Gainy multi-pentode amp acting bad

[Arthur Dent]

After finally winning the battle against blocking distortion by direct-coupling the PI, the clean channel of this amp sounds brilliant.

However, the high-gain channel is still nowhere near where I'd like it to be. The distortion begins at an extremely low input level, at which point it has that sort of cheese-grater-on-concrete sound that nobody in their right mind would like. It then gets louder and sounds better with higher input levels and higher volume knob positions.

Am I right to assume that the reason for this is that something is clipping in an extremely asymmetrical manner? Where would this be happening, and why?

[tubeswell]

My two cents worth of simple understanding is that some assymetrical clipping is the norm for tubes. However, if the assymetry is too extreme, you will get 'grid current limiting' resulting in blocking distortion - if the grid is biased too close to 0V w.r.t. the cathode. If on the other hand the grid is biased too negative (where the -ve AC swing from the grid signal wants to drive the plate to a voltage that it can't get to because it has reached the limit of what the B+ will allow), you will get plate cutoff, which is a hard distortion . To get the cleanest signal, the bias point for the grid has to be somewhere near the middle of the load line.

So you might have to experiment with different biasing for your pre-amp tubes (assuming your output tubes are biased somewhere in the 'working' ballpark)

[Steve Conner]

Sounds like some stage in there is biased either too hot or too cold. (I'm going to guess cold.) Check your preamp tube plate voltages, they should all be in the ballpark of 1/3 to 2/3 of the B+ voltage. If you have anything funky like a Marshall-type cathode follower or a cold-biased stage ala SLO100, be aware that these tricks might not work with a pentode.

Of course with pentodes you have screen resistors and screen bypass caps to worry about. Too much or too little screen voltage could cause badly asymmetrical clipping, and if you're using a pentode that wasn't meant for audio, you'll have no recommended values to get you started.

Finally are the tubes sharp-cutoff, or vari-mu?

[Arthur Dent]

That's odd, the schematic didn't actually get on there. Well, as you can hopefully see this time, the problematic preamp's setup is: triode, volume, pentode, triode CF, "James" tone stack.

Originally, that cathode follower was causing some issues, so that was good advice Steve. I now have that triode biased properly (grid at -1.5Vdc wrt cathode).

I believe the 6BA6's are remote cutoff pentodes. I can also put 6BE6 heptodes in place of the pentodes, as the pinouts of those tubes means that a direct swap will result in a heptode strapped as a pentode. It has a different sound that way, but it does the same thing.

Today I guess I'll be messing with the bias, and if that doesn't work, I'll go and change the screen voltage.

[Arthur Dent]

Okay, I've tried changing around the cathode and screen resistors on the pentode enough to be fairly sure that their values are not the problem. I'm going to have to go ahead and give up on that preamp and rearrange it. The new setup will be: pentode, volume, 12AX7 stage, 12AX7 cathode follower, tone stack.

[Arthur Dent]

Well, I rearranged the preamp, and the problem persisted. It seems I don't really know how to properly wire a 6BE6 as a mixer. I used resistive coupling and used grid 1 for both preamps, with grid 3 wired to the cathode, and the distortion problem disappeared.

Does anybody know how a 6BE6 mixer is supposed to be wired?