The whole point of removing a ground loop is to prevent the signal ground from running through the equipment ground.
Not sure I understand what you're trying to say here, as there's no way to prevent it simply because instrument cables short the two together whether you like it or not.
If you put the pup grounds on the cavity ground lug, then you must bring ALL of the signal grounds there to achieve star grounding, including wires from those pot legs that used to soldered to the shell. Joining all the signal ground wires together (as in the drawing)is star grounding, without connecting them to the equipment ground.
This works, and the drawing is reasonably accurate. (By the way, the red X's in the drawing mean to cut or remove those connections.)
Yeah, I understand the drawing.
You're missing something. I qualified my response due to the fact that the drawing shows a jazz plate, which effectively defeats a true star ground whether you want it to or not. I also said if you have a pickguard then the story is different.
If you REALLY want to isolate the shield can from the signal return then you have to do more work... you have to either use an insulating jack (with a metal plate, like on a J) or cut any shield foil around the jack to open the circuit there (with a pickguard). You didn't mention any of this, so I have to assume there's a short circuit there (which is typical).
I also strongly maintain, and challenge you to contest with data, that, if you have .0x ohms of DC resistance, you "effectively" don't have a ground loop. Period. Because for any significant voltage potential to develop over such a small resistance, the overall circuit voltage would have to be massive... enough to injure someone.
In fact, I think this is dealt with earlier in the thread as well... the whole ground loop thing is so incredibly insignificant, I seriously doubt it'd ever be audible.
You know what a ground loop is, right? It's where you have more than one "reference" and they're electrically DIFFERENT from each other, which means a potential (voltage) can be measured from one to the other. If you don't believe me, look it up. If you've got all reference points in the circuit a fraction of an ohm away from each other, at best (or worst as it were), it's just not going to be an issue. Period. White elephant, red herring, waste of time frettin' over, take yer pick.
