The late observations of Francesca

[Mark]

I finally found time to have a look at the Francesca photo's. While I suppose this has possibly been mentioned before I'll give you my thoughts on the matter in case it hasn't been brought up.

I have been very curious about the earth paths in this amp and of course how to avoid any earth loops. The earth points in the circuit are the input jack, the output jacks, the screws securing the output tubes, the screw providing the earth point for the power supply caps, the buss bar across the back of the pots, and the mains earth in the corner of the chassis.

I think there are a couple of things about the earths that are worth mentioning. The screw that provides an earthing point for the power supply caps has three black wires coming from the tags.

One wire is for the plate supply (two 40uF caps in parallel) and the screen supply (single 40uF cap mounted along-side the other two 40uF caps.)

Another wire is for the pre-amp supply.

The final black wire is for the P.I. driver stage supply.

I don't claim to know much about earth loops, but it is strange that most of the earth points of the amp go to the buss bar across the back of the pots (including P.I. driver stage via the 5K presence pot.) Yet Ken Fischer has used a separate wire for the P.I. driver earth when it would have been just as easy to extend the pre-amp power-supply buss across all three 20uF caps, I find this quite strange?

The bias supply does make more sense then the ones that I have previously seen (a bias pot failure won't cause the output tubes to be destroyed.) However the bias earth goes to the chassis earth point in the corner of the chassis which is odd as it could have just as easily have been routed to the power supply earth point. While I don't discount convenience for the location of this earth point, I am still a little curious about this choice of location.

I'd love to hear any answers or theories on the matter, again sorry for the lateness of this post.

[Elcabong]

The bias supply does make more sense then the ones that I have previously seen (a bias pot failure won't cause the output tubes to be destroyed.)

Mark, can you explain how it is going to be a safer bias supply?

[LeftyStrat]

Well, I'm not Mark, but this is one I can answer. Some bias supplies are wired in such a way that if the pot goes bad, you have no negative voltage and as a result your tubes are biased too hot and can fry.

The better approach is to wire the pot so that if it fails, you end up with more negative voltage resulting in the tubes being biased too cold. That way you potentially save the tubes in the case of a pot failure.

If you take your bias voltage off of the wiper, and the wiper lifts off the resistive pad, you'll have no negative voltage at the bias inputs.

Take a look at the November schematic on the ax84 web site. You can see that if the wiper fails in this configuration, you end up with more negative voltage.

[Elcabong]

That's clever! I guess my eye was not trained enough to notice that detail from Francesca's photo.

I will convert my bias circuit while it is still on paper.

Thanks for the explanation.

[Mark]

Dear Lefty Strat

Thank you that is exactly what I would have said.

I'm a little surprised there aren't any comments on the earth wiring. I suppose in the scheme of things it does matter that much, but I suppose it is would trying.

[Elcabong]

I am not a "ground scheme" expert either but I also see some peculiarities in Francesca's grounding pattern. It is a mix of buss, star and distributed grounding; there is nothing wrong with that but it makes it harder for us to analyze the intent of the builder.

Especially for the points you mentionned: I would have expected the PI supply and stage grounded at the same point. And grounding the bias supply to the chassis earth is for the least unorthodox.

[LeftyStrat]

Actually I do find it interesting. Hopefully I will get to experiment with the how it affects the sound soon. I have my chassis and main board done. I just need to do the power supply.