![[-]](https://theultimatetone.com/images/collapse.png "[-]")
01-05-2020, 06:04 PM
Hi makinrose
Most microphones are low-z and have very low noise as a result. To match this to a tube circuit without compromising the overall noise usually requires a step-up transformer. The Tx provides "free gain" and gets the signal up to where the tube's noise is less problematic and the overall circuit can actually compete with SS performance.
The cathode of the tube can be used as an input with the grid at AC ground. The tube exhibits the same gain this way but z-in is now much lower allowing a direct connection to a low-z source. The tube's noise can be lower this way but you now have to manage the microphone connection so there is no DC through it and low-frequency response is not impaired. The latter requires the use of a high-value cap, preferably polypropylene. Fortunately you can get some seriously high values on PP now and they are not too large.
An alternative to using the cap is to use a split rail supply for the tube and add a DC servo circuit to minimise the DC offset. This involves an opamp and maybe a level shift transistor, none of which is in the forward signal path.
With a grid-driven interface, there is the opportunity to have variable gain for the first stage, where this is difficult to do with the cathode input interface.
Of course, any circuit optimised for low-THD should sound like any other, but that is not always the case. Such optimisation with tubes means you are essentially trying to rid the signal of any tube character which leads to the question: why use a tube then? I would suggest building both circuit types and evaluate them for yourself.
Most microphones are low-z and have very low noise as a result. To match this to a tube circuit without compromising the overall noise usually requires a step-up transformer. The Tx provides "free gain" and gets the signal up to where the tube's noise is less problematic and the overall circuit can actually compete with SS performance.
The cathode of the tube can be used as an input with the grid at AC ground. The tube exhibits the same gain this way but z-in is now much lower allowing a direct connection to a low-z source. The tube's noise can be lower this way but you now have to manage the microphone connection so there is no DC through it and low-frequency response is not impaired. The latter requires the use of a high-value cap, preferably polypropylene. Fortunately you can get some seriously high values on PP now and they are not too large.
An alternative to using the cap is to use a split rail supply for the tube and add a DC servo circuit to minimise the DC offset. This involves an opamp and maybe a level shift transistor, none of which is in the forward signal path.
With a grid-driven interface, there is the opportunity to have variable gain for the first stage, where this is difficult to do with the cathode input interface.
Of course, any circuit optimised for low-THD should sound like any other, but that is not always the case. Such optimisation with tubes means you are essentially trying to rid the signal of any tube character which leads to the question: why use a tube then? I would suggest building both circuit types and evaluate them for yourself.
