There are a lot of mysterious components within any guitar amp, many of which remain puzzling even to hobbyists who have built a DIY project or two. In the sixth part of Mojotone’s series 'What Does This Thing Do?'...
...We’re examining cathode-bypass caps and their impact on any guitar amplifier’s sound.
We’ve already discussed filter capacitors and coupling capacitors in parts 1 and 4 of “What Does This Thing Do?” so this brings us to the third significant role that capacitors et al have in shaping the sound of your guitar amp. Cathode-bypass capacitors (“caps” for short) are found connected in parallel to the cathode-bias resistors in many preamp stages and some output stages, and to many beginner and novice amp enthusiasts they may well be the most mysterious of all three capacitor usages within the guitar-amp circuit. Even so, at times they can have the most dramatic effect on your overall tone, making a significant impact both on gain and voicing.
Cathode-bypass duties are usually performed by filter (aka “electrolytic”) capacitors, albeit usually much smaller ones than you’ll find in an amp’s power supply because these don’t need to handle as high a voltage, and like those larger filter caps no actual signal passes through them from one point to another. Which is to say, they aren’t “coupling together” signal-treating stages of the amplifier—gain stages or tone stages—the way coupling capacitors are. They’re not working like conventional filter caps either, however, and these parts are used because the same value of filter capacitor is usually available in a much smaller physical size than would be obtainable in a coupling capacitor, which otherwise could perform the same task (more of which below).
In brief, and without getting too technical...
...cathode-bypass caps are used to counteract some of the frequency and gain-dampening tendencies induced by the cathode-bias resistors they are coupled with. Both preamp and output tubes need to be biased, and while a resistor is used for the technique in the output stages of only a specific class of amplifier (the eponymous cathode-biased amps, often colloquially referred to as “Class A”), almost all preamp tubes in amps from the late ’50s onward are cathode-biased.
This biasing is achieved by connecting a resistor of a specific value between the preamp tube’s cathode and a ground point (the value of that resistor is determined according to how “hot” the circuit designer wants that tube to run in that stage, relative to operating voltages and in terms of gain factors and so forth). Connecting a resistor to the tube’s cathode, however, also induces a form of negative feedback in that tube, which tamps down the gain and flattens the frequency response. The way to achieve satisfactory biasing while avoiding those tone-dulling side effects is to also connect a capacitor to that cathode, which runs to ground in parallel with the resistor, essentially bypassing the resistor’s own cathode-to-ground connection and emphasizing certain frequencies in the process.
Such cathode-bypass caps almost universally increase the gain of the tube stage in which they are used...
...while also revoicing the stage according to their value (which is to say, the “size” of their capacitance capabilities). The rule of thumb regarding the latter—when considered on a gradient of sorts—is that caps of a larger value emphasize bass response more, while those of a lesser value emphasize treble. They don’t constrict frequencies as completely as coupling capacitors of different values do, but help to emphasize certain bands within the overall response, while enhancing the gain in the process.
For example, the standard guitar-amp preamp gain stage today usually uses a 25µF bypass cap connected to the cathode of one half of a 12AX7. The legendary tweed Bassman used a 250µF cap in that position, and that’s what Marshall copied to build its JTM45. But as the Marshall design evolved toward the Plexi models of the mid to late ’60s, they split the first two gain stages to create independent Normal (bassier) and High Treble (brighter) channels, using a 320µF cathode-bypass cap in the former and a 0.68µF cap in the latter. That crispy, sizzly, crackly high-frequency breakup that is a big part of the cranked-Plexi tone is due in large part to that bypass cap, which is of a significantly lower value than the one in the other channel, or in most other amps’ preamp stages for that matter. The difference in capacitor values across the bassier-to-brighter spectrum doesn’t have to be quite as dramatic as the difference displayed in that example, although small changes (say, swapping for a cap that’s only double or half the current value) often don’t produce a noticeable change in voicing.
We use the term “value” to denote these capacitors’ ratings because...
...while amp-speak sometimes refers to them in terms of “smaller or larger,” that can also be misleading. Often the physical sizes of these components—especially when we’re not comparing like for like, or parts of the same make—have no direct bearing on their capacitance rating. For one thing, a capacitor’s physical size is partly a function of the amount of voltage it needs to handle, and the 25µF filter caps commonly used in this way in preamp stages are generally rated at just 50v or even 25v. That’s why they’re a lot smaller than the 400v to 600v filter caps used in your power supply, which might also be in the 22µf to 30µf range as regards capacitance.
Look at photos of the circuit of a late-’60s or early-’70s Marshall 50W Lead or 100W Super Lead, and you’ll see the disparity in sizes of capacitors used in cathode-bypass positions for the Normal and High Treble channel, which are located at the far-right end of the circuit board when the rear of the chassis is facing downward. The 320µF cap in the bassier channel has a value 470.6 times higher than the 0.68µf cap in the brighter channel, but the latter part is generally two or three times the physical size of the former. That’s because a non-directional (non-polarized) 0.68µF coupling capacitor is used for cathode-bypass duties in the High Treble channel, and coupling capacitors are physically a lot larger than electrolytic capacitors in the first place, when comparing parts of the same voltage rating. In theory, we could also use coupling capacitors when a bassier voicing is required, but such a part would likely be too large to fit into position on the circuit board (and high-quality coupling capacitors of that size and value are also expensive, and hard to come by).
We have focused mostly on the preamp stage here...
...but cathode-bypass caps are also used in—unsurprisingly—amps with cathode-biased output stages. In the general sense, they perform similar voicing duties there, but swapping from one value to another in this position usually has a less audible effect overall, and the tonal difference might be negligible under most operating conditions.
When considering preamp stages, however, you can use your understanding of cathode-bypass caps to better voice a DIY project or amp mod to your own tastes, to produce differences between two otherwise similar channels, or to create gain-boost options through extremely simple changes in the circuit. That’s a lot of variation available from changing just one relatively small component!