Wurlitzer 120/700 Amplifier: Circuit Analysis
The Wurlitzer 120 amplifier represented a total redesign from the previous Wurlitzer model, the 112. The upside is that the amp is a lot smaller, so it fits nicely in the smaller, lighter 120 cabinet. The downside is also that the amp is a lot smaller. It’s missing many of the features that make the Wurlitzer 112 so magical. Regardless, it’s still a great vintage tube amp with a lovely, classic tone.
Features of the Wurlitzer 120
Push-pull output stage. The Wurlitzer 120 is a push-pull, or class AB, amplifier. It uses two 6v6 power tubes, each one tasked with amplifying half the signal. The alternative to a push-pull amp is a single-ended design, such as the single-tube Gibson GA-5.
Here’s how it works. Amplifiers typically have two main sections: a preamp to amplify the signal, and a power amp to drive the speaker. It’s possible to use just one output tube to drive the speaker, but amplifying the entire waveform is so much work that much of the tube’s effort will be lost as heat. For that reason, it’s much more efficient to use two tubes: one to amplify the crest of the waveform, and the other to amplify the valley. In that scenario, each tube can essentially take a break and stop conducting during half of the signal.
How does a tube know which half of the signal it should amplify? Easy. In the previous stage, called the phase inverter, the signal is split into two, each one 180 degrees out of phase with the other. One signal is sent to the grid of one power tube, the inverted, out-of-phase signal is sent to the other, and together the tubes reproduce the two signals as one.
A push-pull configuration is more linear, more efficient, and inherently less noisy than a single-ended design. This is likely why the 120, like all Wurlitzer amplifiers, uses a push-pull output circuit. Wurlitzer intended for their keyboards to sound realistic, and a push-pull circuit is a tried-and-true way to achieve clean tone.
Unfortunately, push-pull designs can be more expensive. First of all, they require a phase inverter, which uses at least one half of a triode and all necessary components. They also - obviously - require twice the tubes as your typical single-ended output, as well as a bigger chassis to mount everything on.
Wurlitzer was a budget manufacturer: everything they made was solid but not particularly high-end. They were unpretentious and marketed all their keyboards, whether acoustic or electronic, to consumers who couldn’t afford or didn’t have the space for an enormous, more traditional piano. The product was always quality - they didn’t exactly cut corners - but, given the opportunity to save on the manufacturing cost of something, they often wouldn’t hesitate to take advantage of it.
The push-pull design of the Wurlitzer 120 is a superb illustration of that. Wurlitzer would never throw their hi-fi principles (along with one of those push-pull power tubes) out the window, instead designing (for instance) the Champ of on-board electronic piano amplifiers. But they would absolutely cram everything on a chassis the size of a postage-stamp and design the most bare-bones phase inverter possible.
Which brings us to our next point.
Paraphase inverter. The whole preamp is just a phase inverter.
Usually, an amplifier will have several gain stages in order to incrementally amplify the signal before sending it to the output section. This is a clean way to achieve the amount of gain that you're looking for - whatever that may be - because you can give each stage a moderate amount of bias that avoids unwanted distortion. If you actually want distortion - as in, overdrive distortion, like a high-gain guitar amp - you can use your multiple gain stages to dial it in precisely, while using attenuators and filters between stages to roll off parts of the signal you don't want. If you tried to get all of your gain in too few stages, you’ll get unwanted distortion, an amp that gets way too hot, and probably still not as much power as you would have liked.
The 120 amplifier has just one gain stage, and it’s not even a stand-alone gain stage. Technically, it’s part of the phase inverter. Its output goes in two directions: it feeds the grid of one output tube, but it also sends an attenuated portion of the signal into a second gain stage that inverts the signal, re-amplifies it, and directs it to the second output tube. This is called a paraphase inverter.
(It’s fine: it gets the job done. We’re just salty because, unlike the 112, the 120 has no extra tubes to hijack for an fx loop.)
How a paraphase inverter works.
Each power tube needs its own signal, and both signals must be 180 degrees out of phase with the other. There are many ways to achieve two out-of-phase signals, and the paraphase inverter goes with one of the more obvious ones by exploiting the natural phase inversion of a conventional gain stage. Any time you apply a signal to a tube grid, that signal appears 180 degrees out of phase at the plate. So, if you send a signal through two cascaded gain stages, you can achieve two outputs, each 180 degrees out of phase with the other.
The problem is that a gain stage will obviously make the signal more powerful, which is undesirable here. We want the power tubes to have pretty much identical signals, not one medium-sized signal and one huge signal. So, before sending the signal from Triode 1 to Triode 2, we have to significantly attenuate it. This way, after Triode 2 is done amplifying, its output will be around the same size as Triode 1’s. The 120 accomplishes this with a 470k/6.8k voltage divider positioned between its two gain stages.
Aggressive bias throughout the circuit. Bias is determined by the network of resistors that is positioned around the tube stage: most importantly, the plate resistor and the cathode resistor. This determines the gain of the stage and also its headroom: basically, whether the signal will be faithfully reproduced or clipped at the output.
The 112 had gain to spare: in fact, its designers clearly tried to dial some of it back with conservative bias and low-gain 12AU7 tubes. For the 120, Wurlitzer went a bit too far in the opposite direction. They eliminated a tube, which removed some gain stages, but then they biased everything super hot to compensate.
Notably, the 6v6 cathode resistor is 125 ohms. This is a very low value, which runs the tubes very hard: in fact, in excess of their maximum plate dissipation. If you have a 700, you should consider increasing the value of this resistor to prolong the life of your tubes and prevent damage to your amplifier if/when they fail. (A well-placed internal fuse - rarely found in amps of this period - can also mitigate damage caused by power tube failure. In case of an output tube short, the mains fuse won’t blow fast enough to protect output components.)
The preamp tube is also aggressively biased. It has 470k plate resistors: clearly an attempt to squeeze everything possible out of the amp’s single gain stage. (Plate resistors in typical amp circuits are closer to 100k. The larger the plate resistor, the more voltage swing the amp is capable of.) For the most headroom, supply voltages should be high: otherwise, you’re likely to hear distortion. In the published schematic, the supply voltage is 250v, which is rather low. The good news is that because wall voltages are higher today than they were in the 1950s, your 120 will almost certainly have higher supply voltages, and therefore more headroom than it did when it was originally manufactured. (Higher supply voltage is, however, another reason to upgrade that 125 ohm 6v6 cathode resistor: it is driving the tube much harder than its designers even intended.)
Another consequence of a large plate resistor: it can sometimes diminish high frequencies. Early 120 circuits had a relatively large bright cap across the volume pot. For some reason, this was removed in later versions of the schematic. You can experiment with adding it back in.
Preamp tube cathode circuit. On the preamp tubes, the cathode resistor is 1.5k, normally a linear value but, because the cathodes are paralleled, it is somewhat of a cool bias here. Sharing the cathode resistor also adds some positive feedback to the circuit. Positive feedback usually causes instability and oscillation, but, because of the large amount of attenuation leading into the second triode, here it only boosts gain slightly.
Early 120 amplifiers did not have a cathode bypass capacitor. A cathode bypass capacitor boosts gain by reducing the amount of negative feedback around the cathode. However, Wurlitzer started including this cap (even releasing a service memo recommending techs to add it to early 120 amplifiers) because it reduces hum by decoupling the cathode from the heater filament within the tube.
Heater circuit. Speaking of heaters, this amp was released before heater circuits were commonly center-tapped. Instead, one leg of the heater was connected to ground. Center-tapping reduces hum by halving the voltage at each leg of the heaters. This can be accomplished by removing the existing ground and instead connecting both legs of the heaters to ground via a small value resistor around 100 ohms. In this case, the heater wires should be twisted together and routed carefully to avoid hum entering the circuit through proximity to sensitive components.
Zobel network. One somewhat unusual feature of the 120 - borrowed from hi-fi circuits - is a Zobel network. This is a .001 uf capacitor in series with a 22k resistor, mounted across the output transformer primary. Zobel networks are intended to keep the output impedance stable across frequencies, which prevents ringing and other undesirable high-frequency noise.
Browse all of our articles on restoring vintage gear. Or, click on an image below.