Tube DIY Asylum Do It Yourself (DIY) paradise for tube and SET project builders.

---

[ Follow Ups ] Thread: [ Display  All  Email ] [ Tube DIY Asylum ]

---

[ Alert Moderator ] [ Edit ] [ Delete ]

---

Further thoughts

155.212.217.78

Posted by Ratspar Cheenk on November 2, 2006 at 13:21:34

In Reply to: Fascinating ... posted by mikeyb on November 2, 2006 at 10:27:47:

Here are some further thoughts.

The reverse current from the second section into the first that I mentioned yesterday does happen, but is basically irrelevant. This is due to reactive coupling between the two sections and should also occur in other types of multi-section filters.

My characterization of the first section as a flywheel or resonant tank is probably misleading. Because of the diodes, current can't flow backwards in L1 and resonance is impossible. To see this, it may be helpful to make a mechanical analogy. We can think of C1 as a balloon with separate inlet and output ports, the former having a one-way valve. Air flows into the balloon when the inlet pressure exceeds the balloon pressure. If we plug the outlet port and run the air pump, the balloon fills up to the peak pressure of the pump and then everything stops. There is no resonance.

The rate of discharge from C1 depends on the DC load current. Assuming that the discharge time is roughly the same with or without L1, the ripple voltage is primarily a function of the value of C1 and the magnitude of the load current. The difference between the two circuits is how the average voltage on C1 varies with load.

Going back to the balloon analogy, the actual effect of L1 is to add some momentum to the air flowing in the inlet port. Think of the pump imparting energy to a hammer, and the hammer driving the air into the balloon. This is what allows the balloon pressure to overshoot the pump pressure -- or the voltage on C1 to exceed the secondary voltage.

I'll need a different catchy name for this design technique.

Over the course of one cycle, on the average, the amount of charge that flows into C1 must equal the amount that flows out. That means that as the load current increases, the duration and peak amplitude of the charging current must increase as well. This increases the amount of energy storage in L1, which explains why the peak voltage on C1 actually goes up with increasing load current.

All of this leads to some general design rules.

An L1-C1 tuning frequency of around 200-240Hz seems to be about optimal but the exact tuning is not critical.

Larger inductor values and smaller capacitor values give lower peak charging currents but higher ripple voltage. A wide range of values will work in the circuit. The inductor should not be so large that its reactance becomes significant relative to the load resistance at several multiples of the powerline frequency.

Increased transformer resistance reduces the effectiveness of L1. Larger L1 values may work better for transformers with higher secondary resistance.

Ratspar

This post is made possible by the generous support of people like you and our sponsors:

Schiit Audio

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---