Sorry about that Bear. A "6A4" is a 6 amp 400 volt diode.
In the particular situation that i'm using these in, the design calls for a "one way valve" in series with the modulated rail voltage feeding an rf output section. These voltages run up somewhere between 25 - 35 volts with an average running at about 30 volts. Obviously, this is WAY above the turn-on threshold of the diode itself. Believe it or not, the difference between one of these diodes with a .4 and a .7 voltage drop is VERY visible in terms of total output and gain characteristics. That is, if you have the right test equipment and know what to look for.
In effect, the differences between the two diodes (.3 volts out of an average of 30 ) is equivalent to 1/100th of the total power supplied to the output section. If we were talking about an audio component, that 1% difference would be equivalent to the line voltage falling from 120 volts down to 118.8, which is negligible to say the least. That same 1% gain in rectification efficiency results in an increase at output of somewhere around 15% - 18% on modulated peaks. I know that this does not sound "logical", but the rest of the circuit is so far "tweaked" that ANY change is highly noticeable. I don't know if we have audio designs that are this far "tweaked out" or running on the ragged edge.
Having seen the difference that such a small change can make and knowing that diodes produce TONS of "hash" and switching distortion, i would think that this could make a difference in a bridge. Using hand selected diodes with a quick turn on rate and fast recovery characteristics could further clean up / reduce the amount of "grundge" fed into the rest of the component and increase efficiency. Not only would the diodes turn on faster and stay on easier ( reduced switching due to the lower turn-on threshold ), ringing should also be minimized due to their quick recovery. Adding a "snubber" would be the icing on the cake.
While i've never tried doing this, i may have the opportunity to do such a thing very shortly. I've to 7 identical preamps that i'm "tweaking", so this may be a great opportunity to try something like this. I'll have to make note to order the necessary parts.
Obviously, the benefits of such a modification would strictly be limited by how well the rest of the component was designed and how revealing the rest of the system was. As such, i'd love to hear your thoughts or those of Ralph, Bob Crump or anyone else that could throw a wrench into the gears i've got turning in my head : ) Sean
>
In the particular situation that i'm using these in, the design calls for a "one way valve" in series with the modulated rail voltage feeding an rf output section. These voltages run up somewhere between 25 - 35 volts with an average running at about 30 volts. Obviously, this is WAY above the turn-on threshold of the diode itself. Believe it or not, the difference between one of these diodes with a .4 and a .7 voltage drop is VERY visible in terms of total output and gain characteristics. That is, if you have the right test equipment and know what to look for.
In effect, the differences between the two diodes (.3 volts out of an average of 30 ) is equivalent to 1/100th of the total power supplied to the output section. If we were talking about an audio component, that 1% difference would be equivalent to the line voltage falling from 120 volts down to 118.8, which is negligible to say the least. That same 1% gain in rectification efficiency results in an increase at output of somewhere around 15% - 18% on modulated peaks. I know that this does not sound "logical", but the rest of the circuit is so far "tweaked" that ANY change is highly noticeable. I don't know if we have audio designs that are this far "tweaked out" or running on the ragged edge.
Having seen the difference that such a small change can make and knowing that diodes produce TONS of "hash" and switching distortion, i would think that this could make a difference in a bridge. Using hand selected diodes with a quick turn on rate and fast recovery characteristics could further clean up / reduce the amount of "grundge" fed into the rest of the component and increase efficiency. Not only would the diodes turn on faster and stay on easier ( reduced switching due to the lower turn-on threshold ), ringing should also be minimized due to their quick recovery. Adding a "snubber" would be the icing on the cake.
While i've never tried doing this, i may have the opportunity to do such a thing very shortly. I've to 7 identical preamps that i'm "tweaking", so this may be a great opportunity to try something like this. I'll have to make note to order the necessary parts.
Obviously, the benefits of such a modification would strictly be limited by how well the rest of the component was designed and how revealing the rest of the system was. As such, i'd love to hear your thoughts or those of Ralph, Bob Crump or anyone else that could throw a wrench into the gears i've got turning in my head : ) Sean
>