drew.spriggs
Well-known member
Working on rejigging my tube workshops, and decided to do a pretty basic triode project - Merlins LV triode clean boost.
Before I got PCB's made, I cut a stripped down version on the CNC only to forget about vertical height requirements - it was a bit of a shitshow to get it into the enclosure as I had to flip the PCB, but we got there (with some kinda dodgy offboard wiring).
The (much better looking) version is off for manufacturing after having a bit of an argument with JLC about whether a daughterboard counts as a separate design or not.
The main circuit between both is the same - a triode with grid leak biasing to 12v +B (instead of ground), leading to a bias of about -0.25v. Second triode is set up as a follower to improve the woeful output impedance when running at such starved voltages, but input impedance still needs something decent driving it (I did think about a JFET buffer, but kind of ruins the entire point of all-tube).
Gives 18dB of clean gain with up to a 0.5v input signal (so 4v out), and anything over that does the nice tube squish. Didn't even add a bypass switch as I kind of figured it would sit on top of your amp at all times which also solves the impedance issues.
The 'fancy' version is practically the same, with additional power filtering and a constant current heater circuit (unregulated will pull almost 500mA at turn on with series filaments, and double that with parallel). Using a 12v supply for the heaters noticeably sounds worse to me than 12.6v, so I figured a regulated 6.3v would work just as well rather than a, say, 15v supply and regulated +B.
Now to figure out some interesting things to do with a tetrode!
Before I got PCB's made, I cut a stripped down version on the CNC only to forget about vertical height requirements - it was a bit of a shitshow to get it into the enclosure as I had to flip the PCB, but we got there (with some kinda dodgy offboard wiring).
The (much better looking) version is off for manufacturing after having a bit of an argument with JLC about whether a daughterboard counts as a separate design or not.
The main circuit between both is the same - a triode with grid leak biasing to 12v +B (instead of ground), leading to a bias of about -0.25v. Second triode is set up as a follower to improve the woeful output impedance when running at such starved voltages, but input impedance still needs something decent driving it (I did think about a JFET buffer, but kind of ruins the entire point of all-tube).
Gives 18dB of clean gain with up to a 0.5v input signal (so 4v out), and anything over that does the nice tube squish. Didn't even add a bypass switch as I kind of figured it would sit on top of your amp at all times which also solves the impedance issues.
The 'fancy' version is practically the same, with additional power filtering and a constant current heater circuit (unregulated will pull almost 500mA at turn on with series filaments, and double that with parallel). Using a 12v supply for the heaters noticeably sounds worse to me than 12.6v, so I figured a regulated 6.3v would work just as well rather than a, say, 15v supply and regulated +B.
Now to figure out some interesting things to do with a tetrode!