When you want to do more...

LA 300B twin

Single-ended 300B with a 300B driver.


Background

LA 300B twin offers an important lesson learned. When I built my first 300B amplifier I thought it was fantastic, I had never heard anything like it. In time, however, I came to wonder why different 300B amplifiers sounded differently. I have heard amplifiers not letting the 300B direct heated triode perform at its best.

Obviously this depends on how the amplifier circuits are constructed and what objectives have been met. Circuit design, topology, physical realization and economics all contribute to the end result. One problem with the 300B, at least for those of us who try to design circuits for it, is that this tube is difficult to drive. Further, its amplification is low, which is why we need quite a large signal swing on the control grid. This puts quite some demand on the driver stage.

I thus wonder whether some of the differences between different 300B amplifiers could be explained by variations in driver stage performance. If true, my different listening impressions could possibly be accounted for by different driver stages, and not by the 300B tube itself.

This is the origin of this project. My intention has been to use the same kind of tube in the driver and output stages. As always, I want these stages to be connected via an interstage transformer. This should enable the output tube to deliver its true performance without being limited in any way by a hard working driver tube. Thus, each tube can develop its distortion signatures without disturbing the other.

The sonic performance of the amplifier indeed confirm these thoughts. The 300B presents itself in a way superseeding all my previous experience. It can now be clearly stated that the driver stage situation should indeed not be underestimated.

Construction

A development version of the amplifier is first built on a book shelf board. This greatly facilitates circuit changes and component replacements. When the circuit is reasonably finished, it is moved to the target chassis.

The chassis is built on an aluminum frame finished off with solid lacquered masur birch. All components are mounted on an aluminum sub-chassis, later covered by a solid 3 mm polished brass panel. The construction is very sturdy and carries all the weight with no hesitation. The extraordinary wood and sheet work is performed by my friend Eive Skoog.

Technical detail

The chassis is divided into two compartments. The rear compartment carries all components related to the power supplies, and the front compartment houses the signal components. Compartments are separated by an intermediate wall, also providing some electrostatic screening.

The amplifier is built around a line stage, a driver stage and an output stage. The line stage load is provided by a constant current source, and the signal is coupled to the driver via a very well sounding Russian paper in oil capacitor. The driver signal is then provided to the output stage grid via an interstage transformer.

Rectification for plate and bias voltages is provided by full wave vacuum rectifiers. Each signal amplification tube has its own power supply, and can thus work with no interdependencies. Each cathode heater, included the line stage indirect heater, is driven by an individual constant current supply.

For anyone interested, lots of detailed information and pictures are provided in the following discussion forum thread (text in Swedish):

Design: silentdesign.se