Positive n negative feedback

[zachtheterrible]

Hello, I'm wondering if someone can tell me what the difference is between positive and negative feedback. Thank u.

 

[Nigel Goodwin]

Hello, I'm wondering if someone can tell me what the difference is between positive and negative feedback. Thank u.

Positive feedback is 'in phase', negative feedback is 'out of phase'. So negative feedback decreases gain, positive feedback increases gain - it's usually used to make an oscillator, although small amounts are also used to make comparators 'snap' more cleanly, and to provide a degree of hysteresis for the comparator.

 

[TillEulenspiegel]

Also neg feedback is used in audio circuits ( especially my fave ..old tube amps ) to give less distortion and more controlled output. Basically it provides a remedy for undesired audio artifacts.

Think of it in these terms positive feedback is used in oscillators and all processes that need elements to "march in step", such as a radio transmitter, a laser ( altho the feedback is optical ).

Neg feedback is a method to control spurious signals and wave shaping.

Damn I gotta go Wife time ..........mothers day and all that

 

[zachtheterrible]

That makes sense. In this schematic, I am told that R2 provides negative feedback. How does it do that, seeing as it is not connected to the base of Q1? Also, can positive feedback be not connected directly to the base? If so, how is it done?

 

[nettron1000]

The emitter resistor (R2) provides a degree of feedback.

 

[zachtheterrible]

When u say that R2 provides a degree of feedback, are you saying that it doesn't provide all of the feedback? And I didn't really get my whole question answered. How does it provide negative feedback if it's not connected to the base of Q1?

 

[Optikon]

When u say that R2 provides a degree of feedback, are you saying that it doesn't provide all of the feedback? And I didn't really get my whole question answered. How does it provide negative feedback if it's not connected to the base of Q1?

Zach.. I understand your question. First, it is important to understand what it is that is being fedback. In the circuit, the current that is flowing through the R2 resistor comes from the emitter of the trans. If you consider the operating point of the transistor at a given current, this corresponds to a particular Vbe of the transistor.

Now, if you were to increase the current through R2, this would increase the voltage drop across it (ohms law) and the emitter voltage would rise as well with respect to ground. This in essence _decreases_ the Vbe drop of the trans. A Vbe that decreases, tends to _decrease_ collector (and emitter) current.

So to summarize, a current that wants to increase in R2, lowers the Vbe causing that same current to decrease. This is voltage feedback to the transistor and since the current tends to reduce, the gain of the transistor is effectively reduced. So we have voltage feedback that has reduced the gain of the trans => negative feedback.

If on the other hand an increasing current in the emitter of the trans caused Vbe to increase, this would be positive feedback and you have a runaway situation on your hands. So emitter resistors provide negative feedback.

This of course is a qualitative way of looking at it. The math will show you the same thing if you choose to undertake that exercise.

 

[zachtheterrible]

hmm, I think that i've got it, so i'm going to try and explain it to you in my own words, and tell me if I'm wrong (or right, hopefully). R2 basically lowers the current that can flow from base to emitter, and therefore lowers the gain of the amplifier, which makes the amplifier more stable.

When you say that positive feedback would cause a runnaway situation, what would it do, and how would positive feedback be attained?

Thank you so very much

 

[Styx]

hmm, I think that i've got it, so i'm going to try and explain it to you in my own words, and tell me if I'm wrong (or right, hopefully). R2 basically lowers the current that can flow from base to emitter, and therefore lowers the gain of the amplifier, which makes the amplifier more stable.

When you say that positive feedback would cause a runnaway situation, what would it do, and how would positive feedback be attained?

Thank you so very much

cant answer yr first bit bit inna way yes.

the second part is:
Positive feedback use to be used all the time when there was no amplification around (pre-valves). It continues to be used but as teh valves started having a bit more kick behind them it would only take a small disturbance that wasnt expect to cause the output to fly off.

Think of it like a mic and a speaker. If you put the mic near the speaker then a high-pitch squeel will get loader and loader. A bit of the input is being added to the output. The output then goes back into the input with that bit extra, this the comes out with a 2x bit extra.... soon maximum output.

Negative feedback came abt in an atempt to provide an ampification setup's who's responce could be calculated by hand, not tuned each time. It had the major benefit of being stable with no runnaway.

As to getting positive feedback, you just need to take a scaled down (or not depending on what you want to do) of the output and sum it in at the input

 

[zachtheterrible]

Ohhhhhh.

 

[Optikon]

hmm, I think that i've got it, so i'm going to try and explain it to you in my own words, and tell me if I'm wrong (or right, hopefully). R2 basically lowers the current that can flow from base to emitter, and therefore lowers the gain of the amplifier, which makes the amplifier more stable.

When you say that positive feedback would cause a runnaway situation, what would it do, and how would positive feedback be attained?

Thank you so very much

I think you've got it. Given an increasing emitter current, the presence of R2 (in the emitter) will cause Vbe to decrease, thus decreasing the emitter current. This keeps things under control.

If the circuit were such that an increasing emitter current caused an increasing Vbe (and thus increasing emitter current further) you would have positive voltage feedback. The transistor emitter current would continue to increase until 2 possible outcomes:

1) the high collector current (supporting the high emitter current) caused the transistor to saturate. Where it no longer operates in the linear region and your circuit would fail since it depends on that.

2) An oscillation would arise due to other parasitics / components or operating modes in the system.

It is ok to create positive feedback as long as it is controlled.

A classic example is providing some hysteresis on a comparator.
Or making an oscillator.

In a more subtle sense, I have also seen amplifier applications where it is clear that positive feedback exists but the circuit has no stability trouble because the net overall feedback is still negative. In otherwords, you can get away with having positive feedback provided you also supply sufficiently more negative feedback so the net result is negative.

These concepts apply across all frequencies as well. The amount of feedback may be frequency dependent.

 

[zachtheterrible]

Thanks a lot everyone, that makes much sense. One last question (hope its the last). What does R5 do? What be the point? I've even seen circuits where that resistor is so small, it would seem to short out the circuit. What gives?

 

[samcheetah]

i think R5 is the load that the amplifier is driving. because usually after the coupling capacitor the load is connected.

 

[Nigel Goodwin]

Thanks a lot everyone, that makes much sense. One last question (hope its the last). What does R5 do? What be the point? I've even seen circuits where that resistor is so small, it would seem to short out the circuit. What gives?

R5 is clearly labelled as the load resistor, it's also not a small value, but a very large one - 10,000,000 ohms. By referencing the output of the capacitor to ground, the signal will be referenced to ground as well.