Opinions, suggestions, comments on a stereo 2A3 design

[Wardsweb]

Let me know your thoughts on this design I originally built back in 2003. What would you change, if anything and why? I have always liked the sound but not happy with the heat from the dual 7.5K resistors to ground.

It was built using a two chassis layout (power supply and output). The dotted line denotes the power supply chassis.

http://wardsweb.org/audio/2A3/DC2A3%20Full%20Schematic.jpg

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwardsweb.org%2Faudio%2F2A3%2FDC2A3%2520Full%2520Schematic.jpg&hash=9680d70fe8881dd4c5d3bf078ae17c7ede231d8b)

[Wardsweb]

Here is what the output chassis looks like.

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwardsweb.org%2Faudio%2F2A3%2Fcomplete640.jpg&hash=01f21d82f3e4449e4a7c9bfe7e66037dd6a549a1)

[Wardsweb]

Here is the power supply built.

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwardsweb.org%2Faudio%2F2A3%2Fpowerwired640a.jpg&hash=14baf8e04bf0c47b6a3dd407b5cc6d50c86cdbab)

[Doc B.]

Hey Luther,

Have you ever measured the frequency response? We did a direct coupled 6SN7/2A3 many years ago, that rolled off above 7kHz. We figured it was probably the Miller C of the 2A3 that was causing the problem. When we changed to a driver tube with a lower output impedance the response came up where it should have been. This was maybe 12 years ago and memory is a little foggy, but I think we must have changed to a 6SN7.

[Wardsweb]

Hey Luther,

Have you ever measured the frequency response? We did a direct coupled 6SN7/2A3 many years ago, that rolled off above 7kHz. We figured it was probably the Miller C of the 2A3 that was causing the problem. When we changed to a driver tube with a lower output impedance the response came up where it should have been. This was maybe 12 years ago and memory is a little foggy, but I think we must have changed to a 6SN7.

Sadly no, I don't have any real test equipment. The entire thing was built at home with all the chassis work being done in the garage. I do appreciate the input. I'll check around to see who might have a generator I could do a sweep with and possible a digital scope to map it out.

[edfallon]

I like the power supply concept.  I'd agree with Doc on the driver.  If it were me, I'd loose the SL7 and also the direct coupling.  I'm not a fan of that topology anymore because the bias voltage of the 2A3 is at the mercy of where the driver is operating.  It's great for a tweaker that is always monitoring the operating points but as soon as that driver tube ages, all bets are off.  Since you are already octal, a 6SN7 would be a natural choice and CCS loading would be nice too.  If you went cap coupled, you could also switch to 300B if you have a little extra current capacity in the power supply and get rid of the big power drop across the cathode resistor. 

-Ed

[Wardsweb]

I like the power supply concept.  I'd agree with Doc on the driver.  If it were me, I'd loose the SL7 and also the direct coupling.  I'm not a fan of that topology anymore because the bias voltage of the 2A3 is at the mercy of where the driver is operating.  It's great for a tweaker that is always monitoring the operating points but as soon as that driver tube ages, all bets are off.  Since you are already octal, a 6SN7 would be a natural choice and CCS loading would be nice too.  If you went cap coupled, you could also switch to 300B if you have a little extra current capacity in the power supply and get rid of the big power drop across the cathode resistor. 

-Ed

Thanks Ed - It has been sitting on a shelf unused for a few years now. May be time to reconfigure with the SN7.

[Paul Joppa]

Thanks Ed - It has been sitting on a shelf unused for a few years now. May be time to reconfigure with the SN7.

I had an amp once with a SRPP front end. It was convenient because the same cathode resistor worked well for both 6SL7 and 6SN7. They could be swapped with just a gain change (10dB of gain change!). Hands down, the SN7 sounded better.

Of course, since then I've abandoned the SRPP in favor of C4S loads, and that amp was converted to a 6J5. Then I got into parafeed so it just sits on the shelf these days....

[Grainger49]

Luther,

Let a non-Bottlehead employee say that is just awesome!  Your work shows a patience that I admire and aspire to.

Is there a medication that would help me?

As always you make me feel, "I'm not worthy!"

Keep up the good EXCEPTIONAL work!  You are a role model!

[Wardsweb]

Luther,

Let a non-Bottlehead say that is just awesome!  Your work shows a patience that I admire and aspire to.

Is there a medication that would help me?

As always you make me feel, "I'm not worthy!"

Keep up the good EXCEPTIONAL work!  You are a role model!

Wow that is some praise. Thank You !

I was raised to do a job right. That in combination with my own obsessive tendencies and things can become meticulous. Actually it isn't hard. It takes simple patience and the ability to map things out in your head. Think of it like a chess game where you think several moves ahead and the possible outcome of each.

My latest project has me veneering my DIY horns. For me that means the glue must be uniform and the left and right sides must be identical. That way the two sides facing each other are mirror images.

[Paul Birkeland]

Your build skills are quite magnificent!  If you can knock the B+ down a little, go with a cap coupled driver.  Try a 6SN7 at about 3ma with an 2.6k resistor under the cathode and a C4S plate load.  This will give you an amp with low input sensitivity, and allow you to really drive the 6SN7 with your preamp, getting adequate drive voltage for the 2A3. 

Also, the high frequency response should be much more even with the low Rp 6SN7. 

[Grainger49]

Luther,

I stink at chess.  My newest project will show that.  A choke almost touching the rectifier tube.

Think about the joke with "Plan Ahead" where "Ahead" runs out of room.  That's me.