K&K Audio / Lundahl Transformers Welcome! Need support, you got it. Or share youe ideas and experiences.

---

[ Follow Ups ] Thread: [ Display  All  Email ] [ K&K Audio / Lundahl Transformers ]

---

[ Alert Moderator ]

---

Mo' Thoughts

Posted by Lynn Olson (M) on December 5, 2003 at 15:50:06

In Reply to: Assorted Thoughts posted by Lynn Olson on December 5, 2003 at 14:21:20:

Some data from John Atwood (my predecessor as Tech Editor at Vacuum Tube Valley, and now head of One Electron) indicates that a current source enforcing balance on a PP output stage can raise 3rd harmonic while reducing even harmonic magnitudes to near-zero.

My guess for what's going on is that with a current source in operation, the PP stages are effectively in series; while with a low-impedance cathode return, they are in parallel. In Class A, this should make little difference, except for the effect on instantaneous output impedance.

If you were to plot instantaneous output Z versus output voltage (as measured at the speaker terminals), for series connected tubes I'd expect to see a plot in the shape of a U-shaped valley. With two devices connected in series, the higher-value one is going to dominate, especially as one power tube heads towards cutoff. For example, as you approach the cutoff condition, one tube might look like 3K, while the other looks like 200 ohms. In a series circuit, the high value will dominate, resulting in 3.2K. So as the circuit as a whole approaches saturation, the impedance rises, and it eventually cuts off completely. It behaves exactly like two SE amps connected in bridge-mode. All of the SE 2nd harmonic properties are translated into 3rd harmonic, since both SE amps are symmetric, but driven in opposite phase.

But things are completely different with a parallel circuit (conventional PP in Class A). One tube looks like 200 ohms, while the other is 3K, but 3K in parallel with 200 ohms is 187.5 ohms. If you plotted this out, you'd have a more or less flat-topped plateau, with sides that sloped downwards (not up) on either side. However, the flatness of the plateau might be critically dependent of the bias value - I can easily imagine little ripples around the center and the sides as the standing current is moved up and down.

Interestingly enough, in a preamp, input, or driver, this variation in impedance is not that important, since the load impedance is much higher than the source impedance. But in the final power stage, the load is our old friend, the loudspeaker, which is intrinsically reactive, and even transmits resonant back-EMF energy back to the amplifier. This is where the output impedance characteristics start to make a difference, especially the instantaneous impedance as the AC cycle is traversed.

One little challenge I've thrown out to Gary Pimm is devise a simple circuit that will make a plot just like I've described - instantaneous output Z versus plus/minus volts, with adjustable gain in the vertical axis.

A tester like this should instantly reveal the true operating mode of any amplifier - SE, PP, transistor, bridge-mode, Class A, Class AB, feedback, no feedback. All feedback would do is flatten the curve, but it could do nothing to change its shape - that is set by the topology and standing currents of the output stage.

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

Parts Connexion

---

[ 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 ]

---