Negative feedback?

Solid Snake · 27th February 2003, 05:24 AM

I've heard a lot of things about negative feedback but I don't understand how you can reduce noise more than you can reduce the signal. If you inject part of the output into the input again, you are reducing the noise, but at the same time, the signal. So when you crank the signal back up, isn't the noise there too? Where exactly does the noise come from?

phase_accurate · 27th February 2003, 05:36 AM

You will of course not reduce any noise that is fed to the input of your circuit.
What your feedback circuit actually does is reducing noise and distortion of the stages that are INSIDE of the loop.

Regards

Charles

Solid Snake · 27th February 2003, 08:46 PM

Does NFB solve the problem of non linear output due to unregulated rails that sag when you draw a lot of power from them?

Jeff R · 27th February 2003, 09:03 PM

Yes - to the extent that it can. If the voltage drops so low that linear operation cannot be maintained, then feedback will lose the battle.

Solid Snake · 27th February 2003, 09:27 PM

For this amplifier I tried to build, I had an NPN/PNP pair with emitters grounded (first stage). The first stage NPN's collector was tied to the secondary stage's PNP's base whose emitter was on the + rail. The opposite was true for the negative rail. In other words, both stages were common emitter (inverting). The first stage was common to ground, the second stage, one the NPN was common to the - rail and the PNP was common to the + rail. Both stages combined formed a non-inverting amplifier, so how could I use NFB in a situation like this? How could I get global NFB without using op-amps?

janneman · 28th February 2003, 08:18 AM

You could put a NFB loop from the emitter of the last stage to the base of the input stage, but it is not strictly global.

To use global NFB you need another inverting stage, which is easy: base to collector of last stage, and give it equal emitter and collector resistors, so its gain is 1, but the collector has the inverted signal. That can be fed back to the input.

Jan Didden

peranders · 28th February 2003, 08:36 AM

Quote:

Originally posted by Solid Snake
I've heard a lot of things about negative feedback but I don't understand how you can reduce noise more than you can reduce the signal. If you inject part of the output into the input again, you are reducing the noise, but at the same time, the signal. So when you crank the signal back up, isn't the noise there too? Where exactly does the noise come from?

You can't reduce any noise, therefore it's important to make the first stage low noise and it must amplify so much so the amplified noise is less than the noise created in the secnd stage.

Feedback can't handle noise, sorry.

Hasn't Texas got a large document, very wellwritten about feedback?

Can't find it right now but here are lot's of info.

http://dspvillage.ti.com/docs/catalo...applicationId=

traderbam · 28th February 2003, 08:37 AM

Snake,
Another way is to put a resistor from input emitters node to ground, then a feedback resistor for output to the emitters node.
Bam

janneman · 28th February 2003, 08:43 AM

Quote:

Originally posted by peranders

You can't reduce any noise, therefore it's important to make the first stage low noise and it must amplify so much so the amplified noise is less than the noise created in the secnd stage.

Feedback can't handle noise, sorry.

Hasn't Texas got a large document, very wellwritten about feedback?

Can't find it right now but here are lot's of info.

http://dspvillage.ti.com/docs/catalo...applicationId=

Per, beg to differ, fb can handle noise as long as it is generated inside the loop. Anything inside the loop is reduced with the fb factor, noise, thd, temp drift, you name it.

Jan Didden

Sch3mat1c · 28th February 2003, 04:12 PM

You know, despite the >40dB (probably 60) NFB in my SS stereo (Onkyo TX-2500), it is still incredibly noisy (part of why I don't use it). Makes me wonder........

Tim

27th February 2003, 05:24 AM	#1
Solid Snake diyAudio Member Join Date: Feb 2003	Negative feedback? I've heard a lot of things about negative feedback but I don't understand how you can reduce noise more than you can reduce the signal. If you inject part of the output into the input again, you are reducing the noise, but at the same time, the signal. So when you crank the signal back up, isn't the noise there too? Where exactly does the noise come from?

28th February 2003, 08:18 AM	#6
janneman diyAudio Member Join Date: May 2002 Location: Where Germany, The Netherlands and Belgium meet Blog Entries: 6	NFB You could put a NFB loop from the emitter of the last stage to the base of the input stage, but it is not strictly global. To use global NFB you need another inverting stage, which is easy: base to collector of last stage, and give it equal emitter and collector resistors, so its gain is 1, but the collector has the inverted signal. That can be fed back to the input. Jan Didden

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Negative Feedback?	Binaural	Tubes / Valves	15	12th March 2009 07:05 PM
Negative Feedback	AmpBuilder225	Tubes / Valves	7	20th February 2009 08:05 PM
Less Negative Feedback Is Better?	CarlosT	Chip Amps	25	14th December 2006 10:18 PM
Negative Feedback	Jack Thomas	Tubes / Valves	6	5th May 2003 12:38 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

27th February 2003, 05:36 AM	#2
phase_accurate diyAudio Member Join Date: May 2002 Location: Switzerland	You will of course not reduce any noise that is fed to the input of your circuit. What your feedback circuit actually does is reducing noise and distortion of the stages that are INSIDE of the loop. Regards Charles

27th February 2003, 08:46 PM	#3
Solid Snake diyAudio Member Join Date: Feb 2003	Does NFB solve the problem of non linear output due to unregulated rails that sag when you draw a lot of power from them?

27th February 2003, 09:03 PM	#4
Jeff R diyAudio Member Join Date: Nov 2002 Location: Florida	Yes - to the extent that it can. If the voltage drops so low that linear operation cannot be maintained, then feedback will lose the battle.

27th February 2003, 09:27 PM	#5
Solid Snake diyAudio Member Join Date: Feb 2003	For this amplifier I tried to build, I had an NPN/PNP pair with emitters grounded (first stage). The first stage NPN's collector was tied to the secondary stage's PNP's base whose emitter was on the + rail. The opposite was true for the negative rail. In other words, both stages were common emitter (inverting). The first stage was common to ground, the second stage, one the NPN was common to the - rail and the PNP was common to the + rail. Both stages combined formed a non-inverting amplifier, so how could I use NFB in a situation like this? How could I get global NFB without using op-amps?

28th February 2003, 08:37 AM	#8
traderbam diyAudio Member Join Date: Jan 2002 Location: Earth	Snake, Another way is to put a resistor from input emitters node to ground, then a feedback resistor for output to the emitters node. Bam

28th February 2003, 04:12 PM	#10
Sch3mat1c diyAudio Member Join Date: Jan 2003 Location: Milwaukee, WI	You know, despite the >40dB (probably 60) NFB in my SS stereo (Onkyo TX-2500), it is still incredibly noisy (part of why I don't use it). Makes me wonder........ Tim

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)