Boucherot cell (Zobel network) values - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Chip Amps

Chip Amps Amplifiers based on integrated circuits

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 21st February 2012, 01:41 PM   #11
diyAudio Member
 
Join Date: Jun 2011
Quote:
Originally Posted by KSTR View Post
The cap isolates DC/LF current and is normally choosen to kick in just above the audio range (but still it must be capacitive at HF, which is actually more important), while the resistor is the part that "does the work", and it must be a low inductance type and also wired with low inductance, btw preferably directly to the neg. supply (which must be bypassed also directly to the pos. supply, or alternatively split up the snubber cap in a center tapped cap between the supplies).

- Klaus
My apologies for reviving this old thread.

I am rather curious about the pros and cons of returning zobel load to the negative supply. Most give a standard config by retuning zobel to ground, or rather the center tap, but could it be more beneficial to return the load to the negative rail?
  Reply With Quote
Old 21st February 2012, 08:13 PM   #12
diyAudio Member
 
Join Date: May 2008
A Boucherot cell is used to terminate a transmission line with a defined impedance to reduce or avoid reflections that interfere with the original signal. That is obviously not the purpose of this RC combination.

A Zobel network is used to linearise an impedance. The only impedance you could linearise at the amp output would be the speaker, but obviously you would need to know which speaker gets connected to find the right values, the values of that RC combination are too small for any speaker, and a well designed impedance linearisation would consist of an additional inductor in series, too. So that is apparently not the purpose of the RC combination either.

That RC combination is simply an RF snubber. The capacitor shorts RF to ground, and the resistor is the current limiter that makes sure the amp does not work into a clean short at high frequencies, where the capacitor's impedance is nearly zero. That is why the National datasheet calls them neither Cbo and Rbo nor Czo and Rzo, but Csn and Rsn.

A more typical value for that resistor is 10 Ohm. The reason to use a smaller value is to make the capacitor more effective at shunting RF to ground at the cost of increasing the frequency where the snubber becomes effective. Drawback is that an amp reproducing corespondingly high frequencies could be driven into its current limit. E.g. an amp that oscillates in the relevant frequency range would certainly suffer even more with 2,7 Ohm than with the usual 10 Ohm. So if the sound of your amp deteriorates due to that resistor, that is an indication of oscillations.

The 100 nF capacitor has an impedance of ~80 Ohms at 20 kHz and ~80 kOhms at 20 Hz. What difference can it make, whether the series resistor to that is 2,7 or 5 or 10 Ohms when it comes to sound quality?
__________________
If you've always done it like that, then it's probably wrong. (Henry Ford)

Last edited by pacificblue; 21st February 2012 at 08:19 PM.
  Reply With Quote
Old 21st February 2012, 08:18 PM   #13
diyAudio Member
 
Join Date: May 2008
Quote:
Originally Posted by nereis View Post
Most give a standard config by retuning zobel to ground, or rather the center tap, but could it be more beneficial to return the load to the negative rail?
Only with single supply, where the negative rail is usually at the same time the ground.
__________________
If you've always done it like that, then it's probably wrong. (Henry Ford)
  Reply With Quote
Old 21st February 2012, 09:00 PM   #14
diyAudio Member
 
noddy55's Avatar
 
Join Date: Feb 2012
Quote:
That RC combination is simply an RF snubber. The capacitor shorts RF to
ground, and the resistor is the current limiter that makes sure the amp
does not work into a clean short at high frequencies, where the
capacitor's impedance is nearly zero. That is why the National
datasheet calls them neither Cbo and Rbo nor Czo and Rzo, but Csn and
Rsn.
A more typical value for that resistor is 10 Ohm. The reason to use a
smaller value is to make the capacitor more effective at shunting RF to
ground at the cost of increasing the frequency where the snubber
becomes effective. Drawback is that an amp reproducing corespondingly
high frequencies could be driven into its current limit. E.g. an amp
that oscillates in the relevant frequency range would certainly suffer
even more with 2,7 Ohm than with the usual 10 Ohm. So if the sound of
your amp deteriorates due to that resistor, that is an indication of
oscillations.
The 100 nF capacitor has an impedance of ~80 Ohms at 20 kHz and ~80
kOhms at 20 Hz. What difference can it make, whether the series
resistor to that is 2,7 or 5 or 10 Ohms when it comes to sound quality?
Welldone! Well Explained! Superb! Marvelous! Thank you very much! Years confusion has gone in minutes, forever. Thanks a lot.
  Reply With Quote
Old 21st February 2012, 09:06 PM   #15
KSTR is offline KSTR  Germany
diyAudio Member
 
KSTR's Avatar
 
Join Date: Jul 2007
Location: Central Berlin, Germany
If you want to partially shunt HF *away* from the chip then those RC dampers should return to GND. Better yet, don't allow HF to get in by symmetrically filtering it at the output terminals.

If you want to control and stabilize the open-loop gain at HF by moderately loading it down then it should return to GND.

Finally, if you want to damp an output section that is prone to parasitic oscillations on its own then the damper should be located close and should return to the appropriate supply pin (neg rail, typically). Taking it to GND is a tolarable detour.

Quite often, we will want all three effects at the same time.

Since GND and the rails must all be shorted together for HF right at the chip it doesn't matter much which route you take.

Last edited by KSTR; 21st February 2012 at 09:11 PM.
  Reply With Quote
Old 21st February 2012, 10:47 PM   #16
diyAudio Member
 
Join Date: Jun 2011
Thank you pacificblue and KSTR, I am putting this thread in my bookmarks!
  Reply With Quote
Old 21st February 2012, 10:56 PM   #17
johnr66 is offline johnr66  United States
diyAudio Member
 
Join Date: Mar 2009
I still wonder how the newer technology audio chips don't require the RC network on the output. They are usually the lower power chips, up to and a bit beyond car BTL chips.
  Reply With Quote
Old 20th September 2012, 07:12 AM   #18
diyAudio Member
 
availlyrics's Avatar
 
Join Date: May 2011
Quote:
Originally Posted by pacificblue View Post
The 100 nF capacitor has an impedance of ~80 Ohms at 20 kHz and ~80 kOhms at 20 Hz. What difference can it make, whether the series resistor to that is 2,7 or 5 or 10 Ohms when it comes to sound quality?
Thanks, that cleared my doubts!
  Reply With Quote
Old 20th September 2012, 01:12 PM   #19
diyAudio Member
 
Join Date: Feb 2011
i might be mistaking here, but if one would be to liearise a speaker's impedance, then it would be done just like if it was part of a crossover.
But the RC zobel can only deal with the rising impedance character of the voicecoil.
But, would not deal with the peak at FS.
Even so, if one adds a zobel to deal with the rising impedance due to voiceciol inductance,
and add a nother compensation network to get rid of the impedance peak, it would still only be able to compenase for a single driver.
Supposedly an active crossover setup (1 driver, 1 amplifier) could benefit from this, as the amplifier would see a close to resistive load. In otherwords it would be easy to drive it, as current spikes would be reduced.
But i think for passive crossover systems this would be verry component extensive to do, and probably would end up pretty bad will all that stuff connected.
Allso i would question if it would lead to any audible effect.
Prehaps a realy general zobel is just to more or less deal with rising impedance characteristics at HF side, and in that case having verry accurate component values are not mission critical.
BTW, one would need an impedance plot of the actual speaker used, be it a single speaker, or one with multiple drivers and a crossover.. witch would add even more complexity to begin with.

Passive Crossover Network Design
This excellent artice shows how to design the compensation networks for a speaker, to make it seem like a resistive load for a passive crossover. Same can be applyd to make the driver seem like a resistive load for a poweramplifier.
  Reply With Quote
Old 20th September 2012, 05:24 PM   #20
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Each Speaker Zobel you add, reduces the speaker efficiency.
Because the added parallel impedance draws more current from the amplifier and this added current draw bypasses the speaker.
The other is that within the passband of each speaker, the efficiency of the driver matches the impedance of the voil coil/magnet. The driver outputs a nearly constant SPL for all supplied voltage when fed from near zero source impedance.

Put a high source impedance amplifier (or an added resistor) in as the feed and the high impedance regions of the driver frequency range and you find big "holes" in the frequency response. You don't want that. Adding a Zobel does a similar thing, if there is any source impedance.
__________________
regards Andrew T.
  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
Boucherot Cell ( Zobel) not required for the TDA7294 ? ashok Chip Amps 4 17th September 2014 09:35 AM
No Boucherot cell? johnr66 Chip Amps 2 5th March 2009 10:45 PM
Zobel Network JiggaD369 Multi-Way 9 5th May 2005 01:27 AM
Zobel Network nuttinbutSQ Multi-Way 6 16th March 2004 09:19 AM


New To Site? Need Help?

All times are GMT. The time now is 08:39 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2