# Filter capacity in 2A3 amps

Is there any reason you can't use high filter capacity in the 2A3 amp power supply ? I notice that some of the 2A3 amps I have seen use motor run capacitors (50mfd 370vac) in the power supply instead of,lets say 500 MFD filter capacitors...How come the ac caps ?? I have been thinking of building a 2A3 set amplifier but I don't understand the filter system they use...I would like any help that is offered....THANKS
autospec
2 responses
 02-24-2006 5:04amIn a 2A3 amp, the mains voltage is stepped up to a range of 250 to 350 volts, which means a load current in the milliamps to provide the output power of about 2 to 5 watts. In solid state amps, the voltage is stepped down to about 30 volts and the current output can be 20 or so amps. The difference is the rating of the diodes and the rest of the circuit components - especially the voltage rating of the filter capacitors.For a 2A3 amp, you want the ripple current to be about 5 to 20 percent of the mains voltage for a time period of 0.01 seconds. Cranking out the math yields capacitors in the 50 to 80uF range. You can go higher in capacitance, but you run into the problem of maximum ripple current. Even at 50uF, or so, ripple current can be more than 10 times the load current. Using higher uF caps will result in that much higher ripple current, forcing you into higher current diodes. You also have to keep an eye on transformer winding resistance, internal wire sizes, and core saturation - all of which limit the ripple current.If you go to a higher value cap, the voltage rating must be the same - which will limit you in sizes. If you're really hell-bent on increasing the capacitance, you can place caps in series to the required voltage rating. Now the problem is leakage current which requires additional circuitry to help the caps operate at a constant voltage. Now the power supply becomes messy - not to mention the space required for the extra caps.And after all that: the load current cannot change (load resistance is the same and the power dissipation of the tube is still what it was) so you do not gain anything but expense. 02-28-2006 4:55amThanks Gs5556.....Thanks for the detailed description of the problem and solution....Autospec