Lead Dress for Star Grounding
Moderators: pompeiisneaks, Colossal
Lead Dress for Star Grounding
I'm considering star grounding my next amp (5E3) and was wondering. Since star grounding can mean more wires back to the post, it could create a spider web in the chassis. Does it make sense to dress the wires together, perhaps twisting them with branches off to the circuit points and back to the grounding post? Would twisting the different grounds from all parts of the circuit (preamp, output, PT, OT, etc.) all together cause problems negating the idea of star grounding through induction problems? Or would twisting them all help with noise more than separate runs?
Re: Lead Dress for Star Grounding
Good planning! Running extra ground wire is the bane of star grounding.
And good thinking - ideal versus real world is always a consideration.
If it were me, I would feel free to bundle the ground wires neatly as they head back to the star ground point, with two exceptions. I would route the PT centertap wire to the negative of the first filter cap as short and direct as possible and keep other things away from it. I would route the speaker jack return with a single wire back to the OT common lead. The others, I'd bundle neatly.
The major point of star grounding is that ground currents in the wires cause voltages from end to end on the "ground" wire because of its resistance and (for high frequencies) its inductance. Splitting the various ground return currents into separate wires greatly diminishes this. Induction of voltages along a ground wire is possible, and more possible the higher the frequency of the magnetic field doing the induction. For audio frequencies, you are probably going to be safe in ignoring this except in a few cases. The M-field leakage from the PT is the biggie there. Don't run the grounds around the PT, run them all to one side of it if they have to pass it. The PT centertap wire to the negative of the first filter cap has big and pulsed (containing higher harmonics) currents running on it. Run that one short and direct, and don't cluster it up with anything else. I don't think twisting the ground wires would help or hurt as a practical matter for audio. They're not differential signals, so this doesn't gain you any particular noise immunity, and might increase the per-wire inductance - which probably won't matter for audio anyway.
And good thinking - ideal versus real world is always a consideration.
If it were me, I would feel free to bundle the ground wires neatly as they head back to the star ground point, with two exceptions. I would route the PT centertap wire to the negative of the first filter cap as short and direct as possible and keep other things away from it. I would route the speaker jack return with a single wire back to the OT common lead. The others, I'd bundle neatly.
The major point of star grounding is that ground currents in the wires cause voltages from end to end on the "ground" wire because of its resistance and (for high frequencies) its inductance. Splitting the various ground return currents into separate wires greatly diminishes this. Induction of voltages along a ground wire is possible, and more possible the higher the frequency of the magnetic field doing the induction. For audio frequencies, you are probably going to be safe in ignoring this except in a few cases. The M-field leakage from the PT is the biggie there. Don't run the grounds around the PT, run them all to one side of it if they have to pass it. The PT centertap wire to the negative of the first filter cap has big and pulsed (containing higher harmonics) currents running on it. Run that one short and direct, and don't cluster it up with anything else. I don't think twisting the ground wires would help or hurt as a practical matter for audio. They're not differential signals, so this doesn't gain you any particular noise immunity, and might increase the per-wire inductance - which probably won't matter for audio anyway.
Re: Lead Dress for Star Grounding
Just to confirm and clarify, the CT of the high voltage winding should be directly connected to the negative of the first filter (reservoir) cap, rather than to a lug at the star ground?
My band:- http://www.youtube.com/user/RedwingBand
- martin manning
- Posts: 13403
- Joined: Sun Jul 06, 2008 12:43 am
- Location: 39°06' N 84°30' W
2 others liked this
Re: Lead Dress for Star Grounding
It might be slightly better to run the CT direct to the reservoir negative lead, and then run a lead to the main ground from there, but I have found that as long as the reservoir and the PT CT are joined at the main ground, preferably on the same solder lug, there is no issue.
Re: Lead Dress for Star Grounding
It might be helpful to think in terms of what is sometimes referred to as "constellation grounding" where each stage has a local ground point which connects to the decoupling cap and whatever other stages are connected to that cap, then from the decoupling cap on to the main star ground. This limits the amount of spaghetti you have to deal with.
I don't know if you will notice much, if any, difference on a 5e3 with a star ground vs a well implamented bus ground.
I don't know if you will notice much, if any, difference on a 5e3 with a star ground vs a well implamented bus ground.