Merlin's next book

[Phil_S]

What Merlin says about filter caps is exactly what I was taught by a couple of old timers who were hanging out at another on line forum when I was learning. In my non-tech mind, it is something that makes perfect sense.

Think of it like this. Each preamp stage terminates at the cathode. Fine all the ground connections that are leading up to that cathode, gather them together, and ground them at one point. If you are using a buss, group them on the buss. If you are using a star, it is (usually) a floating star. Then gather up all the grounds between cathode 1 and cathode 2 and do the same. If using a star, run all the floating stars to a single earth ground. If using a buss, run a wire from one end to the earth ground. Sometimes stages share a ground point, for example, when one filter cap feeds both halves of a dual triode, there is nothing you can do to separate the filter cap ground, so both cathodes and their respective ground wires are all one group. I've employed this design principle in many the amps I've built (about a dozen) and it works well. Because there is an overall guiding logic, it is hard to get lost.

On occasion, I've put an amp into a donor chassis that was cramped and put the filter caps in a bank due to the space considerations and that worked out OK, too. I had one of those that was noisy. I solved the problem by re-routing the grounds to their "proper" respective places, but it was messy. After that, I've always made an effort to place the filter cap with the section it supplies.

FWIW, this is the 2 cents of an amateur builder.

[Structo]

Thanks for the explanations guys. It helps me understand that method better.

But there are more than one way to skin a cat and I have seen a couple other methods and thoughts on grounding that also work well.

One way is the way Dumble grounds his amps sort of multi-star plus the ground buss.
That seems to be a very quiet scheme as well.

Another that is similar is how Larry (the guy from Germany) grounds his Marshall style amps.
His thought was that each section should have it's own star ground around the chassis.
He actually grounds his heater circuit at the input point.
His theory is that that is the quietest point in the chassis and that grounding there will be much more quiet then grounding near the power section.

Not trying to start any arguments because each method seems to have it's merits.

[Cliff Schecht]

Yeah there are dozens of ways to ground an amp and not one of them is necessarily wrong. It's all about keeping the sensitive inputs away from the noisy stuff. It's just like doing any of other form of mixed signal design, you control your grounds not only to prevent ground loops but also to prevent the circuits from crosstalking in anyway.

[Phil_S]

But there are more than one way to skin a cat and I have seen a couple other methods and thoughts on grounding that also work well.

I guess you'll see plenty of discussion, but really no argument on this point.

One way is the way Dumble grounds his amps sort of multi-star plus the ground buss.

You might think of the buss as an extended floating star, if that helps....maybe it actually makes it worse. Well, yes, whatever works. It is hard to argue with success.

[KellyBass]

Just for fun (to me at least) recently, I took a relatively simple schematic...a 2x6v6 Traynor YBA-something, and did a layout using Merlin's grounding and power supply principles.

Man, it sure looks different than Traynor's layout...but it makes more sense. Now, I'm really wanting to build this amp! Anyway, it was a good exercise on a sunday night. I did end up with a dull headache, though.

[aclempoppi]

Yeah, I know what you mean about the dull headache. Turned a simple stand alone reverb kit build into a grounding scheme education. Used a piece from this tutorial and that tutorial. But the combination worked out just fine, dead quiet.