I have experimented on this and found that there is a correlation, however I am skeptical on the 80% rule.
It depends on many factors, to me the correlation was also dependant on the Vce of the device and it's hfe, however I could not get consistent results to form a model but yes, I did liked the sound better with lower rails almost every time (but re-biased the output stage each time to optimal as reducing rails can change the idle bias as well).
Another twist, keep the amp far away from clipping and the difference will not be that clear, i.e. run a 50 watt amp at 5 watts and then reduce its rails, and again listen at 5 watts you will not be able to tell the difference, however test the amp at 50 watts, then reduce the rails 80% (reducing rails from 32v to about 26v) making it a 32 watt amp and now listen to it at 32 watts, it will sound better. (disregarding the SPL change).
i.e. At near clipping, a hi rail amplifier sounds more harsh than a low rail design.
There are many factors that would make this happen, I suspect that we can point out one culprit... beta droop.
Beta droop in the low rail version will be less, hence will have less distortion (perhaps sound better?)...
I am sure there are other reasons as well.
In practical terms many manufacturers made similar topologies and marketed them as different models e.g. Krell KSA-100 and Krell KSA-50, or Adcom GFA-535 and GFA-545 , in both cases, people liked the lower rail version better even though the design was nearly identical (except rails... and more OP devices used).
K-
It depends on many factors, to me the correlation was also dependant on the Vce of the device and it's hfe, however I could not get consistent results to form a model but yes, I did liked the sound better with lower rails almost every time (but re-biased the output stage each time to optimal as reducing rails can change the idle bias as well).
Another twist, keep the amp far away from clipping and the difference will not be that clear, i.e. run a 50 watt amp at 5 watts and then reduce its rails, and again listen at 5 watts you will not be able to tell the difference, however test the amp at 50 watts, then reduce the rails 80% (reducing rails from 32v to about 26v) making it a 32 watt amp and now listen to it at 32 watts, it will sound better. (disregarding the SPL change).
i.e. At near clipping, a hi rail amplifier sounds more harsh than a low rail design.
There are many factors that would make this happen, I suspect that we can point out one culprit... beta droop.
Beta droop in the low rail version will be less, hence will have less distortion (perhaps sound better?)...
I am sure there are other reasons as well.
In practical terms many manufacturers made similar topologies and marketed them as different models e.g. Krell KSA-100 and Krell KSA-50, or Adcom GFA-535 and GFA-545 , in both cases, people liked the lower rail version better even though the design was nearly identical (except rails... and more OP devices used).
K-
Hi,
could the sonic improvement be better ability to drive more difficult loads?
Dropping the Vrail by 20% allows a much wider phase angle to be driven before devices run into Tc and Tj problems.
I also tend to agree with Kamps. Lower output current cascades back through the stages as better gain and lower currents/higher impedance/lower distortion.
could the sonic improvement be better ability to drive more difficult loads?
Dropping the Vrail by 20% allows a much wider phase angle to be driven before devices run into Tc and Tj problems.
I also tend to agree with Kamps. Lower output current cascades back through the stages as better gain and lower currents/higher impedance/lower distortion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.