Grounding points

[WhopperPlate]

I was wondering, is it preferable and does it help grounding efficacy to solder all leads to a single lug on the chassis , as opposed to multiple lugs on a single connecting point?

I have used lugs for each lead on a single point attached to chassis and I have been considering whether or not this is ideal , as I have noticed amplifier layouts like dumble tie every connection to a single lug at the same point. Am I overthinking this? Is it better to tie everything to a single point at the lug before it connects to chassis ?

Thanks

[xtian]

A matter of aesthetic preference only.

[WhopperPlate]

Simple. Thank you .

[brewdude]

I would prefer to put 2 or 3 of this tabs stacked on the same bolt with that many connections. It simply gets too crowded for my taste.

[mhuss]

> 2 or 3
Exactly. Use as many as needed but no more.

[WhopperPlate]

Part of my reasoning for inquiring is the concept presented below. I figure if moving the connection closer between components before they reach ground can reduce noise then it possibly could have an effect elsewhere …

[chaccmgr]

The reason for having a good grounding is to avoid the risk of any high current flows in areas where they can harm. There are many commercial designs seemingly ignoring this and getting away with it (e.g. Dumble placed ground connections wherever they were convenient to reduce/avoid wire lenghts, used the chassis as the return path and the amps are dead silent), but us hobbyists spend too much time chasing errors, hum and buzz when we don't stick to proper grounding.
Grounding components, or in your case, the center tap of the PT at the negative side of the capacitor they feed is in general always a good idea. This is also valid for e.g. controls on the front plate. All ground connections should be connected to the neg side of the filter capacitor that feeds the stage.

Nobody knows, how much time the Fenders, Dumbles and other commercial amp makers invested to get rid of issues related to poor grounding until their designs were good. We just look at them and think, " hey, if they don't stick to star grounding, why should I?"

The problem with us is, that we build most amps one of a kind and investing weeks in troubleshooting is the bad side of the hobby.

there are excellent websites out there dealing with proper grounding like the Valve Wizard, R.G.'s Geofex (a bit cumbersome to search), Aiken Amps and others.

[martin manning]

I would prefer to put 2 or 3 of this tabs stacked on the same bolt with that many connections. It simply gets too crowded for my taste.

I would rather not stack ground lugs since each interface can create some resistance. Ground lugs with two lead holes can easily accommodate four or five leads.

The reason for having a good grounding is to avoid the risk of any high current flows in areas where they can harm. There are many commercial designs seemingly ignoring this and getting away with it (e.g. Dumble placed ground connections wherever they were convenient to reduce/avoid wire lenghts, used the chassis as the return path and the amps are dead silent), but us hobbyists spend too much time chasing errors, hum and buzz when we don't stick to proper grounding.

I have to disagree with your second point, as there are very good reasons why the Dumble ground scheme results in a quiet amp. He seems to have incorporated features from Fender and Marshall, among others. There was some evolution, as you can see that in the early amps the last filter was located on the input end of the chassis, which was later deemed unnecessary.

Grounding components, or in your case, the center tap of the PT at the negative side of the capacitor they feed is in general always a good idea. This is also valid for e.g. controls on the front plate. All ground connections should be connected to the neg side of the filter capacitor that feeds the stage.

This is the "local star" concept discussed in the references below.

Nobody knows, how much time the Fenders, Dumbles and other commercial amp makers invested to get rid of issues related to poor grounding until their designs were good. We just look at them and think, " hey, if they don't stick to star grounding, why should I?
The problem with us is, that we build most amps one of a kind and investing weeks in troubleshooting is the bad side of the hobby.

Fender's ground scheme seems to be derived less from theory and more for convenience and/or ease of manufacture. The most important features are there (e.g. reservoir and PT CT grounded close together and well away from preamp grounds), and It works well enough, but it can be improved. See: https://ampgarage.com/forum/viewtopic.p ... 77#p428877

there are excellent websites out there dealing with proper grounding like the Valve Wizard, R.G.'s Geofex (a bit cumbersome to search), Aiken Amps and others.

+1!

[chaccmgr]

I have to disagree with your second point, as there are very good reasons why the Dumble ground scheme results in a quiet amp. He seems to have incorporated features from Fender and Marshall, among others. There was some evolution, as you can see that in the early amps the last filter was located on the input end of the chassis, which was later deemed unnecessary.

My point was not to say that Dumble did not know what he did - its the opposite. My point is, us mere mortals looking at "non text book" grounded amps might be lead to the wrong assumption that it might not matter where to make ground connections. Grounding wherever convenient CAN (you can get away with it if you're lucky) cause trouble and according to Murphy's Law it eventually will (sooner or later)!

If Fender's layout was intended or not, the forensic analysis of Richard Kuehnel shows, that it is close to the optimum you would do if you did a theoretical evaluation upfront.
...not sure if my english is good enough to point out what I wanted to say...

[martin manning]

Ok, that's clear now. I'm interested in Keuhnel's assessment of Fender's ground scheme. Can you point me to that?

[chaccmgr]

Here you go: https://ampbooks.com/mobile/amp-technology/layout/


Hmm, just read over it again: It's not about the ground scheme, it's about all other layout aspects...sorry, but a good read anyway

[TUBEDUDE]

In my opinion it's best, but frequently impractical, to ground the circuitry associated with a tube to the negative terminal of the capacitor feeding it. Then run those grounds to the Xfmr bolt, buss, or whatever you're using for the chassis ground.
I take the grounds for all signal tube circuitry to a low Z buss grounded at a Xfmr bolt. Separate ground wires run from the output stage tubes to the same bolt, and of course the incoming power ground has its own connection directly to the chassis. I do strictly point to point and this works well for me, YMMV.