Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Let's twist again.
Let's twist again.
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 12th September 2017, 08:38 PM   #11
mchambin is offline mchambin  France
diyAudio Member
 
Join Date: Mar 2011
I know, there are people who connect the speaker ground at the PSU. I think it is wrong, it should be connected at the amp PCB as close as possible to the HF decoupling caps of the output transistors.
Can you point the case( s ), there is current in the ground wire. I think it can only occur from an unbalance between the plus and the minus circuits. ( kind of a three phase electric power )
__________________
Transistor junction temperature is not transistor case temperature.
  Reply With Quote
Old 12th September 2017, 10:27 PM   #12
mchambin is offline mchambin  France
diyAudio Member
 
Join Date: Mar 2011
Quote:
Originally Posted by Mark Whitney View Post
The current through the ground wire of the PSU is never zero.

When the PSU has a common ground for both power rails you should twist V+, V- and ground together.
I see when there is current through the ground wire: Mostly when the audio signal is bass, near and under 50Hz, little when the audio is highs well over 50 Hz.
Am afraid, I opened a can of worms.
__________________
Transistor junction temperature is not transistor case temperature.
  Reply With Quote
Old 13th September 2017, 12:39 PM   #13
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
There are two related issues here:
1. identify the current loop through the speaker, output stage and PSU and keep the loop area small by keeping the relevant wires close together and twisted
2. identify the correct voltage reference point for the amplifier input, output and feedback and ensure that any wire from here is kept well away from the above wires

Alternatively, just use a star ground and hope for the best.
  Reply With Quote
Old 13th September 2017, 02:13 PM   #14
mchambin is offline mchambin  France
diyAudio Member
 
Join Date: Mar 2011
Quote:
Originally Posted by DF96 View Post
There are two related issues here:
1. identify the current loop through the speaker, output stage and PSU and keep the loop area small by keeping the relevant wires close together and twisted
That is the question....How ?
Quote:
2. identify the correct voltage reference point for the amplifier input, output and feedback and ensure that any wire from here is kept well away from the above wires
Alternatively, just use a star ground and hope for the best.
This brings another question....How to implement signal ground, power ground and chassis ground ? if not, all confused at a star ground.
A small question inside: With an input via an RCA plug, is the RCA ground at chassis ground OR the RCA ground isolated from chassis and connected to signal ground ?
__________________
Transistor junction temperature is not transistor case temperature.
  Reply With Quote
Old 13th September 2017, 02:38 PM   #15
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
As I said "identify the current loop". This is part of circuit design. If you cannot do this then you cannot design the circuit. If you are using someone else's circuit then you have to hope that he has done this properly and said what it is.

There are lots of threads with lots of information on grounding. Some of it is helpful. No point in repeating it all here again. I only raised the issue because someone might be tempted to twist a ground reference wire with an output current loop wire and so induce interference into what is supposed to be a clean ground.
  Reply With Quote
Old 13th September 2017, 04:54 PM   #16
mchambin is offline mchambin  France
diyAudio Member
 
Join Date: Mar 2011
The current loops are identified, I did so well before my original post.
Grounding is rehashed all over the net, mostly smart handwaving,with contradictions that show up only when going fully in depth into it.
Unique Star ground.
Multiple: Signal ground Power ground Chassis ground.....and they end up connected together.
My OP is an attempt to find real answers.
So far, we are running in circles.

Let me begin a solution attempt.
My amp is a unity gain output stage with an op-amp as a front end and massive feedback over these.

The power ground is mostly where the two HF caps that decouple the power transistors Vs+ and Vs- meet, secondly where the two revervoir caps on the amp pcb meet next to it.
The loud speaker ground wire and the PSU ground wire go at this power ground.

The signal ground is mostly where the HF caps that decouple the Vs+ and Vs- of the op-amp, secondly where the two associeted electrolytic caps meet next to it.
The feed back network ( I use non inverting mode ) and the input signal ground wire go at this signal ground.

The power ground and the signal ground are linked by a unique link on the amp pcb.

As can be seen the output signal is not exactly the amplified input signal, it is skewed by a possible difference between signal ground and power ground.

Chassis is tied to the mains power ground prong.
Chassis is linked by a unique wire at the power ground wire ( PSU side ) at the PSU reservoir caps.
__________________
Transistor junction temperature is not transistor case temperature.
  Reply With Quote
Old 13th September 2017, 06:53 PM   #17
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
Quote:
The power ground and the signal ground are linked by a unique link on the amp pcb.

As can be seen the output signal is not exactly the amplified input signal, it is skewed by a possible difference between signal ground and power ground.
If that link carries no current and has no magnetic fields coupling to it then you are OK.
  Reply With Quote
Old 13th September 2017, 08:13 PM   #18
mchambin is offline mchambin  France
diyAudio Member
 
Join Date: Mar 2011
I am investigating on the current thru this link and return loops. I do not see clearly, yet, I think I need to reorganize my LTspice simulation to clearly split the part referenced at the power ground and the part referenced at the signal ground; Hopefully, I will identify and trace currents between these two parts.
__________________
Transistor junction temperature is not transistor case temperature.
  Reply With Quote
Old 13th September 2017, 08:27 PM   #19
Mark Whitney is offline Mark Whitney  Netherlands
diyAudio Member
 
Mark Whitney's Avatar
 
Join Date: Feb 2013
Location: Netherlands
If you connect the feedback and input signal grounds together and then connect this to the power ground you have the feedback current running along this link.

Plus any ground loop current when connecting multiple channels.
__________________
Regards Mark.
  Reply With Quote
Old 18th September 2017, 09:57 AM   #20
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by mchambin View Post
The current in the ground wire is zero in the case of a perfectly symmetrical PS and Class B amp.
So #2 should be best, but I am not sure....
No.
Look at your ouput stage and think only of an LF bass note that transistions slowly into one half wave and then crosses over into the other half wave.

Now thinking about that, look at the current flowing from the +ve supply into the output stage, to the speaker and finally returning to the PSU.

Now consider the other half wave of that LF tone.
Consider the current flowing from (to) the -ve supply into (out of) the output stage, to (from) the speaker and finally returning to the PSU.

In BOTH of the above cases the returning current passes along the Zero Volts line back to the PSU. The current MUST return to it's Source. The source of the speaker current is the PSU capacitor/s

Draw a simplyfied PSU + amp + spkr Load (use an opamp symbol for the amp)
Draw in coloured pencil the ROUTE the current must take to get from the PSU capacitor to the speaker and use another colour for the Return current.

Note that the current for each halfwave uses only two wires of the twisted triplet to pass energy to the speaker.
Note also that the Zero Volts wire carries current for the WHOLE waveform.
Finally note that each pair of current carrying wires need to be close coupled to minimise EMI.

Sketch to be attached when the camera battery has charged.
Attached Images
File Type: jpg psu currents.jpg (525.8 KB, 76 views)
__________________
regards Andrew T.

Last edited by AndrewT; 18th September 2017 at 10:10 AM.
  Reply With Quote

Reply


Let's twist again.Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
What Wires to Twist? WntrMute2 Construction Tips 12 18th January 2017 10:16 PM
Twist B+ and B- wires ? RhythMick Construction Tips 2 30th November 2015 10:04 PM
Heater circuit - to twist or not to twist Rob11966 Tubes / Valves 19 20th July 2008 02:08 AM
which wires to twist? Matttcattt Chip Amps 30 1st July 2003 10:16 AM


New To Site? Need Help?

All times are GMT. The time now is 07:15 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki