First amp design / layout

[landmarker]

Updated with v1.2 schematic and layout

I've come up with an idea for what I think will be my "perfect" amp, combining all my favourite features into one horrific Frankenstein's monster - specifically

AA763-style tone stack / tone shaping
Ampeg/Swart style independent transformerless reverb
Single-ended power section, with possibility to parallel tubes on the rare occassion I need 10% more volume
Possibility to tweak/mismatch OT and speaker impedence to compensate when using two output tubes
Choke before power tubes to get as quiet as possible single ended
All this built into a Princetone Reverb chassis, plus a Vibrochamp board with a few extra eyelets added

Please, for the good of my sanity and marriage, talk me out of building this monstrosety... or at least tell me what terrible pitfalls lie within

[EL34]

Posted on wrong board... moving

[sluckey]

I see a couple schematic errors that will prevent the amp from working.

1. V2 pin 2 has no dc path to ground. Bias cannot be established for this triode.
2. V3 pin 7 has no dc path to ground. Bias cannot be established for this triode.

Also V1 pin 8 and V3 pin 8 share a common cathode resistor. It's a bad idea for consecutive gain stages to share a cathode resistor. This may result in feedback oscillation.

The reverb tank output is feeding into a 10K/22K voltage divider. That tank signal is the weakest signal in the amp.You really don't need to knock that weak signal down any more.

Just a general observation about the schematic... Include tube designator numbers and pin numbers. And be consistent with using dots to indicate connection points. These make it easier to talk about your schematic and also make it easier to relate schematic to layout.

[tdvt]

Sounds doable to me.

I built something sort of similar earlier this year;  AB763 TMB pre, with reverb, into a SE 6V6. Power section included a choke.

Mine differs in that I used full-on 2-tube reverb (partly for the 3rd gain/mixer stage) & I did not have a parallel 6V6, though I rebuilt a friend's 3X6V6 SE amp shortly after & the additional tubes worked well.

I used a short 3-spring tank, though you will be able to use a full-sized tank in a PR cab.

[landmarker]

Thanks a lot for the feedback... it looks as if I might have to go ahead and build it.

Sluckey, thank you for the very helpful feedback - I have tried to embody most of your suggestions, I have left the 10K/22K divider after the reverb tank for now, as this is how it looks in the Swart amps (and copies thereof...), nevertheless maybe this is something I can tune to match the reverb level to the dry level, since they are independent in this design.
The 10K is not there in the Ampeg Reverberocket, but this uses a 6SN7 medium gain triode for recovery rather than the 12AX7 triode found in the 12DW7 - maybe this explains it?

[sluckey]

V2A grid still has no dc path to ground. Connect the new 220K directly to pin 2.

[landmarker]

Arrgghh... I got it right on the layout but messed up the schematic.  Thanks again!

[landmarker]

Here is the update

[tdvt]

Maybe you are aware, but I wanted to throw out there that SS rectification shouldn't have an audible effect on an SE design if you were inclined to simplify the power section.



https://www.aikenamps.com/index.php/what-is-sag

[landmarker]

Good point, I hadn't considered a diode rectifier.
My main rationale behind using a 5Y3 was getting the B+ down to reasonable levels for the 6V6s cathode biased with a 5ohm primary, since the only part of the amp I will be re-using from stock is a Deluxe Reverb PT (330-0-330V).  I don't really want to go down the road of using zener diodes or sag resistors, but I'll investigate further...

[Bash]

The reverb tank output is feeding into a 10K/22K voltage divider. That tank signal is the weakest signal in the amp.You really don't need to knock that weak signal down any more.

What would your suggestion be here? Just remove the 10K resistor, or remove them both?


BTW Looking like a great project. For versatility you might want to add a EQ bypass-switch (just a switch the breaks to connection between the mid-pot and ground). I've added one to my AB763 project and I love it.

[shooter]

fwiw;

the signal path tubes, cathode bypass caps;
I've found "starting" at 10uF vs 25uF saves a couple un-solder, re-solder steps when you find your signal is "muddy" (too much bass passing through)


your ears may very 

[sluckey]

The reverb tank output is feeding into a 10K/22K voltage divider. That tank signal is the weakest signal in the amp.You really don't need to knock that weak signal down any more.

What would your suggestion be here? Just remove the 10K resistor, or remove them both?

I would replace both resistors with a single resistor (47K to 220K) from grid to ground.

[landmarker]

I'm finalising the layout to scale for a PR chassis, most of the parts are on their way.

I moved the voltage devider at reverb recovery off the board as this simplifies the layout and will let me test Sluckey's suggestion of using a single resistor to ground without creating rework.

I'm still considering options to switch down the B+ as I've been enjoying playing my PR with 2x6K6 recently, and realistically I'll never need more clean power than 2x6V6 can deliver from this amp.

[tdvt]

Resistor feeding the grid of V3, 4.7K on the schematic, 4.7M on the layout.

Also, for clarity, you might think about only updating the first schematic to the latest version. Since there are several versions posted already, maybe label the others:v1,v2, draft or whatever, but just keep the first one current. 

It makes it much easier down the road to find the final version

[landmarker]

Thank you, I'll keep the first post updated with the latest schematic and layout.

[johnnyreece]

I'm still considering options to switch down the B+ as I've been enjoying playing my PR with 2x6K6 recently, and realistically I'll never need more clean power than 2x6V6 can deliver from this amp.

Sorry to kind of derail your thread, but I'm working on setting an amp up for 6K6 tubes, and would love another data point.  Can you tell me what plate voltage/bias setting you're running your 6K6 at?  Also, what impedance you're using for the OT?  I appreciate any info you can share!

[landmarker]

I'm currently running them in a Class AB fixed-bias output section at about 370v / 16ma, so 70% max dissipation but well above the max rated screen/grid voltages.
I'm not sure if they're as sturdy as 6V6's at handling higher plate voltages, but since NOS 6K6's are cheap, I figured I'll find out the fun way.

[johnnyreece]

Thanks for the input!  Are you also running 8k primary?

[landmarker]

Yes, 8K primary
They sound fantastic and the great thing with this setup is, if I need more power, I can swap out the 5Y3 rectifier I use with the 6K6's for a GZ34, tweak the bias and run 6V6's without any other modification.

[landmarker]

The build is well underway now, I'll share some pictures when it's a bit more presentable and ready for first tests.

I've made a few changes along the way, reflected in the schematic and layout in the OP.

First I realised it might be a good idea to ground the reverb filter stage with the preamp instead of the power amp, so I've added another filter, though it means a section of the can cap is unused.  I can test if it really makes a difference later on.

I've also started with 10uf cathode bypass in the 2nd and 3rd gain stage, again it will be easy to test the effect of adding more capacitance later.

It's been great fun and a good exercise to get back in the groove of doing purposeful and tidy work (rather than modding and prototyping) before tackling a service on my beloved 63 Deluxe Reverb!