[Boatanchors] 5AR4 Rectifier

[Drew Papanek]

James A. (Andy)  Moorer wrote:

>I guess I would have to respectfully disagree - the 5U4 filament pulls 3
>amps whereas the GZ34 filament only pulls 2 amps. That extra amp can heat 
>up
>the power transformer a bit (5 watts to be precise). Most transformers will
>tolerate 50% more current, but on others it will shorten the transformer
>life and can even drip nasty black stuff out the bottom.

>I suggest replacing it with 1N4007 diodes in series with a 50-ohm 20-watt
>resistor. That gives just about exactly the same V-I curve. Some folks skip
>the resistor, but that makes the voltage a bit high.

The voltage would get quite a bit high very early on.

Solid state rectifiers come up instantaneously, directly heated cathode 
rectifiers such as the 5U4 come up in a few seconds as opposed to the 30 
seconds or so for the other tubes in the equipment.
During that time the B+ might exceed the surge rating for the filter caps.

The 5AR4 was introduced because B+ voltages in Hi-Fi equipment were up to 
about the maximum practical value for electrolytic capacitors (without 
resorting to series capacitor stacks). It has an indirectly heated cathode 
to delay application of B+ to around the time the other tubes warm up.  At 
that time the other tubes draw current and keep the voltage from soaring.

Make sure your filter caps are capable of withstanding the extra 
voltage-that includes filters further down the line in multiple section RC 
filters as are common in tube-type stereo amps.  The  capacitor at the far 
end might be rated for 200 volts and normally see only  about 150, but with 
rectifiers that come up fast might see a surge of nearly 600v.

But then, the total resistance back to the rectifier at that point might be 
a couple hundred K and the cap would then break down in a "gentle" 
(current-limited) fashion which does it no harm, much like a zener diode 
would or like the process of reforming electrolytics with a series 
resistor....

Drew