New schematic for Review, Criticism & Errors

[NorCalTuna]

I dont want this to become a life safety or compliance thread, but...

Enzo is right here. Theres just no sure way to know what lurks below the cover plate of every outlet. I used to be an electrician, so I can offer a few insights on common code violations I saw routinely. Please excuse the fact I'm not citing NEC as I don't own a current one (or one on this coast!) or I would.

Reversed hot and neutral (the common and big one)
No or bogus grounds (literally fake bonds to boxes tied to 2 wire (old no gnd) romex)
Loose installations (cables free in walls, very common in retro, they get yanked when people are in the attic, etc)
Receptacles upside down! NEC says all receptacles with a metal plate SHALL be installed with the gnd pin up so when the plate screw falls out they don't short!
GFCI Controlled outlets with no tag (yah! the code says to label any outlet slaved onto a GFCI thats what they include all the stickers for!)
No line of sight disconnect on loads 1/4hp and above, residential- Yep, your disposal and dishwasher need to be on a switch or cord and plug!

So yah, the code is rigorous but that- as Dan Torres points out in his book- doesn't mean anything in america where the guy who wires the stage is working off his bar tab. He goes on to point out that very danger in working on equipment with a variac- that his had a spst in his case and the neutral was hot for whatever reason. Nathan, with all due respect, I'm not sure you make an argument against switching a neutral as much as for due diligence at the bench? I agree, and pull the power cord every time, in fact, I'm a huge fan of IEC plugs for this reason! (with that sweet fuse built in oooooo!)

[mooreamps]

Re: DPST mains switch. I agree that there are some advantages to doing it this way, but I'm almost positive that in the US, switching neutrals is not UL approved.

AND I AM SURE IT'S A UL REQUIREMENT ! !

-g

[octal]

I wonder if there's different rules for different product categories, or if the rules are different for 2-prong vs. 3-prong devices. I looked at some of the current Fender reissue schematics and they DO switch neutrals. Oddly, the exact opposite is true in the many, many pieces of hi-fi gear I've fixed over the years- never a switched neutral for the US market. Anyway, I'll retract my statement about switched neutrals.

[Enzo]

And just to keep the pot stirred up, remember that UL - Underwriters Labs - is not a government agency. it is an independent facility originated for the insurance industry, hence the word "underwriters" in the name. Point being that UL approval is not a legal necessity. Getting a UL sticker is not the same thing as passing the legal requirements.

[Billy R aka DynaFreak]

I agree with the previous poster that the preamp as drawn has a lot of gain. That may need to be tamed in order to keep from slamming the grids of the third stage and the PI with too much voltage and causing ugly distortion. Your large 0.1 bypass caps will contribute to the problem by allowing a lot of low freqs to pass. Rather than try to figure out exactly how to optimize component values in the schematic, a good approach can be to insert placeholders in the diagram. For example, you have no descrete grid-stopper resistors anywhere in the preamp. Draw them in and design the layout with them in mind, even if you are not sure you need them. Same thing with some voltage dividers/ RC networks prior to the later gain stages. That way you are building in flexibility to your layout that allows for easy mods.

In my experience you can build a nice sounding three gain stage preamp for clean sounds, but when you try to push them to produce a good preamp-generated distorted sound, they come up lacking. You really need four stages for good distortion. Or maybe I should say that it is much easier to build a good sounding 4 stage distortion preamp than a 3 stage one.

All that being said, then my advice is that if you are looking for a great clean sound then continue to work with a 3 stage preamp, but provide for a way to seriously attenuate and adjust frequency response of the signal prior to stage 3 and include a grid stopper resistor somewhere btwn 10k-100k. If you want a good distorted sound then consider adding a 4th gain stage.

[Zipslack]

UL tag

Actually, if you plan on mass-producing a product, you are pretty much REQUIRED to have UL certification (CSA for Canada, SEV/TUV, and all the other alphabets for Euro). One thing that adds to the cost of mass-production is the thousands of dollars spent on UL testing and certification. Small custom-job shops can get away with it, but they are exposed to lawsuits to a greater degree (UL cert won't stop lawsuits, but just helps deflect some of the more frivolous stuff). Just because something is not technically covered by law does not mean it is not enforced like a law or vice-versa (anybody remember how Social Security number is specifically NOT to be used as an identification number?). Play it safe and design to meet UL standards if you plan on selling it.

[NorCalTuna]

I'm going to pipe up here and disagree with billy and killer. Yah, there's a lotta bass in that pre and pi and it may be flubby, but too much gain? I'm not so sure. One, it has a cathodyne pi, so there's a bunch of gain it doesn't have- they have a gain of just under 1, like .95 or something. In fact, when you look at it that way, it begins to look like a Bandmaster- a stage, an eq, a stage and a LTPI; but in this case the LTPI is "split", figuratively. Yes, it will have more gain than a BM, but not a ton more: approximately the difference between a stage with and without a bypass cap, i'm thinking. And as for the bass, it might sound good, full and rich at low volume (ever use a "loudness" switch?) until it's pushed, so I don't see what the issue there is until it's upnrunnin and farting. The pi can have it's coupling caps swapped in a few minutes if need be. I'm still wondering why it has the .047 in front of the tone stack, there's a buck i'd spend on liquor or doughnuts...

As for "descrete grid-stopper resistors" they can be- and ideally ARE- on the sockets themselves, and can be added there pretty much at any time should they be needed, which is likely contingent on layout and attention to detail. The 68k (how about 33k or 27k?) definitely should be on the socket, however. This amp is basically a PR with no 3.3m reverb blend resistor, which of course eats up signal, but it does have that second gain pot. I say get er goin! Daylights a wastin!

[kg]

i'll always say built it, then tweak. no use trying to tune a schematic! so many variables go into how it's going to actually sound that going into the nitty gritty details before a baseline listen is likely to just waste time.

[Howlin' Mad Mac]

Thanks there guys for some really great advice. The grid stopper is definitely something I overlooked. As for gain & a lot of distortion, no. I'm hopeful if all goes as planned to turn the master up, the gain low & a nice, bright, chiming high end with a wicked "whoomp" in the bottom & a nice scooped mid. The kind of amp that'll definitely favor single coil, long necked guitars... thru a vintage Dyna Comp set 2 squash & maybe a nice 40ms repeat from an Aqua Puss... or to put it another way I likes my "twang". As for mass-production that's not likely... however building a few here & there out of my home shop ain't out of the question (& keep me out of trouble not to mention the woman's hair.) Building amps from new parts is just my plesant diversion from re-building dirty, dusty old Hammonds. Also got some nice mahogany & swamp ash stashed away to make a couple 'letric lap dawgs soon (will post ya'll some pix once I got 'em just right)... Yee Haw!

[Howlin' Mad Mac]

Hey also if I wanted to bring the gain down, say in the first two stages couldn't I achieve that by lowering the value of the plate resistors, thinking 68k might be a good place to start???

[NorCalTuna]

Build it!!! lol! Don't bother changing a thing! If you want a lower gain mode, try switching those cathode caps out, which may silence some flubbiness at loud vol. So, when it's down low, you get nice warm bass, and then turn it up and kill a cathode cap or two, and it'll tighten up (lose some bass) and sound sweet. Put a resistor, like a 10m, to keep the cap charged so it doesn't POP! when you switch it in... If it's lacking bass, then use SPDT switches to switch from say a 22uf to a .47. Everything can be done once it's up and running. Check out this sheeet from merlin:

The Valve Wizard -Cathode Follower

bon chance-

[Howlin' Mad Mac]

New and Improved (well maybe)

Was counting up parts over the weekend, viewed a schematic or two I hadn't seen before (or at least seen while NOT under the influence of tequila) & had an epiphany (i.e. brain fart) & look what came sliding out. Would love some input on the new "lite" version vs. the old design... hopefully ya'll got lots more good ideas!

12 Watt Dual 6V6GT Lite.pdf
12 Watt Dual 6V6GT Lite - Power Supply.pdf

[octal]

I think your "lite" version will not have enough gain- even to be a clean only amp with no possibility of overdrive.

A very large percentage of the gain of the first stage is thrown away in the tone stack. A cathodyne PI has no gain. 6V6 output tubes are low Gm as well.

Also, both phases of the PI output are shorted together.


Nathan

[Howlin' Mad Mac]

I think this one would need bigger PT & choke supplying more mA's (Deluxe vs. Princeton iron.) I'm still haven't quite got it all figured out how calculate how much gain a particular stage will have or total losses thru a tone stack yet (been 20+ years since electronics school & it's coming back but slowly.) Also this is based on a similar design but it uses a SE OT whereas I put in a UL SE OT... is this what makes both phases of the PI short? Does this design require SE iron? Why?

[octal]

Aaah, you have drawn a push-pull output section, but you're referring to a "SE" OPT.