Someday soon, my Silvertone 1484 will live again ... a long-form thread.

[thorpej]

OK! There was a slight change of plans! Because I wrote down the wrong measurements!

I've had to change the PCB layout I was planning on quite a bit. Starting with the power supply, which now visually looks similar to the original.



All of my subsequent posts will reference this new layout.

 

[thorpej]

Pre-Amp Section



The pre-amp section lives between the tone controls and the AF amp portion of the output section. It essentially serves as a mixer.

Each channel gets its own half of V2, and it looks pretty normal -- 1M audio-taper pots serving as grid-leak, wiper from the pot feeding the grid of the triode, the usual ~1V bias cathode bias voltage you normally see on a 12AX7 gain stage. Note, however, that the plate voltage is pretty low - ~93V, and then look a little further upstream to the resistor junction circled at the bottom of the diagram. There is a 220K dropping resistor (R27), followed by a 68K plate load resistor (R25; ignore R26, the 560K resistor for now). There is another 68K plate load (R24) for the upper triode, and both share a single coupling cap (C12) to the output stage, thus mixing the two channels together.

Now, R26. This is where the wet reverb signal is mixed in. I'll go into greater detail about the reverb circuit in a future post, but this is where it's mixed in. Because of the topology of the circuit, it's not easy to fully bypass the reverb; R26 is the plate load for the 2nd triode in V4, which serves as the reverb recovery amplifier.

It's a little diabolical, but Danelectro was pinching pennies and I admire the commitment to the bit.

The upshot is that the reverb recovery amplifier essentially behaves as its own channel that's mixed in after the volume control for the channel that provides its dry signal. It means you always pay the noise cost of that recovery amplifier regardless of your use of the reverb, but it also means that you can turn the reverb all the way up and the regular Channel 2 volume all the way down and get just an extra-spooky reverb sound with no dry signal at all. The reverb on these amps is objectively terrible, but this is kind of a cool effect.

Here's what it looks like on the PCB:

 

[thorpej]

Input Section



The input section is pretty straight-forward, but besides the presence of effects on Channel 2, both channels have different audio characteristics.

Both channels feature dual input jacks with 68K series resistors before the 1M grid-leak on their respective 12AX7 triodes. That voltage divider results in a ~7% attenuation of the input signal.

Channel 1 has local negative feedback (R5) taken from the top of its tone controls. Channel 2 lacks the negative feedback, but adds a 220K resistor (R9) to form a voltage divider with the upper part of its tone controls which influences C9. Channel 1 is thus the "clean" channel, and Channel 2 breaks up earlier in the presence of hot dual-coils like those on my Les Paul.

On Channel 2, the junction of R9 and R15 is where the tremolo circuit exerts its influence. More on the tremolo in a future post.

The tone controls for each channel are exactly the same. I won't bother with a network analysis here, but as you can see they're pretty straight-forward subtractive tone controls and how they work is fairly obvious.

The input section appear in 3 distinct locations on the PCB, and get connected together by flying leads to/from tube sockets. Note the cut-out in the PCB to make room for the pilot light.