Farting distortion!

[Arthur Dent]

I've got my strange amp built, and everything "works". However, it does not have anything resembling a pleasant distortion. It's completely clean and quiet up to a point, then it does a nasty, farty distortion with a monumental leap in volume once the signal gets loud enough.

Also, the volume pots are real scratchy.

Does anyone have a clue as to what this is? I can't see it being blocking distortion, as the levels involved are so low...

Here's the schematic:

[tubeswell]

I guess you could experiment with reducing one or more of coupling caps earlier on in the signal path (maybe after the first pentode?) to 1nF.

[Arthur Dent]

Actually, since my first post, I've replaced all of the post-pentode/heptode coupling caps with 4n7 caps. That helped a lot, but it's still unacceptable. Do you think going to 1n would be significantly better? Maybe I'll put in some 500p caps and see what happens (I haven't got any between 500p and 4n7 at the moment).

I'm going to go add an NFB loop.

[stokes]

If the pots are scratchy,check the caps feeding them for dc voltage leak.

[EETStudent]

Well, from my experience(although limited) blocking distortion can occur at low levels. As a matter of fact, I was working on adding some extra stages to a recent build at mine, and I also was getting blocking distortion at levels between 2-3.

The scratchy pots are often times a result of dc voltage across them. In this case, you could have a leaky coupling cap allowing dc to get through and into the pots. Hope this helps

[Arthur Dent]

Okay, the pots are no longer scratchy. I think the reason they were scratching was that a 1M grid leak resistor (the pot) was too big for a pentode, giving a small DC voltage on the grids and across the pot. I put 100k's across all the grids, which shut up the pots. I actually have those on the schematic.

It probably was blocking distortion causing the problem, or at least most of it. I changed the coupling cap after the input pentode on the "clean" channel to 470p, which helped vastly.

I also put in a negative feedback loop which, in combination with the reduced caps, has the clean channel sounding excellent. It's the polar opposite of a hi-fi amp, very nonlinear even when clean. It's quite warm with a very smooth break-up. No blocking distortion at all.

The high-gain channel still needs some work though. We're dealing with a LOT of gain here, and as such, there is still some blocking. I'll probably be reducing the caps in that circuit, as well as in the mixer, PI, and driver circuits. We'll see what that does.


Thanks for your help, it's greatly appreciated. It's a lot of fun using weird tubes.

[Bruce / Mission Amps]

I wonder if the tube was getting a little contact or grid leak bias going with the high impedance on the grid?

[Arthur Dent]

Yeah, that was my thought. A 6AU6 is spec'd with a 100k load on the grid for cathode bias, and a mere 10M for grid-leak bias. I'm assuming these tubes are similar. A 1M load is ten times the standard, so I could certainly see it getting a few mV of grid leak.

[Arthur Dent]

Well, I got the amp under control. The high-gain channel sounds superb, very touch sensitive for something having gain well into metal territory. The clean channel is a bit weak, though. I'll have to mess more with the cap in there.

I now have 470p coupling caps on the plates of the clean channel tube, the mixer heptode, and the PI. The cathode of the PI has got a .0047uf cap, which makes an enormous difference to bass response and almost none to blocking distortion as compared to 470p. All the post-triode coupling caps are .022uf.

With all those preamp pentodes and heptodes, there was only one tube I tried that was microphonic: an RCA 12AT7!