Somewhat noob questions: binding posts and preamp filament virtual CT grounds?

Fidot · 2871

0 Members and 1 Guest are viewing this topic.

Offline Fidot

  • Newbie
  • *
    • Posts: 3
Hi guys!

Please pardon my crappy drawing below, and if my questions are idiotic :). This is my early days of getting into amplifiers - and electronics in general.

For illustration purposes; I have attached my extremely shoddy diagram of grounds and preamp tubes' filament center taps - so that I can "talk over it".

- "safety" - chassis - grounds in green lines
- main amp / PSU grounds in dark blue
- filament section I care about in light blue.

Not all grounds are shown, just enough to illustrate my questions.


1. Grounding of OT's outputs to binding posts.

On the schematic, those are grounded into the "main amp" / signal path ground; however, on the amp, they are going to terminal 46 - which is IEC safety ground. Yes, technically it will be connected to "main amp" / signal ground via screw on 46 -> screw on 13 -> ...; but why not run a bit of wire instead? *Logically*, if I'm following what is done here, it would make the most sense to hook up black [ground] binding posts to terminals 5 and 16. Notably, it seems that at least schematic makes it a point of showing ground binding post grounded to the signal path ground.

2. Virtual CT preamp filament reference.

Those are the light blue wires; with filament power going to C4/C9, then over to B5/B9, and, finally, via two resistors to ground. What's confusing me is why resistors are hooked up to different spots, and not, say, B5 -> center socket pin; B9 -> center socket pin.

One possible guess I have is that:
  - it's not really that relevant; yes, path for B5 resistor to ground relatively to B9 resistor (B5 -> 28 -> screw in 28 -> screw in 13 -> 8 ) and (B9 -> 8 ) is longer and can add some resistance, but given that it's a resistive divider and two Rs aren't exactly the same, it's irrelevant, *and*, having filament voltage directly "under" the socket is Not A Good Thing.

.. but I would appreciate others' comments.

Thanks folks!
« Last Edit: October 29, 2024, 05:33:29 PM by Fidot »



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19750
1. Grounding of OT's outputs to binding posts.

On the schematic, those are grounded into the "main amp" / signal path ground; however, on the amp, they are going to terminal 46 - which is IEC safety ground. Yes, technically it will be connected to "main amp" / signal ground via screw on 46 -> screw on 13 -> ...; but why not run a bit of wire instead? *Logically*, if I'm following what is done here, it would make the most sense to hook up black [ground] binding posts to terminals 5 and 16. Notably, it seems that at least schematic makes it a point of showing ground binding post grounded to the signal path ground.
Since no signal current flows through this connection, it isn't all that important where it goes.  You could actually not connect this at all and the amplifier would work fine and you'd never know the difference (unless there was some kind of major failure).

2. Virtual CT preamp filament reference.

Those are the light blue wires; with filament power going to C4/C9, then over to B5/B9, and, finally, via two resistors to ground. What's confusing me is why resistors are hooked up to different spots, and not, say, B5 -> center socket pin; B9 -> center socket pin.
This is primarily to simplify the build process.  Since we do not use the chassis plate to carry audio signal current, it's not the end of the world to have the earth reference for the heater (not filament) winding attached to two different spots that are very close together.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Fidot

  • Newbie
  • *
    • Posts: 3
Thanks Mr. Paul!

Re: binding posts <-> chassis ground:

Thanks again :). But then, if this isn't required, then why was it done in the first place? Sorry for being a bother; but I am genuinely curious _why_ you guys decided to do things one way or another - trying to learn some design principles here in addition to just being a meatbag soldering machine :).



Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19750
Imagine that the output transformer primary and secondary develop a short, then you have 50V DC on the binding posts.  An unsuspecting individual walks up and touches an all metal banana plug on the binding post and gets shocked. 

There are also some minor treble performance benefits to providing an earth reference for the secondary, though your mileage may vary on that one depending on the output transformer! 

The potential shock hazard as a result of a failure is mostly why we have been doing this, though now there are lots of full bridge class D amps running single rail DC power supplies that have 40V DC on the speaker terminals and not too many people complain...

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Fidot

  • Newbie
  • *
    • Posts: 3
Once again - thank you sir!