DIY Audio Projects Forum

If the shrillness is at low listening levels as well as high you are not overdriving the 6SN7. The only time I found this is with a 12AX7 driven direct from a player with no upfront attenuator. The bias on the tube was low and easy to overdrive. Actually changing the tube to a12AU7 fixed the problem.

Try a different 6SN7.

_________________
Projects: "Precision" - 300W Class AB Holton power amp | "retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp |

Yes, different tube brands have their own personality. Some really complex science and math behind it to make one just say, it is what it is

_________________
-= Gregg =-
* Ratings are for transistors - tubes have guidelines*
Home: GeeK ZonE
Work: Classic Valve Design

Yes, the shrillness does seem much lower (or non-existent) at lower volume levels. I will get a new 6SN7 on order.

Thanks...

PS Greg: I am learning how true your note is about the guidelines. 30+ years doing digital stuff and it generally either "was" or "wasn't". I am learning that with the glass it is "maybe" or "perhaps." Bless the glass!

Randy

You need to stop and think about the design. I seriously doubt that your issues are from overdriving the 6SN7. I think they are because you are overdriving the 6V6 (by a lot!). Let me explain.

In the V2 design, the driver is biased to about 4.0v. It also has a gain of approximately 16 v/v. This means that if you overdrive the 6SN7 by giving it an input exceeding 4.0v, the output voltage would be 4.0*16 or ≈ 64v peak. Now the power stage is biased to about 16v (give or take). This means that at 4.0v into the 6SN7 you'd be overdriving the 6V6 by about 48v. This would cause any power tube to become a little "shrill".

One of the ideas behind the V2 was increased sensitivity. In fact, it one takes about 1v peak to drive the amplifier to maximum output. My guess is that you are seriously overdriving the 6V6 output stage. In fact, in this design it is IMPOSSIBLE to overdrive the 6SN7 without first seriously overdriving the 6V6.

Don't bother with a new driver tube unless it just for tube rolling. Play with some other 6V6s and watch that volume.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

You are right Matt. I've spend a couple of weeks now looking at everything with the scope, took the time to study/understand the load line graphs and tube characteristics, and reanalyzed your design with that knowledge to understanding of exactly how things worked. That, as well as a comment MWHouston made caused me to start paying better attention at the lower volume levels.

Everything is working as it should.

I did order some different tubes to "roll" and I am astonished how things can change. None are "bad" I just notice that my ears like some better than others. All of my unfinished digital projects have been shelved so I can continue learning more about this amazing hobby. Looking forward to my next (second) project - perhaps the 4S. With all the Forum activity on that project, it appears a great opportunity to continue my education.

Thanks!
Randy

Hi Matt,
New to tube audio, seriously considering the Lacewood V2.0 for my next project. Couple of questions:
-Why use the bare copper ground bus rather than the nice low impedance chassis for ground?
-Is this common in tube audio? All the tube radios I've built or repaired have a common chassis ground.
-The schematic of the amp shows ground symbols, but the power supply shows only a single point ground with two power supply lines (+,-) as outputs?
-Is this to isolate the AC in earth ground from the circuit except for the single point ground?
-Any other pointers for a newbee?
Thanks,
John Kennedy

Hi Matt,

A little unrelated, but I'm working on a power supply for a single-ended EL84/6BQ5 amplifier (RH84) & have been looking at the power supply for the Lacewood. As shown I've modelled the power supply in PSUD2 giving 3.2mV of ripple at the last capacitor, although the rectifier will actually be a 6BY5GA or 6AX5GT.

As your supply is basically similar initially (CLCLC), do you see benefit in splitting the supply with two 1H chokes as you've done for channel separation?
(Same Triad C-24X as used, and there is a further RC network (10k/10uF) before the ECC81/12AT7 plate resistor.)

The short answer is yes. Although I have not done any specific measurements on channel crosstalk, subjectively, I feel the depth of soundstage is just markedly better with the isolated channel supplies.

I have not settled on any one topology. In general, with the added power stage sensitivity of pentode and UL amps, I don't believe that separate chokes are required for the signal sections. But that is dependent on the level of signal drive required.

In general, because the driver and power stage are generally 180° out of phase, I prefer the method used in the 6CY7 amp and the Lacewood rewire. In both cases the power channels are separated each with an L-C section and then the drivers for each channel are driven from the B+ for that channel with additional isolation. This always seems to produce the best soundstage. That being said, I am working with a new amp, we'll call it the "6V6 Marblewood", that uses four separate filters sections for the two power and two signal stages. I'll have to compare that amp with the 6CY7 and Lacewood to see which sound stage I prefer.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog

Love to see the Maple wood base. Matt, I think you love your woodwork as much as your electronics. Fortunately this hobby allows you to enjoy both. Though I have my bases made by my pensioner amp building buddy, Don, I like to do the final sanding and findihing then decide on a finish. I really don't miss being better with lumber happy for someone else to do that work.

A hot iron and a glass of red and I'm happy.

_________________
Projects: "Precision" - 300W Class AB Holton power amp | "retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp |

Thanks Matt. I will certainly consider it then. Chassis space may be an issue though.