questions about tube rectifiers

[mountianjustice]

I have a 35w4 rectifier tube would there be any benefit to adding it onto the 32vdc plate power supply on a valvecaster with a 12au6 tube? Would it get a more tubey sound than it already has? I realize i'm already running dc but what happens when you rectify dc voltage? Could i boost my plate voltage even higher with this tube in the works somewhere? I have some knowledge of rectifiers ive built a full wave bridge rectifier before to go from ac to dc and I have used diodes as clippers but what all can a rectifier really do

[GibsonGM]

Hi Mountain...good question    

The short answer is no, you won't gain anything by doing this...you're already providing a DC supply, as you said.   "Rectifying DC" will just mean creating a voltage drop for no good purpose.     

Down the road, if you make a 'more traditional' tube circuit (running on a nice high B+...), that rectifier might come in handy, if you like vintage tones.     For the VC, you're kind of already getting out of it what you will...it's running very starved (those tubes want high voltage!), and you don't have many stages.   

People like tube rectifiers, in bigger projects, because they allow the voltage to SAG a bit when playing hard, causing a more compressed sound. You won't be able to benefit from that phenomena at so low a voltage....the VC is what it is...a nice OD, good way into the tube world, but trying to give it a tube recto is probably some real big overkill, IMHO   

A tube recto will do just what a silicon rectifier will, with the exception that its inherent limitations allow for more sag.   

[PRR]

I expect lower output for the same distortion due the the extra drop which Mike explained.

You have been listening to tube nuts. In a HIGH power amplifier, the tubes are run cold at idle and hot at full power (class B or AB). The change of current through a lossy rectifier causes "sag" related to signal level. Technically a flaw, but some over-drivers like the mushy effect.

However your ValveCaster is surely class A. Power demand is *constant* at all output levels. (The tube sucks more than it needs, steady, and diverts as much as needed to output.) Rectifier current is unchanging. No dynamic sag.

http://www.beavisaudio.com/projects/ValveCaster/MatsuminValveCaster.gif

This plan, fed 32V, obviously can not suck more than 0.46mA. Imagine the tubes are shorts and figure the current in R2 plus R3. Shorted tubes are not interesting, we need a "fair fight" between the tubes and their resistors, so the current is *about* half of the shorted-tube condition, say 0.25mA.

Now look at a 35W4 datasheet, such as GE 1960 35W4 (300KB PDF). Page 3 of GE 1960 shows the 35W4 voltage-drop for reasonable currents. At 100mA it is about 11V. At 10mA it is maybe 1.5V. 0.25mA is way too small to even squint, but is surely less than 0.5 Volts drop at 0.25mA. Interpolation suggests more like 0.05 Volts. A 0.05V, or even 0.5V, drop in a 32V supply is "no difference".

If you must fool with "rectifier sag" at ValveCaster scale, at least use a smaller tube. 6AL5 is a classic; however its drop looks like 0.1V at 0.25mA, which is still no-difference. 12AX7 can "be a rectifier", tie grids to plate. Its limits are low but far more than the Valvecaster needs at 32V. Its drop will still be mighty small. Ah! Tie grid to cathode. Drop may now be more like 5 Volts at 0.25mA, which is "something" at 32V (but still not much) (and still not proportional to signal).