Large capacitors decoupling. - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 14th August 2013, 01:23 PM   #1
diyAudio Member
 
Join Date: Aug 2011
Default Large capacitors decoupling.

Hi all,
I doubting at this moment about designed PSU for output stage amp.
I planned to place 10x10,000 uf capacitors per rail into the same output transistors board, so distance from collectors pins to +- rail might be about 10 cm via pcb track. How should I do for decoupling? please advise what option should be good in term of sounding.

1. 10x10,000uf+10x0.1uf MKS4
2. 10x10,000uf+2x0.1uf MKS4
3. 10x10,000uf+2x10uf MKP
4. 10x10,000uf+2x0.1uf MKS4+2x10uf MKP

Are there the negative things of having multiple small value decoupling caps per point?.

Is lowest overall ESR good for PSU?
__________________
My DIY audio site.
  Reply With Quote
Old 14th August 2013, 01:44 PM   #2
diyAudio Member
 
east electronics's Avatar
 
Join Date: Jan 2006
Location: Athens GREECE
10x10.000 per rail ...or could it be per amplifier ?
you need to be more specific about that ...

one way or another this looks like an overkill or at least over design ...it could be nicer to know the all project you are working on .

Implementation of over design and pcb rooting and how after all is possible to feed and distribute correctly all that power that comes form such a huge bank to a circuit while preserving a star ground and/or as minimum /possible ground all around splatter is possible is you concern here ...not decoupling and bypass as is better to be called .

You need to also study benefits and problems that might come from both before you decide about that..

To cut the story shorter i would like to see a pcb for that No details just a general view

Kind regards
Sakis
__________________
SERVICE ΕΝΙΣΧΥΤΩΝ ΚΑΙ ΙΑΠΩΝΙΚΩΝ ΜΗΧΑΝΗΜΑΤΩΝ ΗΧΟΥ www.eastelectronics.gr
  Reply With Quote
Old 14th August 2013, 03:40 PM   #3
diyAudio Member
 
Join Date: Aug 2011
Quote:
Originally Posted by sakis View Post
10x10.000 per rail ...or could it be per amplifier ?
you need to be more specific about that ...

one way or another this looks like an overkill or at least over design ...it could be nicer to know the all project you are working on .

Implementation of over design and pcb rooting and how after all is possible to feed and distribute correctly all that power that comes form such a huge bank to a circuit while preserving a star ground and/or as minimum /possible ground all around splatter is possible is you concern here ...not decoupling and bypass as is better to be called .

You need to also study benefits and problems that might come from both before you decide about that..

To cut the story shorter i would like to see a pcb for that No details just a general view

Kind regards
Sakis
Thanks Sakis,
I can actually share what I'm doing.
I'm working on circuitry at this moment so the actual pcb is not designed yet.
It is 3xx watt at 8 ohm and about 600 watt at 4 ohm amp works as balanced, so the PSU is lower voltage but will consumed more current in order to reach that aimed output power.
100,000 uf per channel(for inverted and non-inverted amps). Because of I don't want to have just 2 very large caps per channel, due to the physical issue.

You can share what are you thinking about this mine.
__________________
My DIY audio site.
  Reply With Quote
Old 14th August 2013, 03:46 PM   #4
diyAudio Member
 
east electronics's Avatar
 
Join Date: Jan 2006
Location: Athens GREECE
As said i am not interested in your circuit ...Implementation though will be fun to watch

Kind regards
Sakis
__________________
SERVICE ΕΝΙΣΧΥΤΩΝ ΚΑΙ ΙΑΠΩΝΙΚΩΝ ΜΗΧΑΝΗΜΑΤΩΝ ΗΧΟΥ www.eastelectronics.gr
  Reply With Quote
Old 15th August 2013, 03:03 AM   #5
diyAudio Member
 
Join Date: Aug 2011
Ok let do it this way to my simple question.

Is one small cap per rail or one per each big cap in PSU?
__________________
My DIY audio site.
  Reply With Quote
Old 15th August 2013, 03:13 AM   #6
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Probably you already know that a large amount of capacitance will require careful design of the parts upstream or you'll end up with over-stressed rectifiers and transformers.

I believe you want to worry about inductance of the wiring. Large caps take up space and lengthen traces. Small caps near the power devices offer short traces. So a small cap near the power device is a good idea; the larger caps cab then be further away.

Big caps have low ESR, so on main PSU I prefer to use small number of big caps instead of large number of small caps because it's easier.
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.

Last edited by Bigun; 15th August 2013 at 03:15 AM.
  Reply With Quote
Old 15th August 2013, 03:46 AM   #7
diyAudio Member
 
Join Date: Aug 2011
Nice to know guys, it is not easy to design the pcb tracks, but could somebody answer my simple question?
__________________
My DIY audio site.
  Reply With Quote
Old 16th August 2013, 08:24 AM   #8
diyAudio Member
 
east electronics's Avatar
 
Join Date: Jan 2006
Location: Athens GREECE
you need to separate your question between bypass and decoupling ...

All capacitors will benefit from a bypass and all amplifiers from a decoupling while bypass is an upgrade thing to do and it is expected to enhance performance and decoupling is necessary for proper operation of the amplifier .

All benefits from both may seriously degrade and often create more trouble than good if wiring /traces/ pcb / implementation is not proper

Does this answer your question ?
__________________
SERVICE ΕΝΙΣΧΥΤΩΝ ΚΑΙ ΙΑΠΩΝΙΚΩΝ ΜΗΧΑΝΗΜΑΤΩΝ ΗΧΟΥ www.eastelectronics.gr
  Reply With Quote
Old 16th August 2013, 12:58 PM   #9
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
In an amplifier the OUTPUT device connects the LOAD to the PSU.
It is PSU current that gets modulated to drive the load.

That OUTPUT device is what modulates the current.
ANY INDUCTANCE between the PSU and the OUTPUT device will interfere with the way the current gets delivered to the LOAD.

Decoupling is the best way to help the OUTPUT device deliver CURRENT to the LOAD.

The analogy that has been used by a few Members goes as follows:
Imagine that your PSU is disconnected. The whole CURRENT to the LOAD comes from the decoupling. It behaves as a very short term battery. Just microseconds, s or us.

Look at the loop that takes current from the decoupling through the output device to the output Zobel and now back through the return paths to the decoupling. That loop will have inductance. The BEST way to minimise the inductance of that loop is to minimise the loop area. That means short lengths and close coupling.

I did not include the Speaker in that load loop ! Because the UHF does not go to the speaker. The UHF part of the signal goes around the Zobel R+C.

The bigger capacitors (100uF to 1mF), the MF decoupling on the PCB, recharge the tiny ceramics (22nF to 220nF) that gave the s supply, ready for the next transient. The PSU recharges the MF decoupling. This recharging of the MF decoupling does not need to be faster than a few ms. A bit of inductance in that recharge loop is not a performance problem.
To my mind the PSU does not need decoupling, nor bypassing. For me this is a big change in my attitude from what I believed 40years ago.
__________________
regards Andrew T.
  Reply With Quote
Old 17th August 2013, 03:18 AM   #10
diyAudio Member
 
Join Date: Aug 2011
Quote:
Originally Posted by AndrewT View Post
In an amplifier the OUTPUT device connects the LOAD to the PSU.
It is PSU current that gets modulated to drive the load.

That OUTPUT device is what modulates the current.
ANY INDUCTANCE between the PSU and the OUTPUT device will interfere with the way the current gets delivered to the LOAD.

Decoupling is the best way to help the OUTPUT device deliver CURRENT to the LOAD.

The analogy that has been used by a few Members goes as follows:
Imagine that your PSU is disconnected. The whole CURRENT to the LOAD comes from the decoupling. It behaves as a very short term battery. Just microseconds, s or us.

Look at the loop that takes current from the decoupling through the output device to the output Zobel and now back through the return paths to the decoupling. That loop will have inductance. The BEST way to minimise the inductance of that loop is to minimise the loop area. That means short lengths and close coupling.

I did not include the Speaker in that load loop ! Because the UHF does not go to the speaker. The UHF part of the signal goes around the Zobel R+C.

The bigger capacitors (100uF to 1mF), the MF decoupling on the PCB, recharge the tiny ceramics (22nF to 220nF) that gave the s supply, ready for the next transient. The PSU recharges the MF decoupling. This recharging of the MF decoupling does not need to be faster than a few ms. A bit of inductance in that recharge loop is not a performance problem.
To my mind the PSU does not need decoupling, nor bypassing. For me this is a big change in my attitude from what I believed 40years ago.
That was interesting AndrewT I liked you short article.
But what about PSU impedance at mid and hi frequency. I have once tried with add/remove decoupling/bypass at VAS regulator-output. The sounding really different, with decoupling it's better treble. I never try the same thing with hi-power unregulated PSU.
__________________
My DIY audio site.
  Reply With Quote

Reply


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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Bipolar capacitors for decoupling? rhythmsandy Solid State 2 30th December 2012 09:54 PM
TDA1541A decoupling capacitors value Zoran Digital Source 25 22nd April 2010 10:18 PM
Film capacitors for decoupling rotation Tubes / Valves 4 6th November 2007 09:02 PM
Gainclone - why use large value decoupling capacitors ? ash_dac Chip Amps 11 20th August 2006 03:49 PM
Good Coupling and Decoupling Capacitors fmak Solid State 42 20th June 2004 09:05 AM


New To Site? Need Help?

All times are GMT. The time now is 10:34 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright 1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2