Searching \ for '[OT] Noise in discrete components' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=noise+discrete+components
Search entire site for: 'Noise in discrete components'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Noise in discrete components'
2000\03\08@215740 by JB

flavicon
face
I'm working on a 16f877 a/d project that reads audio from an electret(sp?) microphone which has been amplified by an lm386 set for a gain of 100. The mic itself seems to be low on noise, but the 10k resistor to V+ is a standard carbon 1/4W variety and it seems to be producing excessive noise. I conclude this based on the fact that I can short the mic and the noise persists, but disconnecting the resistors will stop the noise. Test circuit is below (use fixed-pitch font to view). Would a metal-oxide resistor help, or are there other options?

Lots of noise:

__________>> +9V
|
\
/
\ 10k
/
|
|-| |--->> amp (lm386 gain ~100)
|
[ ] mic
|
\
/
\ 4.7k
/
|
|________________>> GND

Lots of noise:

__________>> +9V
|
\
/
\ 10k
/
|
|-| |-->> amp (lm386 gain ~100)
|
|
|
\
/
\ 4.7k
/
|
|________________>> GND

Minimal noise:

|-| |--->> amp (lm386 gain ~100)
|
|
|
|
|________________>> GND

2000\03\08@224627 by Peter van Hoof

flavicon
face
Try the cap from the input of the amplifier to the 9volt line
I expect , lots of noise probably because of power supply noise

Peter

[snip]

Noise in discrete components

I'm working on a 16f877 a/d project that reads audio from an electret(sp?)
microphone which has been amplified by an lm386 set for a gain of 100. The
mic itself seems to be low on noise, but the 10k resistor to V+ is a
standard carbon 1/4W variety and it seems to be producing excessive noise. I
conclude this based on the fact that I can short the mic and the noise
persists, but disconnecting the resistors will stop the noise. Test circuit
is below (use fixed-pitch font to view). Would a metal-oxide resistor help,
or are there other options?

[snip]

2000\03\09@003702 by Robert A. LaBudde

flavicon
face
<x-flowed>At 07:57 PM 3/8/00 -0700, you wrote:
>I'm working on a 16f877 a/d project that reads audio from an electret(sp?)
>microphone which has been amplified by an lm386 set for a gain of 100. The
>mic itself seems to be low on noise, but the 10k resistor to V+ is a
>standard carbon 1/4W variety and it seems to be producing excessive noise.
>I conclude this based on the fact that I can short the mic and the noise
>persists, but disconnecting the resistors will stop the noise. Test
>circuit is below (use fixed-pitch font to view). Would a metal-oxide
>resistor help, or are there other options?

1. Yes. Carbon composition resistors are noise generators. The noise is
proportional to the current. Use a 10k resistor to bias the FET. Carbon
film is better than composition, metal film is best.

2. You should use a low noise pre-amp. Either a low noise op-amp or a low
noise transistor. Why are you trying to do all the gain in the LM386? If
it's a minimal parts count, you're going to have to live with the amplified
noise of the LM386.

3. Bypass all power supplies (+/-) to ground with 0.1 uF capacitors.

================================================================
Robert A. LaBudde, PhD, PAS, Dpl. ACAFS  e-mail: spam_OUTralTakeThisOuTspamlcfltd.com
Least Cost Formulations, Ltd.                   URL: http://lcfltd.com/
824 Timberlake Drive                            Tel: 757-467-0954
Virginia Beach, VA 23464-3239                   Fax: 757-467-2947

"Vere scire est per causae scire"
================================================================

</x-flowed>

2000\03\09@005437 by David VanHorn

flavicon
face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


>2. You should use a low noise pre-amp. Either a low noise op-amp or a low
>noise transistor. Why are you trying to do all the gain in the LM386? If
>it's a minimal parts count, you're going to have to live with the amplified
>noise of the LM386.

The 386 is very much NOT a low noise part.

Try any generic FET input op-amp, you'll get much better performance.
You may want to filter the audio as well, those mics are amazingly
broadbanded.

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.2 for non-commercial use <http://www.pgp.com>

iQA/AwUBOMdmiIFlGDz1l6VWEQIs1ACgvvmmVbmWpvbkTubf/MiJ06FjuxcAoLFg
rJLWCqIqvmT/5s0cH1+IeD3Q
=o/GL
-----END PGP SIGNATURE-----

2000\03\09@063429 by Russell McMahon

picon face
As well as other good advice here, you should decouple the mic supply as
power supply noise will enter via the 10K to Vcc.
Split the 10 K into 2 x 4k7 and add an electrolytic (say about 10uF to
100uF) from the junction to ground.
Chances are you can in fact use 2 x 10k with OK results with most electrets.
This isolates the filter action from the low impedance of the supply
itself - you are then not trying to pass all the supply noise present at
this point.

Also, IF you are using a differential amplifier try taking an equal sized
capacitor from the BOTTOM of the mic to the "other" amplifier input.

The bottom 4K7 you show really doesn't do much UNLESS you do something like
the differential input mentioned above.



regards


     Russell McMahon
_____________________________

>From other worlds - http://www.easttimor.com
                               http://www.sudan.com

What can one man* do?
Help the hungry at no cost to yourself!
at  http://www.thehungersite.com/

(* - or woman, child or internet enabled intelligent entity :-))


{Original Message removed}

2000\03\09@084855 by jb

flavicon
face
I haven't had a chance to try it out yet, but you have pointed out several
areas I need to address. I assumed that being battery powered meant that no
bypass cap would be needed, but I guess there will always be something to
bypass due to internal battery resistance (especially with the PIC running
when I get to that point). The quiescent noise level of the lm386 seems
pretty decent for the purposes at hand, but if a FET amp would be better, I
should pursue that as well. I would like to have it in one package if
possible due to size constraints.

Thanks to everyone for the assistance with this.

J

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...