diytube.com

Looking for ideas/suggestions on this circuit. I "borrowed" the pre- section from the Frankenhome schematic, then substituted a 6L6 power section in place of the 6V6 pair. I have never seen a 12AT7 used in this manner so I don't know if the values or layout are correct, but I'm looking for a simple, three-tube system to start with. I'm not looking for a lot of volume, I've got a pair of RTR bookshelf speakers to drive and I'm hoping something like this will lead me to a decent, simple stereo outfit.

You'll need feedback with pentode 6L6s. I don't know how you can put in global negative feedback with that phase splitter. But then you can put the 6L6s in triode and there you go. That 12AT7 might have about 20 - 25 gain, don't know if that's gonna be enough.

You can lower the number of tubes if you use a 14GW8 (PCL86) they are pretty cheap $5. Each tube has a triode driver and a power pentode good for about 12 watts out PP.

Something like this?

I wonder how well a 6U8A would work in place of the12AT7.

Your "NFB" point is going directly to ground, there.

You'll have to make a divider to the grid of the triode that's currently bypassed right now. You'll also have to get the anode connections right on the 6L6's, ot it'll oscillate.

Ah, you are correct. This is actually how it's connected on the other schematic I was looking at.

Still not seeing how that would work, as one tube will have positive feedback at that point.

Also, your bias takeoff is in the wrong spot. You'll have no bias to the 6L6 that way.

Also, your bridge is shorted.

I'll post a schematic suggestion in a bit.....

Cheers!

Try this....

Rfb will be used for your feedback. Cx across it will be a picofarad value under 1000p (depending on size of Rfb) and it's chosen for the best 1KHz square wave on an oscilloscope.

Rdiv is part of the feedback divider. I'd start with 1K. If you're getting crazy high values for Rfb, make Rdiv lower. If you're getting too low of a value for Rfb, make Rdiv higher. (choose Cx AFTER you settle on the resistors for adequate NFB. For 6L6 in pentode and HiFi, you'll be looking at 20dB+ or so of NFB)

Choke was in the wrong place for HiFi amps as well.... did you get this circuit from a guitar amp?

*edit*
Better make that 0,1uF cap on the feedback line 1uF (250V polypropolyne film), just so the gain doesn't skyrocket at LF

Cheers!

Ok, that makes a lot more sense. And yeah, I did take the bias feed off the wrong spot on that schem. It's in the right spot on my actual circuit layout.

Yes, the power section was from a guitar amp design.

That gives me a whole lot more to work with, thanks! I'm still wondering about that 12AT7 not providing enough gain, but we'll see. I have boxes of tv/radio tubes to play with too, but again I was hoping to keep it simple and use common parts.

If you put another tube in front, that would simplify the NFB even further (it'll tie to its cathode and no need for a blocking cap) and give you the extra gain you'll need because the amount of NFB required to linearize a pentode-mode 6L6 has been in the 25-30dB IME.

Cheers!

If you have a 6D10 in that box of radio/TV tubes you will have the equivalent of 3 12AT7s in one tube so you could have an amplifier front section and a long tail pair phase splitter.
Tube count still 3.

No 6D10 in the box.

The circuit above works. It's actually louder than I thought it would be but I've got a serious hum issue that I need to track down. BUT! I've got a working circuit that I can play with and that was the goal. Thanks guys!

Good news!

With an oscilloscope, check to see if your hum is an oscillation. If not, then it's just a layout issue.

Cheers!