KX1 speaker output with superimposed oscillation

[wb2srf]

Hi Paco,

Sorry to read that it did not work for you.

LM386 amplifiers can do strange things when over driven. There is a lot
of available gain.

If the amplifier is operating properly, the Idle DC voltage with no audio
signal on the output pin 5 should be  half of the DC voltage measured on
the power supply pin 4 if not you may have a  leaky coupling capacitor on
the input circuit or a bad IC internal circuit throwing off the bias.

I have also found at times from building my own equipment that a de
coupling capacitor with short leads is needed across the power supply
pins 4 and 6  to keep it stable at high output levels especially if the
impedance of  the power supply is not low enough. I use a 10uF tantalum
capacitor  when connecting it is very important to observe the correct
polarity.

Also the RC network connected between the output pin 5 and ground is
required to keep the amplifier stable at high frequencies at different
loads.

GL 73,
Bob Johansen WB2SRF
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Francisco Barros]

Hi Bob,

Finally, I got it !.
It is related with "the RC network connected between the output pin 5 and
ground, required to keep the amplifier stable at high frequencies at
different loads", as you mention on your previous post.
I checked the RC network and I even installed a new one: same behavior.
If I removed this RC network completely, the oscillation now started with
lower volumes !.

Using the technique "try and error", the problem was solved with a capacitor
of 68 nF in parallel with this RC network, that is, just from pin 5 to
ground. No more high frequency oscillation with different loads !.

By the way, the DC voltage at pin 5 was half the supply: 3 V. I even
installed a short leads 10 uF tantalum across the supply of U4, without
positive results.

I don't know the reason of this behaviour, maybe, as you say on your post,
is due to the high gain of this stage (46 dB). Who knows... .

Many thanks Bob and have a good day.

73,
Paco

_________________________________________________________________
Descarga gratis la Barra de Herramientas de MSN
http://www.msn.es/usuario/busqueda/barra?XAPID=2031&DI=1055&SU=http%3A//www.hotmail.com&HL=LINKTAG1OPENINGTEXT_MSNBH

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[wb2srf]

 
Hi Paco,
 
Great detective work, glad to read that you tamed the beast. Hi Hi
 
I am in the process of trying to model the LM386 circuits  using LT spice
( program free from Linear Technologies web site SWCAD III) to try to
understand the feedback mechanism that results in these amplifier's
becoming unstable and turning into an oscillator.
 
Many times I have had success through trial and error getting circuits to
work properly but with the help of some engineer friends, I have been
encouraged to try circuit analysis. It is strange that placing a
capacitor at the output to ground solved your problem because it can
cause a phase shift to result acting with the output resistance internal
to the IC. Perhaps the phase shift placed the gain-phase margin outside
of the area in a Bode plot to stabilize your amplifier
 
These amplifiers have a problem. Before they run out of gain at high
frequencies, if there is 360 degree's of phase shift from input to output
of the amplifier it turns into an oscillator.
 
 
73,
Bob Johansen WB2SRF
 
 
On Sun, 09 Sep 2007 16:28:43 +0000 "Francisco Barros"
<[hidden email]> writes:

> Hi Bob,
>
> Finally, I got it !.
> It is related with "the RC network connected between the output pin
> 5 and
> ground, required to keep the amplifier stable at high frequencies at
>
> different loads", as you mention on your previous post.
> I checked the RC network and I even installed a new one: same
> behavior.
> If I removed this RC network completely, the oscillation now started
> with
> lower volumes !.
>
> Using the technique "try and error", the problem was solved with a
> capacitor
> of 68 nF in parallel with this RC network, that is, just from pin 5
> to
> ground. No more high frequency oscillation with different loads !.
>
> By the way, the DC voltage at pin 5 was half the supply: 3 V. I even
>
> installed a short leads 10 uF tantalum across the supply of U4,
> without
> positive results.
>
> I don't know the reason of this behaviour, maybe, as you say on your
> post,
> is due to the high gain of this stage (46 dB). Who knows... .
>
> Many thanks Bob and have a good day.
>
> 73,
> Paco
>
> _________________________________________________________________
> Descarga gratis la Barra de Herramientas de MSN
>

http://www.msn.es/usuario/busqueda/barra?XAPID=2031&DI=1055&SU=http%3A//w
ww.hotmail.com&HL=LINKTAG1OPENINGTEXT_MSNBH
>
>
>
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[dave-281]

Paco,

Thanks for posting the solution to this problem.   I feel sure there are
a number of us who read about it with interest but did not comment.  
Good to know the answer to the problem!!


73 de dave
ab9ca




Francisco Barros wrote:

> Hi Bob,
>
> Finally, I got it !.
> It is related with "the RC network connected between the output pin 5
> and ground, required to keep the amplifier stable at high frequencies
> at different loads", as you mention on your previous post.
> I checked the RC network and I even installed a new one: same behavior.
> If I removed this RC network completely, the oscillation now started
> with lower volumes !.
>
> Using the technique "try and error", the problem was solved with a
> capacitor of 68 nF in parallel with this RC network, that is, just
> from pin 5 to ground. No more high frequency oscillation with
> different loads !.
>
> By the way, the DC voltage at pin 5 was half the supply: 3 V. I even
> installed a short leads 10 uF tantalum across the supply of U4,
> without positive results.
>
> I don't know the reason of this behaviour, maybe, as you say on your
> post, is due to the high gain of this stage (46 dB). Who knows... .
>
> Many thanks Bob and have a good day.
>
> 73,
> Paco
>
> _________________________________________________________________
> Descarga gratis la Barra de Herramientas de MSN
> http://www.msn.es/usuario/busqueda/barra?XAPID=2031&DI=1055&SU=http%3A//www.hotmail.com&HL=LINKTAG1OPENINGTEXT_MSNBH 
>
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft   
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
>
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Francisco Barros]

Hi Dave,

Thanks to you.
IHope Bob can tell us, with the modeling of the LM386, why this strange
behaviour.
>From the knowledge point of view, the technique "try and error" is not the
best way, we would like to know more... .
Now, my KX1, at the LM386's speaker output, pin 5, has the original RC
network (10 Ohm + 100 nF) to ground plus one 68 nF to ground also.
I even tried to put the 68 nF in parallel with the original 100 nF but it
did not worked. Neither with a value of 100 nF, so finally I installed the
68 nF capacitor directly to the output and ground, allways with short leads
(I learned this after several years working with RF techniques :)).

73,
Paco

_________________________________________________________________
Grandes éxitos, superhéroes, imitaciones, cine y TV...
http://es.msn.kiwee.com/ Lo mejor para tu móvil.

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com