High Impedance cartridges and SUT's
Posted: Thu Jul 28, 2022 3:20 pm
I have noticed a trend lately of people asking for high ratio SUTs for higher impedance cartridges. The current request was for a 1:50 for a 22Ω cartridge and the following is why I think that is a bad idea. There are two different places where you can get into trouble here and they both have to do with the reflected load. Lets start out with the ideal with a mythical 22Ω cartridge feeding a 1:20 (green) 1:40 (Blue) and a 1:60 (red) into an open secondary where the full expected gain of the SUT will be seen.
The gains are the expected 26dB, 32dB and 35.6dB and this is about as far as the average SUT user gets. It is critical to dig a bit deeper. All MM phono stages have a 47kΩ input resistor and when we add this to the mix the problem of the reflected resistive load becomes apparent. The addition of the 47kΩ resistive load causes all three of the outputs to drop with the 1:60 actually giving less output than the 1:40! This all makes perfect sense when you consider that the 1:60 will reflect back a 13Ω load to the 22Ω cartridge which quickly explains where all the gain goes. For a more thorough explanation have a look at: SUT Design .
The simple solution to this is to understand that the typical 47kΩ input resistor is there for MM cartridges and can only be a liability for a properly wound SUT. Simply increasing the value to 300kΩ goes a long way to get closer to the gain predicted by the turns ratio. Unfortunately the reflected resistive load is only 1/2 of the battle and we also need to look at the reflected capacitive load. The input of a typical mm phono stage also has a fair bit of capacitance in order to deal with resonant behavior at the top of the audio band. This too is a liability for a SUT and the capacitance (both cable and phono input) should be kept to an absolute minimum. The next set of plots shows what happens when we simply add 100pF of capacitance to the 300KΩ resistive load. It is the reflected capacitance that has a severe impact on the high frequency extension. This is what limits my suggested maximum step up ratio to 1:20 for this case. .
To finally complete things, lets look at this situation for a typical MM phono input of 47kΩ + 200pF of input+cable capacitance. In this 'worst case' scenario the 1:20 is -1dB @ 50kHz which is right at the edge of what I would call acceptable.
dave
.The gains are the expected 26dB, 32dB and 35.6dB and this is about as far as the average SUT user gets. It is critical to dig a bit deeper. All MM phono stages have a 47kΩ input resistor and when we add this to the mix the problem of the reflected resistive load becomes apparent. The addition of the 47kΩ resistive load causes all three of the outputs to drop with the 1:60 actually giving less output than the 1:40! This all makes perfect sense when you consider that the 1:60 will reflect back a 13Ω load to the 22Ω cartridge which quickly explains where all the gain goes. For a more thorough explanation have a look at: SUT Design .
The simple solution to this is to understand that the typical 47kΩ input resistor is there for MM cartridges and can only be a liability for a properly wound SUT. Simply increasing the value to 300kΩ goes a long way to get closer to the gain predicted by the turns ratio. Unfortunately the reflected resistive load is only 1/2 of the battle and we also need to look at the reflected capacitive load. The input of a typical mm phono stage also has a fair bit of capacitance in order to deal with resonant behavior at the top of the audio band. This too is a liability for a SUT and the capacitance (both cable and phono input) should be kept to an absolute minimum. The next set of plots shows what happens when we simply add 100pF of capacitance to the 300KΩ resistive load. It is the reflected capacitance that has a severe impact on the high frequency extension. This is what limits my suggested maximum step up ratio to 1:20 for this case. .
To finally complete things, lets look at this situation for a typical MM phono input of 47kΩ + 200pF of input+cable capacitance. In this 'worst case' scenario the 1:20 is -1dB @ 50kHz which is right at the edge of what I would call acceptable.
dave