the circuit design
The first block is the input section. It is not bypassable. It consists of a single 12AX7 stage, very similar to the first stage of a Marshall, except that there is an optional negative feedback (NFB) path, controlled by a switch on the back panel. When NFB is activated, gain is lower but linearity is higher: i.e., the sound is cleaner, and less colouration is added. With NFB off, the stage adds more tube colouration to the sound, even when the overall tone is "clean". The tone will be more similar to a conventional Fender or Marshall, which use no NFB except in the power amp.
Next in the signal chain are the two bypassable gain sections, the linear preamp and the M-section. These are controlled by the two-button footswitch, or by toggle switches on the back panel. They are independently switched, so either gain section can be active by itself, or they can both operate in cascade ("pre" into "M"). If both gain sections are off, then the master volume is the only control that functions, and the amp becomes a very simple and "pure" configuration, just the input stage straight into the power amp, similar to early (pre-1960s) tube amps such as the Fender Tweeds.
The linear preamp is a relatively low-gain section, intended to act as a "boost" ahead of the M-section. Like the input section, it has switchable NFB, so that if desired, it can provide a very clean boost with almost no colouration added, or with NFB off, it can contribute some non-linearity and crunch. The linear preamp includes controls for the "formant", which is just a variable-frequency midrange boost or cut. This can be used to provide a mid boost similar to many overdrive pedals such as the "Tube Screamer", or it can produce a mid scoop, for a Fender-like tone. In the middle, frequency response is actually flat, which is not a response curve that most guitar amps can do!
The M-section is so-called, if not already obvious, because it is pretty much a direct copy of the heart of the Marshall preamp, specifically the 2203/2204. Active by itself, it's like the low-gain input on one of these amps. With the linear preamp also active, it becomes more like the high-gain input of a 2203 with a boost pedal in front. There are a few changes from Marshall, however. For one thing, I added a "coldness" switch to the cold-clipper stage. Instead of Marshall's 10k resistor here, the choices are 39k (like Soldano), or 8k. Generally speaking, the gain is higher in the 8k position, but there are other subtle differences to the tone and both are interesting settings, for different applications. Also, I eliminated the cathode-follower stage which usually drives the tone-stack in Marshalls. This was just out of necessity: I felt the stage was needed more for the effects loop; but I was pleased to find that there was almost no perceptible difference in the tone (I did a direct A/B test). (Perhaps it would make more difference inside an actual Marshall amp, but to a first approximation, the downstream loading should be the same, so I'd be inclined to try it if I had one to play with (and then use the extra stage for More Gain, of course!).)
Overall, the frequency response of the linear preamp section is broad, with plenty of bass and mids, in contrast to the M-section. So switching between these sections is almost like changing to a completely different amp, sonically. The linear preamp sounds full and warm, whereas the M-section sounds scooped and sharp-edged.
After the gain sections comes the FX loop. This section can be completely bypassed via a switch on the back (at least in my first prototype, it's not controllable via the footswitch, but I may add this feature on the manufactured version). As mentioned, I did considerable circuit-shuffling in order to free up a tube stage to use in this section. Originally, I had planned to just have a fully-passive loop, i.e., simple "pre out" and "pwr in" jacks with no gain or attenuation. However, given the high signal levels of tube circuitry, this would have meant that an external active device would have been needed, akin to the "Dumbleator" for Dumble amps. I tried it without, and there was no way it was going to cut it, with a normal line-level solid-state device in the loop (such as my reverb unit). It's an unavoidable necessity, to attenuate the signal going out of the "send" jack, and then to correspondingly amplify the signal coming from the "return" jack. So to make the loop usable with only one tube stage, there are some sacrifices necessary. The loop is not truly "transparent" and "gainless" as it should be. For one thing, it is inverting. So flipping the bypass switch between "in" and "out" (even with nothing plugged into the jacks) changes the tone: not tremendously, but perceptibly. Which is why I'm glad to have the switch. (Actually, being able to flip the phase at this point in the circuit is interesting in its own right, especially given the single-ended power amp and the cold-clipper preamp. Theoretically, there might be a much more prominent difference in tone when flipping the loop-bypass switch at loud volume levels, due to the asymmetrical waveforms. This is perhaps a matter for future research, sometime when I won't disturb family members and pets!)
After the FX loop comes the master volume control, and the power amp section. This is a single-ended EL84, cathode-biased, with no negative feedback. The Fender Champ does have NFB in its power amp, as does Marshall; so in terms of being an EL84 with no NFB, this amp has that bit of commonality with the Vox AC15 and AC30, but really it's quite different from any of these other amps.
The built-in 6-inch speaker can be bypassed by plugging in to the "ext spkr" jack. Driving other types of cabinets such as 4x12 should be interesting and is likely to provide a very different range of tones from this amp; but I haven't tried it yet.
The Jensen Mod 6-15 speaker is rated for 15 WRMS, much more than this amp can produce. However, the Hammond 1750AX output transformer is only rated for 3W, and this under-rating is intentional, to enable a bit of core-saturation at high volume levels. Because of the under-stressed operation of this amp, with lower voltages and intentional power supply "sag" designed in, I'm not sure it's even hitting 3W of output, although it's plenty loud and definitely sounds like more than 1W. (I'll make better measurements when I get back to my "real workshop".)
Comments
Post a Comment