2 new Brit inspired designs ready for inspection

[Howlin' Mad Mac]

Been reworking a couple of Brit inspired designs I've dreamed up. Please take a look, tell me if there are any glaring errors, etc. & let me know what you all think. Thanks.

Abbey Road - Schematic.pdf
Abbey Road - Parts List.pdf
Mersey Beat - Schematic.pdf
Mersey Beat - Parts List.pdf

[redelephant]

Cool. The only thing obvious to me is the PIs. The Abbey Road has that diode misplaced. Are u sure a single diode provides enough voltage drop? How'
bout a combination diode-resistor? The Mersey Beat needs the grid lifted. Cathode followers are for pussies. Just add that triode as a common cathode stage post EQ and connect the PI to that, kinda as in the Abbey Road.

[Howlin' Mad Mac]

Hmmmm... Well with the Abbey Road design I was going for more agressive gain, kinda like a Vox procreating with a Dr Z deal, so figuring on some overdrive happening at the PI. According to what I was able to take from the "Valve Wizard" site this PI design should eliminate any frequency doubling & provide some arc protection. Beyond that I'm still trying to understand exactly how it's supposed to accomplish its task. Now as for the Mersey Beat I'm shooting for a sort of lower power plexi but very vintage-like performance complete with tube rectifier & 40H filter choke. Here I'm thinking there really shouldn't be a lot of overdrive happening in the PI so I went with about as simple a cathodyne inverter as possible. Now as to why I would want to lift (disconnect) the PI grid I'm having a difficult time following your reasoning. Are you suggesting I should ground the grid (I've read it's generally considered better to ground a grid than disconnect it) so the signal would enter the stage thru the ground?

[redelephant]

Your ideas and designs are cool, it's just that the PIs need a little fix. What I mean about lifting the grid is not hanging it in the air, but placing it on the bottom resistor so you bias that stage correctly. See drawing. I think with 100kohm resistors, a single diode will be too little and the current too high, droping too much quiescent voltage over the resistors. Either go down in resistorvalues or increase the grid-cathode voltage with more diodes (or a LED).

Edit: No need for high voltage diodes in this case, isn't it better to use small signal diodes such as 4148?

Btw, on the lower gain design, having the EQ before the PI without a gain stage means you risk having to turn the preamp gain up so much that it distorts before the output has reached full power. Although with EL84s I guess you need not worry since they don't need a whole lot of drive, so never mind. But generally for a cleaner style of amp you want the output stage to run at full steam way before the preceding stages.

[pdf64]

The Abbey Rd B+ level seems too high to me, certainly too high for class A operation, and with such a high B+ and 220 ohm cathode resistor, my guess is that at high signal levels, the el84s will be closer to class B, with significant crossover distortion when overdriven.
Also there's an oversight on the heater circuits, they require an ac ground reference / dc reference.
The Abbey Rd tone control seems unusual; have you tried it; what's the design rationale behind it?
For me, the single tone control arrangement on the 6G3 is hard to improve on.
http://www.webphix.com/schematic%20h..._6g3_schem.pdf
Pete.

[Howlin' Mad Mac]

Changed directions slightly

After reading your comment & learning a few more things on my own I started drawing this up instead. I haven't worked out all the component values just yet but figure B+ @ 265-280V, input 1 bright, padded slightly (-2 to -3dB so 48K to 56K), 100K on the plate, 4n7 coupler, biased slightly hot (1K2), partially bypassed (1u) & input 2, not padded (33K), 100K plate/.022u coupler, again biased slightly hot (1K2), partially bypassed (10u). Stage 2 incorporates some local feedback & buffers the tone stack so 680K on grid inputs, 100K to the plate, 1M2 feedback, .1u coupler. Tone stack is pretty standard old Marshall. Stage 3 is where I went a little wild... 220K on the plate, biased pretty cold (8K2) with curve selected closer to cut off & finally a AC Long Tail PI. So what do you think?

18watt.pdf