Simple Chip Amp for P to P wiring

[ratza]

That is a great double triode. Have fun with it.

[danielwritesbac]

Originally Posted by ratza

That is a great double triode. Have fun with it.

Hey man, thanks. Question: Would you look over the schematic at post 74 and do a sort of "fill in the blanks" for the LM3886 resistor values? The point (I think) is the least gain from LM3886 and the most gain from the tube, all within reason, of course.

[coldcathode]

All,

This project is taking on a life of its own. While going thru my junkbox I found a good size transformer. I think it is from an old Receiver. It has a secondary of 62VCT. It is of the size and shape that I "like" for my projects. I am thinking that from its size and weight it is good for about 300VA or so...maybe more. It would free up some real estate on the chassis for the tube complement to be added. I think I can Point to Point what is needed to get this together rather than purchase pre-built stuff. I am thinking that I can built a BRIDGE using Ultra Fast Recover Diodes on a piece of Perfboard for the PS. This is simple to do just use the center tap as 0V and +/- off the "bridge" correct? Should be about 40V =/- ??

Daniel,

6922 is I nice tube, where did you buy it and what did you pay?

I am not sure that you really need a tube with that high a Mu (amplification Factor).
What experience do you have with tubes? I can certainly be of help to you.

Running that tube at 90V on the plate with 12mA of current gives a Gain of about 33. That is a lot of gain unless your source is very weak.

I will explain the schematic from #74 shortly.

Let me look at 692 data sheet for a minute or two I can give you revised values for that tube.

What is the minimum gain that a 3886 likes in its circuit? That is why I left those values blank.

[danielwritesbac]

62vct + MUR860 will stick open one of the output transistors aboard LM3886 and output a steady 45vdc to the speaker after approximately 5 months time.

Long term testing suggests that max is 52vct transformer. And the heatsinks will be much less expensive--so much less expensive that the right size transformer is free (in comparison).

This is approximately the right size transformer for LM3886 stereo amp with dual bridge rectifiers: http://www.antekinc.com/details.php?p=79 28+28, 7a

And this is less expensive for single bridge or dual bridge rectifiers:
http://www.antekinc.com/details.php?p=53 25+25, 5a

[coldcathode]

To all,

This thread has officially been HIJACKED by myself (the OP) and has now become a collaborative effort (between myself & Daniel) to design a relatively simple TUBE/Gainclone Hybrid.

I will be adapting the design for my own needs but for the purposes of the project we will be looking at an outcome of:

40Watts/Channel @ 8Ohm load
Dual Triode input section
Relatively low voltage on the "tube" section to allow for easily sourced transformers.
Vacuum tube or SS rectification of the Tube B+

Other specs TBD

[coldcathode]

I am thinking that for the purposes of this project we want to look at the following tubes

12AU7
6CG7/6FQ7
6SN7 for a really "Vintage" look and the "Octal" Fans out there

These are relatively low "medium-mu" Dual Triodes (meaning one tube for both channels)

They will need roughly 150V @ 30-50mA's, so this can be easily done by placing two RatShack 12V Trannies "Back to Back" 120 to 12V then 12V to 120V and use a Bridge or Half Wave tube rectifier.

Without getting into the age old debate of Tube Rectification VS. Solid State I will offer the following suggestions as to the choice.

Solid State bridges are:
Cheap
Reliable
easy to wire
Allow for greater capacitance immediately following them
BUT,
They aren't TUBES
Sound a little too "tight", "Tinny" or "Bright"

Tube Rectifiers are
Smooth sounding
TUBES
Look like tubes
But,
Expensive
can be "noisy"
need a filament supply
harder to wire and design

For the sake of project simplicity I plan to design with SS Bridge rectification. Anyone interested in the tube rectification will most likely have the "know how" to design it and if they like I can help if needed.

[danielwritesbac]

One ideal in first class audio is to have the gain stage run either its own private transformer or on regulators. It avoids having the speaker slam the gain stage with high amperage noise.

In this case, we need to keep up with a "vacuum tube mids and treble amplifier" that was already a high dynamics build. Now freed of its bass duties, its fierce competition to the subwoofer amp.

This is quite similar to a purpose-made TDA7294 mids and treble amplifier run along with a purpose made LM3886 woofer amplifier. Illustration: Mids and treble fall from the heavens as if a snowstorm of music, and that might be terrible if the woofer amp didn't compete / hold up.

Thus, the subwoofer amp (woofer amp) must Not sink the voltage of its own gain stage.

P.S.
I think that others who have no interest in vacuum tubes may substitute "LM1875 Class A Heaphone Amp on Regulated Power" for some of the vacuum tubes, including the gain stage of the LM3886 subwoofer amp. Results should be similar, but the only way to know for sure is buy a vacuum tube. lol! I did!

[danielwritesbac]

Originally Posted by coldcathode

To all,

This thread has officially been HIJACKED by myself (the OP) and has now become a collaborative effort (between myself & Daniel) to design a relatively simple TUBE/Gainclone Hybrid.

We need the more experienced solid state designers too!! Their maths are as beautiful as any artists painting, and this pleases the electronics greatly.

Let's please do get as much input as possible from AndrewT and Ratza.

I try to pay attention as far as possible because happy electronics are much easier to deal with! Lets please remember that its not audio science until both audiology and electronic engineering are done simultaneously.

[coldcathode]

EXACTLY,

The plan would be for the two sections to be "discrete" ie; The tube will have its own PowerSupply and merely send its amplified signal to the chip amp via a coupling capacitor and ground. Ground in this case is purely a voltage reference for the signal so it would be the "signal ground" of the chip amp. The Rin of the chip amp would also function as the output impedance for the tube. A Cin would be sized accordingly to the value of this resistor. ie; it forms a first order High Pass filter. In my case I will set an F3 quite low (big cap) for mids-highs or smaller speakers you could size as appropriate.

The output impedance has an effect on the gain of the tube so we need to size this a little bigger than 10K (I think, will know more as I look at the tube specs).


I DEFINATELY welcome any help/ideas/criticism etc from the other members of the forum.
Consider this an "Open Source" forum and chime in anytime. Please PM if you have attachments so I can download and file them.

[danielwritesbac]

Originally Posted by coldcathode

I am thinking that for the purposes of this project we want to look at the following tubes

12AU7
6CG7/6FQ7
6SN7 for a really "Vintage" look and the "Octal" Fans out there

These are relatively low "medium-mu" Dual Triodes (meaning one tube for both channels)

They will need roughly 150V @ 30-50mA's, so this can be easily done by placing two RatShack 12V Trannies "Back to Back" 120 to 12V then 12V to 120V and use a Bridge or Half Wave tube rectifier.

Without getting into the age old debate of Tube Rectification VS. Solid State I will offer the following suggestions as to the choice.

Solid State bridges are:
Cheap
Reliable
easy to wire
Allow for greater capacitance immediately following them
BUT,
They aren't TUBES
Sound a little too "tight", "Tinny" or "Bright"
. . .

While you're at the Radio Shack, their 6A05 diodes can make for an incredibly dull and inefficient bridge rectifier. The "looks like fork" SCE KBU808 is the tinny and bright. May I suggest a simple bit of decency with 1n540x series for the gain stage? http://www.diodes.com/datasheets/ds28007.pdf

Also, it would be really nice to run the gain stage regulated, if possible.