 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
Well - if the quality of the wire is as important - what about an 'optimum' ground scheme (if there is a universal method)?I'm up against grounding a dual-mono pentode tube amp which has two main B+ transformers (one/channel) with heater transformer grounds etc. My heath is rather crappy energy and coordination-wise so I only want to do the exercise once.
One shcool of thought will say a buss system is ok and can be attached to the chasis at either the power supply area or near the inout jacks, others prefer a star/'galaxy' system with a 'star' for each involved stage and these are grouped to ground somehere to the shassis.
Theres also multiple point chasis grounding shich surprisingly can be usable - I've heard worse things happen then using a buss! - a friend build an elaborate amp with toroidal power transformers, adn a #4 buss sytem - hummed and buzzed like holy hell!
Sakuma likes to use a '2-point' system - with the major power-supply grounds tied to the chassis near the power transformers and has another ground systme for 'signal' including the input and output connectors and various cathode returns - can you see any advantage in his method? - it must not be terrible as Sakuma employs Altec VOT speakers which would show the 100Hz rectified hum component nicely.
Might the 2-point ground dissipate some 'hash' from the power supply into the chassis?
Also - I do suspect long ground returns to invite noise and oscillation -my parallel 2A3 amp has a chassis 33" by 18" - those returns to the buss are long and the buss itself probably 2 feet long.
best wishes and hope for a nice thanksgiving!
Freddy (listening to John McCormick 'Sentimental Songs')
PS - with the ground returns - what wire would you suggest without getting into exotic areas or being physically unmanageable?
Follow Ups:
It is very hard to predict exactly which grounding method will provide the best results. What works best for hum and ground loops, may not sound as good as another grounding scheme.Theory says a true star is the best, but practical real world constraints will not allow a true star to be used in many cases, just too many ground wires, and they get too long. Litz wire might help here, I once braided a ground line of several feet, and got the induictance down to 1/4 of the single wire value.
Essentially, whatever experience and practice shows works best.
I am leery of grounding chassis at more than one point, and this point should be very carefully chosen.
Jon Risch
Hi Jonthanks so much for the reply - very interesting that you went with a braided ground (was each wire insulated? - I'm a bit confused on your low inductance scheme)
Since you seem to have a lot of exprience and savvy for layouts, may I ask what places withing a poewr amp which you would avoid grounding? - sorry to phrase that so poorly....
A few years back, I built a push-pull parallel 6550 bass guitar amp for a friend as a present - and used a buss system perhaps 11" long and the various ground returns went to this 'floating buss' then i gorund taht to the chassis near the fist power supply cap - there was a lot of 'hash' pickup - don't know if that wsa realated to the ground or to a isolation transformer under the chassis used for a bias supply. At any rate the finished amp was noiser than the Tryanor amps which have NO shielding of their input wires and perhaps 3-soldered grounds to the steel chassis!
Also the same friend built a dual mono pentode/UL/triode amp using toroid power transformers and a very thick (#4 or heavier) ground buss - this had a lot of buzz too but some of his problems may have been related to the close proximity of the toroids to the input tubes and his insitance of using DC ont eh 6550C heaters. My Eico HF-87 with little B+ capacitance and multiple grounds was much quieter.
Both of these amps lead me to believe the chassis grounding isn't too bad - at least for musical instrument amps.
I appreciate any generalities you might make on what should and shouldn't be done in a ground system for a tube power amp with the transformer(s) on chassis.
Best regards!
Freddy
I employ a simple method;
1. I make all rectifier diodes and reservoir capacitors return their grounds directly to the respective power transformer secondary then attach all of this to earth.2. Make a "clean" ground starting with the active regulators if any and use this as the start of a buss
3. Then I will make a star for the output circuit and connect this close to the regulator
4. Lastly I will make a star for the input section and connect to a point on the buss that is furthest away form the buss.
5 The chassis gets only one point connected to earth ( a true shield passes no current)
6. Filaments get their own ground bus following the same rules above
Other than this there is no way to predict ground loops that I know of.
As for materials a nice piece of fat silver wire is a great ground. as for oscillations a fatter piece of wire for the buss rather than thinner should keep this in check.
check out this link for a silver wire source.
http://www.meta-gizmo.com/Tri/Dancing/DrG_silverT.htmlRobert Morin
Thank you very much Robert - this and the follow-up paint a clear grounding scheme. I've seen where some prefer to tie the final ground to chassis at the input jack rather than near the input cap and CT - what happens when your ground scheme has its chassis connection point 'flipped' from PS to input jack area? - is it anything predictable from one case to another?Also when using two trasnformers in a 'dual-mono' amp - would you make a common ground near their CT/rectifier points or try the tedious method of keeping the two (left and right) busses separate?
best regardsFreddy
As long as you can keep the ripple artifacts from contaminating the audio portion of the ground, than anything you want to try is fair game. I have looked over the shoulder of some really smart guys and they can't predict what will happen either.
As for the dual mono amp, a twin set of grounds will hold crosstalk in check better. I don't know if it is worth the trouble though. I do try to do twin grounds on mic preamps 'cause the signal comming off of a mic is just a whisper. I have not tried this on a power amp.Robert Morin
Hello, Morinix
Thanks for your suggestions. I am trying to learn more about grounding, since my DIY preamp hums a bit.I have a few doubts regarding your post:
> 1. I make all rectifier diodes and reservoir capacitors return their grounds directly to the respective power transformer secondary then attach all of this to earth.
> 2. Make a "clean" ground starting with the active regulators if any and use this as the start of a buss
> 3. Then I will make a star for the output circuit and connect this close to the regulator
> 4. Lastly I will make a star for the input section and connect to a point on the buss that is furthest away form the buss.
This is where I got lost...away from what?
5 The chassis gets only one point connected to earth ( a true shield passes no current)
Īs this earthing point the same as "1" above ?
By the way Sakuma recommends a simple method to check for ground loops:
-make all connections to the PS bus and signal buss;
-connect PS bus to signal buss;
-do not connect any buss or RCA connector to the chassis;
-with a DMM, test the resistance between the buss and chassis, should be > > > 10 meg ohm;
-if OK, there are no ground loops within the amp; connect one point only to chassis and you are done.Thanks in advance for your help
Carlos
4. Lastly I will make a star for the input section and connect to a point on the buss that is furthest away form the
buss.~~~I meant to say "furthest away from the power supply end of the buss"~~~
5 The chassis gets only one point connected to earth ( a true shield passes no current)
Īs this earthing point the same as "1" above ?
~~~Yes usually it is.~~~
~~~As far as the Ohm meter test, it sounds good to me.~~~~~~One thing about grounding is that only the reservoir caps ground path is pretty much etched in stone. The other ground techniques are up to the skill and preferences of the designer. As soon as you think one methode works all the time than a circuit will come along and spit in your face. This is why you should just be ready to try different schems on each design.~~~~
Robert morin
Hi CarlosHow would you impliment the Sakuma scheme?? would there be two short busses, one for 'signal ground' and one for 'power supply ground' eath with whatever length returns were necessary (like a 'star' system)?
Also - with regard to hum/heater bakance pots - should they connect to the signal ground which I assume is connected to the chassis between the input jack area?
I've built long buss grounds and haven't been all that happy - think a star may be better but i've only used tht in an outboard supply for a phono preamp. I think perhpas a long buss invites 'stacked' loops - please help me understand.
Also - when using a one-buss system I think it should in theory tie one point to the chassis at the input but remember the late Manny Horowitz saying to tie at the main filter point - which if either is right?
best wishes
Freddy
ps - my main computer seems dead after being only 9 month old - guess I have to use my kid's computer and the yahoo address
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
