Pro Opinions On A // 6L6 Build I'm Working On, Please

[AHeck]

Here's an amp on paper, It's my first time, so...  I used .05VAC for signal, based on strumming a normally wound neck an bridge single coil configuration with no real attack or excessive force, measured right after the pick clears.  I used .1 as the factor for 12 o'clock on the volume pot and -15dB loss in the tone stack. 
To be honest, I'm not sure this is right or will work.  That's why I'm asking.
Same B+ all down the row, again it might be foolish or even stupid, but I made the math work even though I'm convinced that that is but a small part of this art.
I haven't gotten into the particulars of the PSU, I wanted your opinion first.  The plan is to wire it up on my prototype board , dial in the voltage and see how it sounds.  However, I don't want any fires or exploding components. 
I am very good with constructive criticism, it's a great way to learn.

[sluckey]

Using a single B+ rail to supply every stage is going to be problematic. If it would work, every amp you see would be doing that. Stick to the tried and true circuits.

Even if a single B+ rail would work, look closely at your circuit. You have the 6L6 plates tied directly to the B+ rail for all the little tubes!

[AHeck]

I do have the have the 6L6s and the other three gain stages operating off of a 370ish volt B+.  Aren't most amps power sections and pre sections working off of the same B+?  If you wouldn't mind, where'd I go wrong with the gain stages?  I was pretty sure I dropped the voltage into their working tolerance.

[sluckey]

You have the 6L6 plates tied directly to the B+ rail for all the little tubes!

[AHeck]

Thank you for the response, sluckey.  When I read this forum, you are always giving well rationed and experienced advice.  I appreciate you taking the time to answer my post.

[PRR]

The first tube does not need much current but it must be really clean.
The big tube needs lots of current but it does not have to be real clean.

My well-water is not so clean. I have a screen to catch the fish-turds, a fine screen to catch sand before it gets into valves, a paper filter for cooking and bath water, and I used to have a super-filter for drinking. If I used the paper filter on the garden water it would clog constantly and be more expensive than it needs to be.

You think the grid is the Input. Actually everything is an input. The big chunks of current in the 6L6 make wobbles in the B, which is allowed to sneak-back to the small-signal stages will oscillate like a mike on a PA system.

And yes, I think you should practice your circuits.

[Ethan]

What PRR is alluding to (with a pretty great analogy, I might add) is stage decoupling.
Each gain stage has AC current demands which it places on the power supply.  Those current demands will cause ripple current which will affect all the other stages both before and after the stage in question.  Since the gain is increasing with each stage, the effects of power supply noise on the first stage will be amplified a lot and will be really noticeable.

If you look at the power supply topologies of classic amps you'll see that the main B+ supply to the power tubes almost never goes directly to the preamp sections.  It will first go to a resistor and capacitor (an RC filter) which performs two functions: it drops voltage and it decouples the stage from the later stages in the amp, meaning it minimizes the effects of power supply ripple caused by later stages. 

It is the combination of the resistor and the capacitor that allow this to happen.  Neither part will do it alone.  The resistor is there to drop voltage but sometimes it's only dropping very little.  What it's really there for is to form an RC filter so the stage can be decoupled.  That's what you're missing from your circuit.

Sometimes the resistor will be replaced with a choke (an LC filter) but the principle is the same.

When designing from the ground up, it's good practice to start with the power stage and work backwards.  In your case it might be a good idea to take a step even further back and start with the power supply.  Preamp topologies are neat and there are a lot of them to play with, which makes them really fascinating, but the meat and potatoes of the amp are in the power supply and the power stage.  Without a strong foundation there, the tower is going to fall over no matter how much fun you have designing the preamp!

[AHeck]

Alright.  Now to rethink how to rethink this.  It seems that I'll forever be designing fish turd filtration systems in my mind, and for that I thank both of you, Ethan and Sluckey.  Looks like I've got a lot more reading to do on PSUs.

[Ethan]

Nobody wants their amp to sound like fish turd.

[DummyLoad]

i don't think that cathode follower is biased correctly.


--pete

[PRR]

i don't think that cathode follower is biased correctly.

No; and the grid bias divider on the "cascoding" stage looks like a buzz-magnet (although there may be significant rejection here).

While it is a different plan, Kuehnel's study of the DeLuxe covers stage-gain and power filtering very well.

[AHeck]

Okay, I've taken the advice and done some cursory study on some of the issues pointed out and have come up with a revised circuit.  I'm hoping it is improved.  I'd be very happy if any of you would stick with me and point out problematic areas and poor or uneducated decisions.  If this is not something you'd like to spend your time on, I , of course, understand.
Voltages I'm trying for, which should have been on the schematic, are:
270 at the transformer/297 at the 47uf cap/ 288 at the 22uf/ 277 at the 16uf/ 266 at the 10uf.
In all honesty, I read through a treatise on PSUs at Eliot Sound Project and checked out a copy of Duncan Amps PSU Designer.  It seems I need all the help I can get.  So if These voltage expectations are wrong, or the level of filtering is insufficient, please let me know.  Again, many thanks.

[sluckey]

Try this. I untangled your power supply and fixed the biasing on V2A.

[AHeck]

Thanks, very much better on the PSU knots I'd made.  If I understand this, the 68K RL on V1a went away because I've got the high impedance of the CF as a load?  The RL of the CF is gone because of the 1M Pot of the TS? 
The biasing on the CF now allows for a little over 3.5VPP?  Am I following? 
I dropped the coupling cap after the CF to .47uf to help with really low frequencies and could certainly go lower, but I don't want to thin it out.

[sluckey]

The 68K went away because you need some method to properly bias the CF. I chose direct coupling because it was the easiest method. Notice the cathode resistor has also been changed to 56K.

[PRR]

220 Ohms and 10uFd is not great value for preamp rail filtering. It is 75Hz so gives very little filtering of 100/120Hz wall ripple, also bass frequencies which the amplifier passes well. And it's maybe 1% drop of DC voltage, you can be greedier. If you steal other designs you find 10k to 22k in single preamp filters. And 10uFd even 16uFd is pretty 1947.

[thetragichero]

maybe it's a bit "hacky" but since there is a wealth of 'prior art' available for known good sounding amps, spend some time looking at various schematics (hoffman has a nice amount) to see what works. unlike semiconductor technology, tube technology hasn't changed in several decades so there's a lot to learn from

[AHeck]

I have been working very hard on this circuit.  After the much appreciated help getting it to a safe design without catastrophic elements, I have built it on my prototyping board and have been tweaking and playing it on a daily basis.  I'm back for another round of critical input and face palming if necessary.  It sounds quite good at overdriven levels, if a bit polite and nondescript when clean and never gets out of bounds to my ear.  However, I am wondering why I cannot seem to get more than 5W RMS at the output (by my measurements- (peak squared x 2/secondary imp (8 Ohms)) x .707).  After measuring my signal input to the power grids higher than I had planned, I though that I could change the bias and get some more out of that stage, but I didn't measure any change.    I had increased Rk from 390 to 600 Ohms.  What am I missing?  I had only anticipated feeding the 6L6s 20Vpp, but I'm getting 33 when opened up.  Also, is the 1M resistor parallel to the 1M pot a redundant mistake?  LNF on the final gain stage, bad idea? If I were to go global, could anyone give me pointers at how to safely accomplish that?   Just trying to learn here.

[PRR]

600r does not feel right. What is your power tube(s) dissipation?

[DummyLoad]

with the B+ you indicate, you'd need a cathode R of about 165Ω for a pair of 6L6GCG or a 330Ω under each 6L6GC. use a pair of 330Ω 5W in parallel for a 165Ω.


--pete
 

[AHeck]

Yessir, I’ll try that today. Thank you for the response.

[AHeck]

As per given advice, I have changed the bias on the power tubes with very good results.  I'm now outputting 22 volts/pp to the speaker and the over all character of the amp is richer and more complex. Chiming cleans and very smooth overdrive.  Thanks for that pointer.
My total plate dissipation is 125.4 mA so just under 20W per tube. 
It does have some issues though, the causes of which I'm supposing are mainly due to the nature of my prototyping  set up, i.e. lead dress and quality of connections.  Neither of which are too terribly bad and I've taken measures previously that have actually made the circuit quite quiet as far as hum or hiss, but what I'm hearing in the background of the guitar noise is a ghosting distortion.  It is distinctly not attached to the actual tone of the instrument but seems to follow it or exist parallel to it.
Also, when I removed the local negative feedback from the second triode of the 12AX7, all hell breaks loose.  Many non guitar noises appear which are mostly but not exclusively tied to the volume and control pots.  My input and runs to and from the tone/vol circuit are shielded and tied to ground at one end of each run. I'll be honest, I don't know what oscillation sounds like, but I believe if I did, I'd know this was it.  I put the resistor and cap back, thinking maybe I knocked something askew when removing them initially and the offending sounds are gone, minus the strange background distortion tagging along. 
Any thoughts about the possible causes for these symptoms?  I was thinking that there might be phase issues, but that is a very uneducated assumption. 
Thanks much for the pointers so far. I'm very happy with the tone I'm getting and I'm going to do some basic housekeeping and a few tube swaps as trouble shooting measures.