On Thursday, December 20, 2018 at 11:33:27 AM UTC-5, Ralph Mowery wrote:
>
> This is one of he dumbest statements I have ever heard about how a
> capacitor works in a power supply. In the normal tube or simple diode
> supply the capacitor never discharges anywhere near 100 %
>
Let's go back to first principles.
Back when AK designed their power-supply, they used 2uF paper caps. Which did discharge to nearly zero at each cycle. And, they designed "onward" accordingly. And they warned specifically about increasing capacitance. And they used other means to clean AC out of the line, typically by way of a massive choke.
Not so much time later (about 4 years, in their case), they moved onto almost exclusively electrolytic caps for filtering, typically at about 10uF, whereupon their field coils got much smaller (same size speaker) and they stopped warning about increased capacitance. In most cases, they still used the 4-pin 80 tube as their rectifier. Rated at about 10uF. Note that in the Handbook, the 80 will accept much higher capacitance values "if voltage and current are controlled" - which is entirely the point.
Now, taking what you state - a larger capacitor will draw a larger charge. True *upon first charge*. After which, it will cycle just as with any other cap based on downline draw. If the downline draw is not excessive, no mater the size of the capacitor, current through the transformer will not be excessive - assuming voltage and current are stable.
Now, moving into semi-modern times - one of the most popular hacks for the venerable Dynaco ST70 is to greatly increase capacitance - some times as much as 200%. The theory behind it is to increase peak response and reduce clipping. That particular unit uses the 5AR4/GZ34 rectifier, nominally rated at 60uF, Power-supplies using this rectifier using 4-section caps totalling as much as 320uF are common. The downline draw is not changed on-average, the output transformers prevent any stray DC from reaching the speakers, and there is some momentary additional peak-response capacity.
A note here from a discussion on chokes vs. caps vs. input size vs. output side and so forth:
Probably because the d.c. output voltage from the same transformer-rectifier system will be higher when a capacitor is added in front of a choke-input filter. This neglects transformer heating, which is higher for the same d.c. power output with capacitor input than it is with choke input, even though the output voltage is lower with choke input.
Leading to the point on voltages.
This is straying away from the original discussion on modifying a vintage radio. But, there it is.