I will be building a zen balanced line stage using a uC and relay attenuator, and I was wondering about the trade-offs between attenuating the input vs. the output.
It seems to me that the output network uses some very small value resistors for the low volume settings to keep the output Z low, but has the benefit of reducing the noise of the circuit, as described in the original article.
On the input side I was considering using a matched Nchannel JFET buffer between the major (10db steps) and minor (1.25db steps) attenuators, keeping isolation and preserving the loading impedance so one doesn't affect the other. The problem there is no output noise attenuation.
I have also considered splitting the major and minor steps- minor on input and major on the output (or the opposite).
Anyone have any comments or suggestions?
It seems to me that the output network uses some very small value resistors for the low volume settings to keep the output Z low, but has the benefit of reducing the noise of the circuit, as described in the original article.
On the input side I was considering using a matched Nchannel JFET buffer between the major (10db steps) and minor (1.25db steps) attenuators, keeping isolation and preserving the loading impedance so one doesn't affect the other. The problem there is no output noise attenuation.
I have also considered splitting the major and minor steps- minor on input and major on the output (or the opposite).
Anyone have any comments or suggestions?
Steve, where in USA???
Well, the JFET input may keep the noise level down if you attenuate at the input. Recently I have seen a few threads relative to this subject(volume, gain, levels etc. May have been X or SOZ topologies). A search would be fruitful I think...
I have a question though. Is'nt there a "gain" control between a diff pair of sources in the design you have in mind. What is wrong with using the gain control as the level adjust??? There will be changes in measured performance by using this component that way but, I don't think they will be a showstopper. You might therfore want the ideal input level with the level control at the most typical setting to acheive close to optimum performace at typical listening levels. Just some idea's.
Well, the JFET input may keep the noise level down if you attenuate at the input. Recently I have seen a few threads relative to this subject(volume, gain, levels etc. May have been X or SOZ topologies). A search would be fruitful I think...
I have a question though. Is'nt there a "gain" control between a diff pair of sources in the design you have in mind. What is wrong with using the gain control as the level adjust??? There will be changes in measured performance by using this component that way but, I don't think they will be a showstopper. You might therfore want the ideal input level with the level control at the most typical setting to acheive close to optimum performace at typical listening levels. Just some idea's.
I'm near Akron / Youngstown Ohio.
I plan on implementing an "X" topology, so the gain resistor is out. I'm not sure that it's the best idea anyway because distortion performance would change with gain settings. With more feedback at lower gain settings you would get more upper harmonics with less lowers to mask them, and it might start to sound thin and brittle.
I'll give the search a try, but there are just SO many threads and I'm feeling lazy.
Thanks for the ideas
I plan on implementing an "X" topology, so the gain resistor is out. I'm not sure that it's the best idea anyway because distortion performance would change with gain settings. With more feedback at lower gain settings you would get more upper harmonics with less lowers to mask them, and it might start to sound thin and brittle.
I'll give the search a try, but there are just SO many threads and I'm feeling lazy.
Thanks for the ideas
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.