Step Up Transformer Question

While the Epoch may certainly be superior sounding to the Koetsu, for a given listener, this may be the first time I ever read that someone thought a Koetsu cartridge, especially a "KRSP", was "too analytical". But then again, that was the guy who sells the Epoch. (Don't get me wrong, I love MI cartridges, and I would love to hear an Epoch.)

How are you reducing phono gain in the Bellari?

Alex, your opinion is your opinion, but it is not helpful if you don't provide more information, like what turns ratio, what cartridge with what internal resistance and voltage output, and what phono stage with how much gain? 

Not to make an already very confusing thread even more confusing, but Jensen make great  step up transformers, too, at a reasonable price.  And their staff of engineers are very knowledgeable and helpful to DIYers. Go for their top of the line, and you won't be sorry.

Dogberry, The information contained in the referenced thread from Steve Hoffman Forum echoes precisely what Atma-sphere and Jonathan Carr (two authoritative sources) have been writing on this forum for years, loading an MC (at any resistive value above some multiple of 10 or greater of the iinternal R of the cartridge) should make very little to no difference in frequency response. Yet, there are an endless number of posts stating otherwise. I suggest to anyone interested in this topic to search these archives for their words. Even below a load equal to 10X the internal R of the cartridge, where you start to get significant loss of gain, due to shunting of the cartridge output to ground, the gain penalty is very gradual as you go down from 10X to 1X (cartridge internal R = load R), where 50% is lost to ground but also there is a potential loss of HF response due to capacitance at the interface, and almost all phono stages have at least some input capacitance.  The input Z does affect the phono stage in ways described by Atma-sphere.

mdmoore, I advised against replacing a 12AX7 with a 12AU7 first and foremost because those two tubes are electrically very different from one another. Upon further thought, I do seem to recall that there is something out there where the manufacturer officially OKs such a tube swap. I can conceive of doing it if there is some sort of switch that changes the load resistor and other parameters, when the user makes such a drastic change in tube types. Is that so, in the case of your unit? Do I recall correctly that you have a Bellari? The second reason I reacted so negatively is that I and many other hobbyists do not much like the sound of a 12AU7, when it is used for gain, typically in a linestage. Using a 12AU7 in a phono gain stage seems impossible; you would get very little phono gain, nowhere near even the 40db typically associated with MM stages. So, to sum up, the 12AU7 would be current starved in a 12AX7 socket, would result in very inadequate gain even for an MM cartridge, and would sound bad even if changes were made to the circuit so the 12AU7 would run properly. Better candidates would be 5751 (check me on the numerical designation of this tube, which is a standard replacement for 12AX7 when users want slightly less gain), 12AT7 (which also would run better if the circuit for a 12AX7 was slightly modified), and those are your best bet. There are some others in the 12A?7 series, e.g., 12AY7 and 12AV7 that might work, especially 12AY7, but the circuit really ought to be re-worked for such swaps.

4 lines from the bottom of my post I wrote “in proportion to the square of the turns ratio of the cartridge “. I of course meant “in proportion to the square of the turns ratio of the SUT”.

What has always boggled my mind is that SUT manufacturers so rarely publish the turns ratio(s) of their products. Which would make it very simple indeed to at least begin to match a SUT to a particular cartridge. Instead, and especially in the golden era of vinyl, most SUTs were given a value in ohms (see Elliot’s compendium above). Typically, the value in ohms is meant to indicate the recommended internal impedance of the cartridge to which you want to match the SUT. For example a "3 ohm" SUT is meant to be used with cartridges having an internal impedance close to 3 ohms (probably 2 to 5 ohms cartridges would work). It also means that the turns ratio of that cartridge is such that when mated to that group of cartridges will result in an acceptable ratio between the internal R of the cartridge and the input impedance that will be seen by the cartridge when using that SUT hooked up to the standard MM stage with a 47K ohm input impedance.

The turns ratio is equal to the voltage step-up ratio. A SUT with a turns ratio of 1:10 will increase the output signal V of the cartridge by 10-fold. Etc. The resulting impedance seen by the cartridge will be equal to the input Z of the phono stage (47K ohms) divided by the square of the turns ratio (100, in this case). So the cartridge sees 470 ohms (47,000/100). Since you want the ratio of the internal R of the cartridge to be about 1/10 of the input impedance of the phono, you can see that any cartridge with an internal R of less than 47 can drive a 1:10 SUT. Nothing explodes if you break the rule of thumb. The actual loss of gain and high frequency response that occurs when the match gives you a ratio less than 1:10 is on a continuum, losses get gradually worse as the ratio approaches 1:1. At a ratio of 1:1, 50% of cartridge output goes to ground instead of driving the phono stage, and there is audible loss of HF response, due to input capacitance.

Back to our "3 ohm" SUT. A theoretical cartridge with an internal R of 3 ohms can drive a net 30 ohm load before you cross the line of the 1/10 rule of thumb. In order to create a 30 ohm load, you just have to work the formula for the load seen by the cartridge backward. I calculate 1:40. A cartridge with a 3 ohm internal R can drive a SUT with a step-ratio (voltage gain) of 1:40 into a 47K load, if you want to rigidly obey the 1/10 rule of thumb.

By the way also, an ideal SUT or any transformer does not have an impedance per se. Transformers merely reflect the impedances from secondaries to primary windings, or vice-versa. I wondered that Elliot mentioned there are SUTs that let you choose both voltage gain and impedance. That can only be done if the SUT has built in resistors typically across the secondary, which would be in parallel with the standard 47K load.

Whart, you wrote, "My dumbed down approach is to look at the internal impedance of the cartridge and multiply by 10 for the load seen by the cartridge. The secondary (or output) from the SUT will be 47k."  In your first sentence, I suppose you mean that you multiply by 10 to determine the (minimum) load that should be seen by the cartridge.  I agree. In your second sentence, you may be confusing others.  The secondary is connected to a 47K standard load at the input of an MM stage. The SUT acts on that impedance to reduce it as seen by the cartridge, in proportion to the square of the turns ratio of the cartridge.  I am guessing you know this but there is a semantic disconnect in what you wrote. I point it out only so others will not be confused.

Can you boil down your questions to two or three sentences ?

The only thing I’m sure of is DON’T replace a 12AX7 with a 12AU7.