AB763 Deluxe scratch build

[Craw]

Hi guys !  I'm looking to build a modified AB763 Deluxe for my next project.  I have nothing but tube sockets and a chassis and an assortment of passive electronic parts on hand at this point, so its an open book.  I know that many or most of you are guitarists as well as amp geeks, so I have a few questions from both perspectives.  I don't see any need for the tremolo circuit, so that's out from the get-go (my apologies to the trem lovers).  I do however want to keep the reverb in the project.  I also see no need for two identical preamp sections, so I'd like some input on that issue.
1)  I was thinking of building a D**ble style preamp section (instead of the other identical preamp) with some sort of switching or gain stage cascading capability.  Any suggestions on what circuit would work well in that scenario ?
2)  In the event that idea "1" isn't a particularly good one, can anyone recommend an alternative fender preamp section that might have a substantilly unique flavor.  Most of the Fender preamps I've printed out so far look roughly the same on paper.  Is there a substantial difference between the way the AB763 and 5E3 preamps sound ?  Are there more unique alternatives ?
3)  There are some substantial differences between the 5E3 and AB763 output stages, most notably the feedback loop from the OT secondary in the AB763 design.  What is the advantage of this over the 5E3 output section ?
4)  I've noticed that Mercury Magnetics is offering OTs with 8K primaries for dual 6V6's.  Is this a significant step forward in terms of what a guitarist would like to hear out of his amp ?
I realize that some of these questions are subjective, but your input would be appreciated regardless.
Thanks,
Charlie

[sluckey]

I built a single channel AB763 DR with reverb and no tremolo. It's the TDR on my website. There are several guys on the forum that can offer advise on dumble style preamps.

3)  There are some substantial differences between the 5E3 and AB763 output stages, most notably the feedback loop from the OT secondary in the AB763 design.  What is the advantage of this over the 5E3 output section ?

There are some other BIG differences also, like LTP PI and fixed bias. All these changes work together to provide more/cleaner power.

4)  I've noticed that Mercury Magnetics is offering OTs with 8K primaries for dual 6V6's.  Is this a significant step forward in terms of what a guitarist would like to hear out of his amp ?

I have MM iron in my TDR. I haven't used MM in any subsequent builds though. The biggest thing that stands out about MM is the price!

[Fresh_Start]

I'm not sure that I understand the question about an 8 ohm primary for two 6V6s operating in push-pull.  Hoffman's transformers for the Princeton Reverb and Deluxe Reverb both have 8 ohm primaries:
http://www.hoffmanamps.com/MyStore/perlshop.cgi?ACTION=enter&template&thispage=Transformers&ORDER_ID=!ORDERID!

I would not hesitate to use Hoffman's iron BTW based on two almost identical Princeton Reverb builds, one with Mercury Magnetics and one with Hoffman transformers.

Here's a schematic for a Super Reverb preamp that I built with a not-normal Normal channel and reverb on both channels:


The idea was to get a brown Vibroverb sound from the Normal channel.  The cathode & gain switching may be useful.  However, in this case the "Bright" switch actually is a guitar/harmonica switch.  Also, the "Lo" input is intended for a harmonica mic.  Might be helpful as a starting point though.

Adjustable or switchable negative feedback might be something you want to experiment with.  Maybe couple it with a "raw" switch that increases the "mids" resistor on the bottom of the tone stack.

Hope this helps,

Chip

[Craw]

I'm not sure that I understand the question about an 8 ohm primary for two 6V6s operating in push-pull.

Hi Chip,  It appears that most OTs designed for dual 6V6's come with a 6.6K primary.  I'm trying to get a handle on why MM is offering OTs with an 8K primary.  It must have something to do with freq. response or efficiency/headroom (or something entirely different).  I'm just too newbie-ish to figure it out right now.

The idea was to get a brown Vibroverb sound from the Normal channel.  The cathode & gain switching may be useful.  However, in this case the "Bright" switch actually is a guitar/harmonica switch.  Also, the "Lo" input is intended for a harmonica mic.
Adjustable or switchable negative feedback might be something you want to experiment with.  Maybe couple it with a "raw" switch that increases the "mids" resistor on the bottom of the tone stack.

I like these ideas and will look into them.  I'm not familiar with the vibroverb sound.  I'll ask some of my guitarist friends about that.  I also like the idea of having the reverb available on both channels.  Thanks for your thoughts and schematic, Chip

[Craw]

I built a single channel AB763 DR with reverb and no tremolo. It's the TDR on my website. There are several guys on the forum that can offer advise on dumble style preamps.

There are some other BIG differences also, like LTP PI and fixed bias. All these changes work together to provide more/cleaner power.

1) So which ends up being cleaner with more output, the AB763 ?
2) I took a look at your web-page. Nice workmanship !  I didn't have a lot of time, so I'll look to see what you drew as a schematic, later.
3) I'll do a search for the dumble preamps later as well.
Thanks for the input, Steve

[Fresh_Start]

Craw - I was thinking over your questions while driving around and had a thought.  You may be able to put a 12AY7 in the V1 spot for the "Normal" channel.  Use the "Hi" input jack for one channel and the "Lo" input jack for the other "5E3" channel.  Make the "Bright" switch jumper the two channels when it's "on".  IOW both triodes operating in parallel with the Bright switch on. 

I haven't figured out yet how to isolate the "Normal" (IOW 5E3) channel from the "Vibrato" channel though.  It wasn't necessary in the UberVerb.  Actually, the build originally HAD isolation resistors between the coupling caps and the junction of the two channels at the 3.3meg/10pf point.  That created some unexpected voltage dividers with the grid return resistor for the reverb circuit (1 meg) and each mixing resistor was in series with the 3.3 meg reverb mixing resistor.  Maybe SLuckey will come back and give us an idea of how to mix the two channels.

Oh, on the power amp side, there's no reason you couldn't have a fixed/cathode bias switch.  You might even want to combine it with some way of dropping the voltages on the power rail for the cathode bias setting.

This could be a pretty cool concept if we can just solve the mixing problem...

Cheers,

Chip

[38Super]

I've noticed that Mercury Magnetics is offering OTs with 8K primaries for dual 6V6s.

In general for guitar amps,  you'll get the most power transfer from output tube, through OT, to speakers if the amp output impedance and the load (speaker) impedance are matched.

6500 Ohms seems to be the most commonly referenced load resistance for a deluxe, but does that mean you'll get the best sound (the sound you like most) by loading the 6V6 output pair using speaker and OT combination which results in 6500 Ohms?  Maybe...  You may like the sound of loading the output pair more heavily or more lightly, e.g., 8K Ohms.  There are a lot of factors such as tube type, B+ voltage, cathode current, etc.

A lot of guys like Edcor, Pacific or Heyboer transformers.  They're cheaper than MM and have found their way into many boutique amp clones around the world.  There is a site rjguitars.net that can help you find these trannies for a good price.  I have no financial interest in this outfit.   

cheers,

rob

[Craw]

I emailed a rep. from Mercury Magnetics about transformer questions related to this project.  This re-post of our conversation is just to help educate any other newbies asking the same questions that I am.

I'm sure you receive many requests like this, so I'll try to be brief.
I will be building a Fender deluxe amp from scratch.  Its likely to be some sort of AB763/5E3 hybrid.  I'm a little confused about the many available OT choices, so I'd like to ask a few questions.
1)      Is the design intent of the 8K primary as opposed to the standard 6.6K, simply to optimize the impedance for a pair of 6V6s ?
1a)      Yes it is, but it is also a Clone, meaning we found an original with that primary which is ideal for 2 x 6V6’s.
2)      If the primary impedance is not listed, should I assume it to be 6.6K ?
2a)      Not necessarily, we just do not always reference it.
3)    I was hoping to find a transformer with both 8 ohm and 4 ohm taps.  Are there any available with an 8K primary ?
3a)   Not unless you go Blackface Deluxe.
4)      If I were to use a tweed designed transformer on a blackface output design, would there be a substantially audible difference ?.....  and could you describe it, if possible ?
4a)      Yes there will be. The Tweed is the Tweed sound that we all know, very thick and swampy, and not as articulate. The Blackface is stiffer, with more headroom and better low end.
4aa)      Basically, it really depends what you are after tonally.

[Craw]

Thanks 38SUPER.   I'll take a peak at RJ when I get a chance.

[Fresh_Start]

More references for 8K ohm OT primary:
Take a look at the GE data sheet for two 6V6s in push/pull, Class A/B1 on second page, "Effective Load Resistance, Plate to Plate"  8K ohms
http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6V6GT.pdf

My guess is that Leo Fender followed the data sheets, especially if that's what Mercury Magnetics offers in its "Tone Clone" series.

On the flip side though, I see that Dave Allen offers 6.6K primaries on his output transformers intended for dual 6V6 amps:  http://allenamps.com/parts.php#transformers

Cheers,

Chip

[jojokeo]

I mean this in the best way but all of this talk about pri Z isn't really worth it. In three recent scratch built lower powered amps, I find it very difficult to hear OR feel any signigicant real differences when switching between values this closely. If I was to put someone in a blind A/B test I think many others would say the same. So many other things that actually do affect sound, tone, and feel like tubes, speakers, coupling caps, NFB, B+ voltage, etc... all make a heck of a lot more noticeable difference.
my .02

Craw,
You may want to consider one channel 12ax7-ish and the other pentode-based w/ either an EF86 or 5879. Lots of tone, mojo, and functionality. You also could make it so each could be played seperately and they could also both be combined or cascaded.

[Craw]

Chip,
I like the way you think.  It seems that the 5E3 input/tonestack is not all that interesting, now that I've looked at it a bit more.  However the Bassman 5F6A has a unique enough input/tonestack to qualify for this project although the gain staging is pretty different.  I might be able to use the concept for an overdrive stage.  I still have to print and study Frank's Uberverb schematic and SLuckey's TDR schematic to see how the puzzle will fit together.  I like the bias method switch idea.  I don't really understand it yet, nor do I have any idea how switching bias methods will impact the amp's overall sound.  Crucial for me, is a solid low frequncy response, so if one bias method robs the amp of solid lows, I won't use it.  I'll also be looking into coupling cap values to make sure I'm not bleedin off too much low end.  Next, I'll have to dig up the dumble OD info.
The more I learn, the more I need to learn..................... Very Cool
PS....Where do I find the GE datasheet that you were talking about ?

[Craw]

Jojokeo,  Thanks.  I know very little about the EF86 except that they have a reputation for being easily microphonic.  I do know that they are used vey successfully by at least one boutique builder(I can't remember which one).  Absolutely worth checking out. Thanks again.

Charlie

[Fresh_Start]

Charile - I added a link to the 6v6 data sheet in the post where I forgot to put it earlier.  Here's a link to a great, searchable database of tube data:
http://www.mif.pg.gda.pl/homepages/frank/vs.html

You have a Bassman/Marshall flavor in the "Normal" channel fairly easily.  5F6-A schematic  Use a MOSFET in place of a tube triode for the cathode follower before the tone stack.  Use 5F6-A tone stack values and Mid pot configuration (it's different from Blackface Fenders).  See MOSFET Follies and Follies 3rd Edition.  Note that you need a 220 ohm "gate stopper" resistor which is not shown in the drawing but is described in the text.  You won't get the compression a tube triode cathode follower produces, but you will reduce the signal loss from the tone stack and be able to use Bassman cap & resistor values.  The output from those first two triodes might even need a voltage divider to bring the level down, but I'm not sure about that.

Going back to my earlier Tweed/Blackface idea, here's another potential twist.  Set up the "Normal" channel as parallel 12AY7 triodes with Tweed Deluxe volume & tone controls.  Route the output from the "Normal" channel into the "Vibrato" channel.  You'll need to knock the signal down some or you'll overdrive the first triode of the Vibrato channel too much when the channels are cascaded.

Jojokeo's pentode idea is a great suggestion too.  Guys here have had good success with the 5879 tube.  Apparently it's not as susceptible to microphonics as the EF86.  Dr. Z among others uses an EF86 in several amps BTW.

Sorry if throwing out too many options is a problem - I love this kind of brainstorming.

Chip

[Craw]

Steve, Chip, Rob, and JOJO,
Thanks a bunch for all of the insights, ideas and links.  I have a ton of information to digest before I even order the first resistor.  Between the Kuehnel books and all of the info I just downloaded, I should have this thing up and running in about a decade.  ;-)  Sure beats letting senility set in.........