocities can exceed the 5 cm/sec reference velocity that is the basis for these specs, although I don't know to what degree.

My other comment, which you most probably realize, is that if you make BOTH changes, restrapping the xfmr to 10:1, and also changing the preamp to a 100K input impedance, the cartridge will see a 1K load, which goes too far toward the other extreme.


-- Al