Fun with Feedback

What people are working on at the moment
Post Reply
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#91 Re: Fun with Feedback

Post by Nick »

Not sure it's about keeping the feedback equal but I have look at this more with solid state so may be different. Normally the group delay in the amp will mean that as you go up the frequency range the delay will turn the negative feedback into positive feedback and so will oscillate. By ensuring the gain of the amp is less than unity by the time the delay creates positive feedback means the amp is stable. It's why making a solid state amp stable into reactive and low loads is not trivial as the load alters the loop phase shifts.

Btw. I would argue that ignition advance is not a feedback system as there is no error signal. It could be considered feedforward though.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#92 Re: Fun with Feedback

Post by JamesD »

Agree with Nick - the small value bypass caps are there to ameliorate the phase angle at high frequencies and to keep the feedback negative at those frequencies - it is a side effect that one might consider this as maintaining values of negative feedback across the positive gain bandwidth of the amplifier...
Cressy Snr
Amstrad Tower of Power
Posts: 10552
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#93 Re: Fun with Feedback

Post by Cressy Snr »

JamesD wrote: Sat Jul 09, 2022 5:16 pm Agree with Nick - the small value bypass caps are there to ameliorate the phase angle at high frequencies and to keep the feedback negative at those frequencies - it is a side effect that one might consider this as maintaining values of negative feedback across the positive gain bandwidth of the amplifier...
Yes but that’s what I’m trying to get across. :?

Here’s what I thought I’d learned:
The phase shift network on the input stage anode has the effect of reducing the loop gain at frequencies from about 20 kHz upwards without affecting the phase shift in the critical region and the phase advance cap across the feedback resistor is an attempt to steer the phase curve (that Nyquist snail thing) away from the 180 deg point at very high frequencies as the loop gain is passing through unity This increases the stability margin a little bit further. The cumulative effect of all this manipulation is that you get the feedback to be equal in strength across the audio band but the VHFs get pushed right the way down outside it and kept negative, so oscillation is prevented.
That’s how I understand it but not being an EE I seem not to be using the right frames of reference. I’m trying my best….honest. I don’t know how else to explain it. It’s all the technical language I’ve got. It’s nearly driven me mad and all things considered I’d thought I was getting there :cry:
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#94 Re: Fun with Feedback

Post by IslandPink »

No that's fine, I'm up to speed so to speak, Thanks all.
"Once you find out ... the Circumstances ; then you can go out"
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#95 Re: Fun with Feedback

Post by Nick »

No, I think we are all saying the same thing. Its just when you talked about keeping feedback equal, I imagined you were talking about way above 20k or so. Partly because I have spent more time playing with this stuff with solid state where its above 200kHz where you normally start to find things get interesting, and often in the low MHz range where it gets really fun.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#96 Re: Fun with Feedback

Post by Nick »

Oh, and with push pull you can find all sorts of little local areas of instability as the devices are swept across their ranges and gm and self capacitance varies and you can find little RF burst hidden in the audio waves.

Oh, and then there is the fun of taming the clipping behaviors and stopping the stage locking up as it comes out of clipping (or at least making all the amp come out of clipping at the same time). And then you start having to fight not saturating junctions and you discover the fun with Baker clamps (which is one of those circuits where you look at it and wonder why it does anything at all), and having to find fast enough diodes to do the job.

All good stuff covered in Bob Cordell's book.

Buy anyway, thats a little burst of the fun I had trying to make a solid state amp that did what I wanted.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Cressy Snr
Amstrad Tower of Power
Posts: 10552
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#97 Re: Fun with Feedback

Post by Cressy Snr »

Yikes. :shock:
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#98 Re: Fun with Feedback

Post by izzy wizzy »

The Bob Cordel book is a good read and I have no intention of building a SS amp.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#99 Re: Fun with Feedback

Post by JamesD »

Hi Steve,

Its not that your understanding is wrong - what you describe happening is correct but the cause and effect is described the wrong way round for conventional description - although the mathematics of sine waves frequency and phase response allows both descriptions... Conventionally we would describe the frequency response changes as causing the phase response changes whereas your description suggests the phase response changes drives the frequency response changes... mathematically they both happen at the same time as they are 'two sides of the same coin' but conceptually it is the frequency response change that drives the phase response change as frequency response is an absolute parameter than can be measured at each unique point by itself and the phase response is a relative measurement and can only be measured by reference to some agreed standard - typically the phase of the stage input signal for a circuit stage phase measurement and because of this we give frequency response precedence...

As Nick says, we are all saying the same things but saying it differently - a bit like talking with an American - our explanations are separated by a common language :)
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#100 Re: Fun with Feedback

Post by Paul Barker »

Ill just stay in the stone age. I’ll get my coat.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
Cressy Snr
Amstrad Tower of Power
Posts: 10552
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#101 Re: Fun with Feedback

Post by Cressy Snr »

JamesD wrote: Sun Jul 10, 2022 12:11 am Hi Steve,

Its not that your understanding is wrong - what you describe happening is correct but the cause and effect is described the wrong way round for conventional description - although the mathematics of sine waves frequency and phase response allows both descriptions... Conventionally we would describe the frequency response changes as causing the phase response changes whereas your description suggests the phase response changes drives the frequency response changes... mathematically they both happen at the same time as they are 'two sides of the same coin' but conceptually it is the frequency response change that drives the phase response change as frequency response is an absolute parameter than can be measured at each unique point by itself and the phase response is a relative measurement and can only be measured by reference to some agreed standard - typically the phase of the stage input signal for a circuit stage phase measurement and because of this we give frequency response precedence...

As Nick says, we are all saying the same things but saying it differently - a bit like talking with an American - our explanations are separated by a common language :)
Thanks James. That makes things much clearer. I’ve never been easy to teach. :)
Sgt. Baker started talkin’ with a Bullhorn in his hand.
Cressy Snr
Amstrad Tower of Power
Posts: 10552
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#102 Re: Fun with Feedback

Post by Cressy Snr »

Paul Barker wrote: Sun Jul 10, 2022 8:43 am Ill just stay in the stone age. I’ll get my coat.
:mrgreen:
Sgt. Baker started talkin’ with a Bullhorn in his hand.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#103 Re: Fun with Feedback

Post by JamesD »

Paul...I'll swop my amplifier knowledge and experience for yours... Anytime!
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#104 Re: Fun with Feedback

Post by Paul Barker »

Lol. I know nothing, except to learn from out dated old ways!
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#105 Re: Fun with Feedback

Post by Dave the bass »

Ta for the answers re the 200pF cap all.

Aah, So it's the capacitive reactance of that cap that causes the change in frequency dependant load on that 1st stage.

I dun learned.
"The fat bourgeois and his doppelganger"
Post Reply