Okay, here is that spring reverb driver I designed for a friend's 1474. Here are a couple of points about the circuit:
• I tried to design it so as not to disturb the interface and the function of 2 available channels and inputs (although I would have liked to get rid of 1 of the channels)
• This will not place any higher load on the heater supply (the load on the B+ might be nominally higher), and the 6U8A can use the available socket of the removed 6CG7.
• This modification to the 1474 - the internally jumpered channels (with a single guitar in channel 1), and increased mixer resistors w/switch option will add pretty significant gain. The original ground scheme will not work and will add way too much ground noise. This can be solved by isolating the standoff grounds from the chassis, and running the ground leads to a star point.
• Lastly,
and most importantly to me, the close friend who I designed and built this for was diagnosed with a rapid form of ALS just weeks after I gave the amp back to him. Unfortunately, he lost the ability to play guitar within weeks of his diagnosis and his life just months after that. I would like to dedicate this to his memory.
Here it is in it's first iteration. It uses the same location to mix the wet/dry signal as the original 1474:
6U8A Reverb Drive & Hybrid Cascode Recovery Circuit.jpg
I never really liked the way the reverb recovery signal was mixed in the 1474, so I modified it. It's mixed in a later stage just before the phase inverter using a common plate load mixer (similar to the direction Silvertone went with the 1484). While I like the
idea of this circuit better than the previous iteration, I haven't had the chance to test it and check actual operating conditions vs calculated ones. Or, of equal importance, give it a listening test. One important thing to note is that I opted to redesign the stages for V3 (mixing and PI) to use a 12AT7 vs a 12AX7. The reasoning was that a 12AT7 still provides high gain, high input sensitivity, but has a higher transconductance which I though would be a better fit with the 6U8A's triode for the common plate mixer. So, the changes to component values reflect that as well.
Experimentation is needed in determining the final values of the shared load resistor, individual plate load/mixing resistors, as well as the cathode resistor in the first section of the 12AT7. The 6U8A/LSK170 cascode needs to be biased regardless due to manufacturing tolerances of JFETs.
I don't love placing the reverb control in the drive circuit, but it was a compromise.
(Schematic edited to reflect corrections)
Silvertone_1474_-_6U8A Reverb_Drive_&_Recovery_Mod.jpg
Please feel free to comment; corrections and suggestions for improvement are all welcome.
You do not have the required permissions to view the files attached to this post.
