One of my first guitar technology projects is designing a small amplifier. This seems like a fairly logical thing, since I currently have a set of criteria and design goals which are not satisfied by the commercial offerings. Call it my own quest for tone. Supporting the certain features I want is really not such a great challenge. That's mostly cut and dried, and the process of working it all out it is the same as most of the things I design. Just add and remove features, shift stuff around so that it physically fits into the available space, looks aesthetically pleasing and generally works well. It's mostly a game of problem solving with a little bit of artistic sensibility thrown in for good style. I tend to think that I've gotten fairly good at this by now.
So, my design was going along quite well. I discussed it with a friend and he had some good suggestions which I factored in. For the physical design of the amplifier housing I worked a lot in 3D CAD to get something that seemed to look decent, fit all the parts and was small. (portability is one of my overall goals) I built the main part of the box and it turned out really nicely. But then I sat down and started prototyping the preamp, and that's where things started to become not so straight forward. I was out of the logical design process and into the subjective arty part, and that's when there didn't seem to be any right answers anymore. There are always multiple ways of solving technical design problems, but I've always considered an optimum design to be the simplest, most cost-effective and elegant. (elegance to me means that when you look at a solution, it is either devilishly simple or solves a problem in a unique and beautiful way) But when I started working on circuits to make great guitar tone I started to realize that I was getting into totally new territory. The measures of a good solution that I'm used to: cost, simplicity, and good (technical) performance were now overshadowed by some sort of unmeasurable thing: tone. Never before had electronics felt so much like writing music!
I know that I'm after some sort of magical, excellent guitar tone that fits into a small lightweight box. I also know that part of my goal is to not rehash the same old thing. So many boutique amp builders are using the same tubes and same circuit topologies that people have been using for generations. Especially with hand-wired tube circuits (which are inherently simple) it's unlikely that there's much difference between most models. And most people seem to be interested in making clones of classic designs. To me this is completely boring, and there is really no point to an exercise like this.
My main problem though is that I don't really know what tone I'm going for. I've heard a lot of excellent sounds on recordings, but I don't believe I've ever heard an amp in real life sound like that. Part of my problem is that my experience with amp sounds exists almost exclusively on recordings, as I haven't had an opportunity to use very many different kinds of amps for myself. I own a Fender Hotrod which I like. It only makes a small range of tones which seem good for rock and blues. I also have a Pod XT which I don't like. It attempts to model a wide range of classic amps and speakers, but I don't really like any of the sounds it makes. And I've heard my friend's Fender Performer which doesn't sound nearly as good as my Hotrod. I've never played a big Marshall stack or anything "vintage" so I'm somewhat at a loss in terms of comparison. A friend suggested I rent some different types of amps to try, which I might do.
After reading a lot of discussion boards and hearing a lot of people talk about guitar amps, it appears that the general consensus is that tube amplifiers are better. But tubes don't really fit into my design criteria of small and light. A preamp made with tubes can be pretty small, but a power amp needs power supply and speaker transformers. It also wastes a great deal of power and makes a lot of heat. I'm also skeptical as to whether tubes really make as much difference as people seem to think. So I set out to see.
A recent experiment that I did involved comparing my Fender Hotrod's tube power amp to my Rotel solid state hi-fi amp. In one test I used the Hotrod as normal and recorded the result with a mic placed in front of the speaker. In the second test I used my Rotel as the power amp instead of the tube amp in the Hotrod. The results surprised me. I figured that the tube amp would be warmer and silky smooth since this seems to be the general way people describe tube amps. (compare: sterile and brittle as words used to describe solid state amps) The results were the opposite. The tube amp sounded crispy and brittle and the solid state amp sounded warm. (and slightly muffled) By using the Match EQ plugin in Logic Pro I was able to make the Rotel sound like the tube amp with EQ alone by letting the computer match the timbre of both amps.
Ok, so then I started thinking that maybe the magical part was the preamp. The Hotrod has a tube preamp with overall low gain. It's not suited for hard rock, and in fact the "More Gain" setting, which increases the gain of the main drive stage, actually sounds like crap and I rarely use it. But for bluesy and rock stuff the overdrive sound is really nice. I put the preamp output (using the effect loop on the amp) into a DI and fed it into my mixer. I was really surprised that the sound I was getting was really shrill and buzzy. Obviously the natural high frequency rolloff from the speaker was quite critical to getting a good sound. That seemed fair, so I opened the datasheet for the speaker I'm looking at using in my amplifier (the Eminence Lil' Texas) and had a look at the frequency response plot. Using the quite capable digital EQ in my mixer I set up the major parts of the EQ curve using the DI channel's EQ. I used high pass and low pass filters to set the overall usable frequency range, and then used the two parametric mid bands to accentuate the two major response peaks visible in the datasheet. The result was quite astounding! It went from sounding horrible and buzzy to sounding quite like a real guitar amp. Adding a bit of reverb made it really nice and I ended up jamming out with headphones on for several hours with my Les Paul playing through this setup.
This weekend I got my old electronics lab equipment out of storage and set up a little workbench at home. Since starting my job and building a new lab at work I've not been doing electronics at home. I got everything set up and dusted off my old analog scope, which is simple and refreshing to use after spending a lot of time with an expensive digital scope at work. My first project was a very simple distortion circuit using an op amp with diodes in the feedback path to make some soft clipping. I made a recording with my DI directly out of the preamp out on my Hotrod using the drive channel. Then I made a recording with the DI from the output of my circuit. In both cases I used the EQ curve that simulates a speaker which I had previously tweaked and saved. The results sounded surprisingly similar. The op amp circuit was a bit more shrill in the high end, but once again after using the Match EQ to adjust the timbre (it mostly decided to take down some of the highs) my circuit sounded convincingly like the tube overdrive in my amp. And amazingly other than the input impedance converter, it only had one stage of actual audio processing.
So by this point I was starting to wonder if all this tube stuff was a big joke. Certainly I took the bait and bought an expensive tube amp that's too big to carry around. But I was starting to wonder if the allure of glowing tubes is attracting people more than the sound. (which appears to be quite similar to simple solid state circuits) As an aside I once heard a player say that tubes made a warmer sound LITERALLY because they were hot inside! It was a friend of my dad's so I kept my mouth shut and laughed inside. Mind you I must add that if tubes, glass, retro technology, glowing filaments and so on make people feel better about their music making and feel more comfortable then I'd say it's as good a reason as any to use them. Just like wearing a comfortable pair of jeans or something like that. However the point of my quest is for tone, which has nothing to do with feeling good for a completely unrelated reason.
After some more reading on the web I came across a lot of people stating that the real tone doesn't come from the preamp but from the power amp overloading. This didn't seem to make sense to me at all. The preamp is designed with a few gain stages that are purposely meant to distort, giving that bluesy overdrive. In the case of high gain amps, more gain stages are strung together to give more distortion. Now I'm finding out that what people really love isn't that sound at all, but the sound of the actual power amp overloading. And there are actually products designed for loading down the power amp and putting a small amount of that power into the speaker so that you can have cranked tone at low volumes. It all seems quite wasteful in terms of power and especially wear on the tubes and other components. But this bit of knowledge was somewhat of a blow to my idea of making a lightweight amp, since I really couldn't afford to put a tube power amp inside it and was hoping to use a solid state amp, leaving the tone to the preamp and speaker itself.
Since I've never had a chance to crank up my Hotrod (due to living in a condo) I figured that until I took my amp somewhere that I could do that, I should at least see what the power amp distortion looked like on a scope, and what it sounded like. I know that the speaker's reactance affects things, and the fact that a speaker has a wide range of impedances almost all of which are much higher than 8 ohms. This reality definitely affects how an amp overloads because if a speaker doesn't have the same impedance at every frequency, the load on the amp is going to vary a lot depending on the sound being played.
Keeping the above facts in mind, I set out to do a fairly simple test. Although I didn't have any way to simulate an actual speaker's reactance at my disposal I settled for a bunch of power resistors which I configured to be around 8 ohms, with enough power handling to not burn out with the full output of my 40W tube amp being soaked up. I tapped off a signal from the resistors, which I fed into a DI for monitoring and into my scope for viewing. Once again (as with all of the other tests so far) the results surprised me! The over easy soft-clipping effect I was expecting didn't occur. Instead the amp appeared to clip quite like a class AB solid state amplifier. The waves were mostly squared off. I saw a little bit of the tube saturation that is quite visible in tube preamp circuits. But mostly it looked like square waves. And it didn't sound all that great either. Interesting was the fact that despite being a push-pull amp, the clipping was asymmetrical. The bottom of the wave clipped about 2V before the top. Certainly it's possible that there was some loss of symmetry in the single-ended tube preamp stages, or the phase inverter. Another interesting thing I noticed was that with heavy overloads there existed a lot of crossover distortion. I was really surprised by this because this is universally considered to be a sonically bad thing, although it is mostly considered a fault of bad push-pull transistor amplifiers and I've never heard anyone talk about tube amps doing that.
It seems that we learn the most when we make mistakes, and also when seek out to prove things we assume to be true. Certainly the latter has been the case for me in my quest for tone and understanding of guitar amps. All of the magical things about tubes that I've been told appear not to be true at all. I learned that preamps made with tubes or op amps can be made to sound pretty much the same with a bit of equalization. And I also learned that apparently power amp distortion is more important than preamp distortion, but found that unlike tube preamp distortion, (which really does have a characteristic sound) tube power amp distortion looks surprisingly like solid state power amp clipping.
Obviously I have only scratched the surface, and my experiments so far are incomplete and not perfect. But so far I've been surprised at every turn when the results turned out not only differently than I expected, but usually completely the opposite. I expect that more experiments will yield even more surprising results.
Go to the Projects Page