zobel question

[sakis]

we had extensive talks and posts about zobel networks ....there is one question that might sound silly but you never know

does it make any diference in a zobel network if the arengement is inductor and then capacitor and resistor to ground .....

will it make an change if its inductor resistor and capacitor to ground ?????

does it make any diference which of the parts of the zobel is first attached to ground either the resistor or the capacitor ????

can it work visa versa ????? i think yes but on the other hand there might be reasons i dont know

thanks

[lineup]

Does not make much difference,
if the output inductor comes before the Zobel or after.

And if the cap or resistor is next to ground does not matter, at all.
This is for sure!
-------

I have seen both ways to put inductor in many amplifiers.
Can be interesting, though, to see what a Professor in Electronics use.
As we all know,
a real Professor is always 10 times more right
than any www.diyaudio.com besserwisser

You do not become Professor in Electronics at
Georgia Institute of Technology
School of Electrical and Computer Engineering
only because some people are nice .... I think


http://users.ece.gatech.edu/~mleach/lowtim/output.html
http://users.ece.gatech.edu/~mleach/lowtim/

[lineup]

This is Professor Leach comment to the bove output stage
of his amplifier.
One pratical thing with R50 C25 after the inductor
is that they can be put across the amplifier output terminals (Speaker OUT).
Leach have had bad experiences with putting these 2 components, The Zobel, on the PCB.


Zobel is an equalization filter for flatten the impedance curve for the LOAD.
This makes sense to me.
To put it more close the LOAD, than close to the amplifier feedback point.

R49, L1, R50, and C25 suppress parasitic oscillations that could be induced by shunt capacitance in the loudspeaker load.

R50 and C25 mount on the loudspeaker output binding posts.
On the first amplifier I built, I had these on the circuit board
where they caused violent oscillations because the current through R50 and C25 generated positive feedback when it flowed through the circuit board signal ground lead.
When the oscillations occurred, the heat sinks would get very hot.

Moving R50 and C25 to the loudspeaker output binding posts solved this problem.

I have seen amplifier circuit diagrams with different variations of the circuit formed by R49, L1, R50, and C25. Some have R50 and C25 on the circuit side of R49 and L1. Some have a series resistor and capacitor to ground on both sides of R49 and L1. One amplifier I saw had a big inductor for L1 that was wound on a toroidal core and there was no resistor in series with C25. In the Low TIM amplifier, L1 is a very small inductor. It measures about 1 uH and looks like a short circuit for audio frequencies.

[ferencz]

Lineup, the question was related if C or R should go to the ground!
My answer is that it makes no difference.

Anyway I've to clash with Prof. Leach, 'cause the L//R must be after R-C.

P.S.: Sakis, would you please ask more collected?

[richie00boy]

One might question why Leach was allowing dirty Zobel currents to flow in signal leads.

Self has a completely opposite recommendation and one that I agree with, that the Zobel is to provide load and low impedance to pickup, at RF, to the output stage. Therefore it should be on the PCB as any inductance due to wires or traces will spoil it's effectiveness.

You should have dirty ground for the Zobel and decoupling capacitors and a clean ground for the signal and feedback network. These two should return to the power ground star by their own wire.

To answer Sakis, in a series connection it does not matter in which order components go. In accordance with the recommendation by Self, the Zobel should go before the output inductor, as the output inductor is to just protect against highly capacitive loads from crossovers and speaker cables.

[PMA]

I would ground the capacitor, when I think about parasitic impedances.

[sakis]

diy audio ..... even a silly question can bring so many opinions from so many people

thank you all very much

my regards to PMA and LINE UP ..... my favourites !!!!

[ostripper]

You are very correct, Lineup , the Zobel/ inductor order is NOT
critical

Being a perfectionist, one has to consider that the zobel's true
function is to protect the amps output from too inductive
a reactance (loudspeaker). Knowing this, one could speculate
the best place for it is RIGHT at the output.

The output inductor's job is block the effect of capacitance
FROM the loudspeaker being introduced back to the amp.

Both effect's as mentioned above will affect the NFB loop and
cause various instabilities if left unchecked.

Another aspect of the zobel I found interesting is
where to "dump" the HF "garbage" that it filters out.
Most DIY designs I've seen couple the zobel right to the PC board
ground injecting this undesirable noise directly to the amp.

In my designs I keep the zobel network ground separate by connecting it directly to the main chassis star ground which
results in a much "cleaner" board/amp ground.


In commercial amps I've seen all 3 implementations .
(including my separated zobel scheme.)

Like in the prof. leach design, the extra damping resistor
in parallel with the output inductor reduces resonance
(ringing) at the output.

OS

[lineup]

Any bypass capacitor and filter cap attached to ground
will inject Current spikes into GND.
Where is Smaller caps, often smaller current.
Big caps often big Currents.
AC signal currents can not just dissapear.
And ground rails are often transporting this to power supply.

Ground rails have some resistance, inductance.
No matter what geometry / size of copper.
Copper has got resistance per volume & length.

Current x [Resistance+Inductance] = Voltage.
What we have done is SCREWING UP the 0-Volt ZERO Reference.
And if 0 is +0.1
then follows that +2 is no longer 2
but +2.1.
All other values are dependent on ZERO Reference.

And if REF is changed, all other values are changed.


Make this experiment.

Connect your 1000uF caps, that is to bypass/filter/feed output transistors with current.
Connect the Ground plane legs of those
to same GND rail as your INPUT impedance resistor.
Usually 22 kohm - 47 kohm.

And listen.

When working with Op-Amps, this can sometimes be an issue.
We use often two 100 nF very close to chip V+ V- pins.
Where we put the 'ground legs' of those 2 caps
can sometimes make a BIG difference.
This is why sometimes can be better to put only 1 capacitor.
From Opamp V+ pin to V- pin.
This way we avoid current injection at high frequency
into the GND.

My Lineup Bottomline:
.. There is nothing more important in one AC Amplifier
than to have a clean ZERO AC 0-volt Reference.
Star Grounding is one good way, to try to achieve this.

Note .. that ZERO DC Ref .. is not necessarily same as ZERO AC Ref.
Many times DC Ref is not abolute zero.
But AC Ref should always be as close to 0.000 Volt, as possible.

[traderbam]

Originally posted by lineup
My Lineup Bottomline:
.. There is nothing more important in one AC Amplifier
than to have a clean ZERO AC 0-volt Reference.
Star Grounding is one good way, to try to achieve this.

Note .. that ZERO DC Ref .. is not necessarily same as ZERO AC Ref.
Many times DC Ref is not abolute zero.
But AC Ref should always be as close to 0.000 Volt, as possible. [/B]

You may want to explain this a little. Of course, there is no such thing as absolute zero except, for example, when it comes to temperature.