The Gain Cell

Join Our Community Subscribe to Paul's Posts

The critical nature of a preamplifier’s heart and soul, the volume control, has vexed designers concerned with sound quality for years. Time and again we face the dilemma posed by it: What’s the best device possible for the smallest degradation? Turns out, the answer to that is no device.

Instead of compromising audio quality by inserting an attenuator in a purely analog preamp, better results are obtained by eliminating the device altogether. Instead, it is possible to reduce a preamp’s essential elements from three to two by varying the gain of the output stage, rather than attenuating the signal into a fixed amplifier.

But how do you design a transparent sounding gain-variable output stage in a preamp? You turn to the recording industry who has wrestled with this problem for years.

In the heyday of analog mixing boards, (they’re mostly digital now) engineers faced the same problems as we in audio: How to remote control potentiometers without sacrificing audio quality. They first used motors. It didn’t take long before that solution became unwieldily so they turned to something else. A variable gain amplifier. Based on an older circuit design known as a Gilbert Cell, this unique amplifier topology uses multiple differential pairs in a balanced configuration and a voltage to vary the gain. According to Wikipedia, “The Gilbert cell was invented by Howard Jones in 1963 but usually attributed to Barrie Gilbert (before joining Analog Devices) in 1968.”

My first experience with this unique topology was gratifying. I had long wondered what sonic compromises might be attributed to this device and set about testing a number of them. At the time I was interested more in sound quality than functionality (they all worked well for gain setting—not all sounded good). I wanted to find a Gilbert Cell that had the sonic richness and transparency I insisted upon to go into one of our products. None lived up to my expectations until our chief engineer, Bob Stadtherr made a suggestion to me. “Maybe it’s not the cell but the way you’re using it.”

Aha! Of course. It took me six months of hard work to design a fully balanced input to output amplification stage whose gain could be varied by simply changing a voltage, but hard work was rewarded with amazing sonic performance.

And thus, the Gain Cell was born.

You’ll be reading much more about this sonic wonder in the next few weeks as we launch the Stellar Gain Cell DAC for beta testing next month.

Stay tuned.