<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
	<title type="html"><![CDATA[ChipMusic.org - How to stabilize the parasitic oscillations of a 386 amplifier]]></title>
	<link rel="self" href="https://chipmusic.org:80/forums/feed/atom/topic/2321/"/>
	<updated>2010-09-07T18:54:47Z</updated>
	<generator>PunBB</generator>
	<id>https://chipmusic.org/forums/topic/2321/how-to-stabilize-the-parasitic-oscillations-of-a-386-amplifier/</id>
		<entry>
			<title type="html"><![CDATA[How to stabilize the parasitic oscillations of a 386 amplifier]]></title>
			<link rel="alternate" href="https://chipmusic.org/forums/post/38210/#p38210"/>
			<content type="html"><![CDATA[<p>If you&#039;re building an oscillator that runs through a speaker, chances are, you&#039;re using a 386 to amplify it. Why not? They&#039;re cheap, general, and easy to use.</p><p>Well, the fact is, they suck. For the most part, they usually work fine, but there&#039;s the occasion where you get unwanted oscillation when you want an amplifier. BUT, there are a solutions for when this happens.</p><p>We design circuits with ideal behavior in mind, but build them in a much less tidy, but ultimately more beautiful world, and old adage in electronics reminds us that</p><p>&quot;oscillators amplify and amplifiers oscillate&quot;.</p><p>Assuming you use the common gain of 200 from the LM386 datasheet:</p><p><a class="postimg" href="http://i53.tinypic.com/66g4eh.png" title="http://i53.tinypic.com/66g4eh.png" id="forum_image_80210062"><img src="http://i53.tinypic.com/66g4eh.png" /></a></p><p>Reducing the overall gain is an option. This is accomplished by removing the capacitor between pins 1 and 8. This drops the gain down from 200 to 20.</p><p><a class="postimg" href="http://i53.tinypic.com/egr9qe.png" title="http://i53.tinypic.com/egr9qe.png" id="forum_image_33076675"><img src="http://i53.tinypic.com/egr9qe.png" /></a></p><p>Be advised that this will also take away some of the pleasant harmonic distortion possible with higher gain.</p><p>Because of this sacrifice, I found a solution that is much better.</p><p>In my personal device in which I had this problem, the oscillations occur only when driving a speaker, not when using the output directly at the line output. The fix is partially to stabilize the &quot;load&quot; of the speaker. This is done by adding a resistor and capacitor to ground at pin 5, also known as the &quot;zobel network.&quot;</p><p>Another possible solution if is maybe even adding another larger capacitor to ground at pin 7.</p><p>These two solutions are illustrated here:</p><p><a class="postimg" href="http://i51.tinypic.com/2a0e3p1.png" title="http://i51.tinypic.com/2a0e3p1.png" id="forum_image_97596166"><img src="http://i51.tinypic.com/2a0e3p1.png" /></a></p><p>I hope this helps some people out.</p>]]></content>
			<author>
				<name><![CDATA[hotmessization]]></name>
				<uri>https://chipmusic.org/hotmessization</uri>
			</author>
			<updated>2010-09-07T18:54:47Z</updated>
			<id>https://chipmusic.org/forums/post/38210/#p38210</id>
		</entry>
</feed>
