NFB capacitor

We all start somewhere
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#16 Re: NFB capacitor

Post by Nick »

BTW, the ones I found are only 22uf 50v, I use 100uf 50v for my amps, but the ones I will send should be ok to try.
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.

#17 Re: NFB capacitor

Post by Cressy Snr »

OK.
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.

#18 Re: NFB capacitor

Post by Cressy Snr »

Interestingly the DC offset on the output stage varies with the position of the volume control. As the pot is turned towards its maximum clockwise travel, the DC offset falls.
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#19 Re: NFB capacitor

Post by Nick »

I would guess in the absence of of a cap on the input the resistance looking into the pot from the LTP base is whats causing the change. What I would suggest trying is a resistor to ground on the front of the amp so the resistance to ground on the +ve LPT base is the same as the resistance seen to the virtual ground of the output by the -ve LTP base. Then isolate from any other DC path to ground with a input cap.

But again, guessing because of not having the circuit.
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.

#20 Re: NFB capacitor

Post by Cressy Snr »

Nick wrote: Mon Jun 05, 2023 8:26 pm I would guess in the absence of of a cap on the input the resistance looking into the pot from the LTP base is whats causing the change. What I would suggest trying is a resistor to ground on the front of the amp so the resistance to ground on the +ve LPT base is the same as the resistance seen to the virtual ground of the output by the -ve LTP base. Then isolate from any other DC path to ground with a input cap.

But again, guessing because of not having the circuit.
I obviously can't put up the circuit as it's a commercial product, but I've been using it in conjunction with an old transistor tutorial to try to work out what is going on but I'm new to this and probably wrong. Basically (I think) there is an LTP input sitting on a CCS. This pair feeds into a cascoded VAS that sits on a degenerated current mirror. To prevent it getting all upset, the VAS is buffered from the output stage by a power transistor emitter-follower which takes over the heavy lifting and drives the output stage.
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#21 Re: NFB capacitor

Post by Nick »

Ok, so if you consider the two transistors in the LTP. The current through those two transistors will be controlled by the current into each of the bases. So if we assume that the input is at 0v, and the output is also at 0v. The base of each NPN transistor will have to be below 0v for current to flow into its base junction (assuming conventional current). The amount of current with be governed by the voltage at the base and the resistance between that base and the 0v point. Assuming you have say a 47k resistor between the input and ground, that will set the current into the base. If you have a 9k feedback resistor into the other base, then feedback will adjust the voltage on the output of the amp, such that the PD across that resistor will match the input current. If you put a cap on the input of the amp, the only DC path to ground for the input base current will be through the 47k resistor, hence I would expect that adding the blocking cap would alter the output voltage as you have seen.

I would also expect that without a blocking cap on the front, the DC on the output will be different when a input is connected or not because of the resistance looking into the source output being in parallel with the 47k.

What I do (FWIW), is keep the input resistor (26.1k) the same as the feedback resistor (26.1K), then isolate the input with a cap. That way the current conditions for both sides of the LTP will be identical.

I assume that the feedback point of the LTP will have a blocking cap and resistor to ground, and at AC, the actual feedback ratio will be set by the feedback resistor and that resistor to ground via the cap. So that way you can keep the AC feedback ratio as you want, and the DC conditions equal to give you 0v output (assuming perfectly balanced LTP) without needing a servo. I also use identical caps on the input and between the -ve base to ground to attempt to keep any distortion generated in the caps equal and hopefully cancel out.

And because the cap on the LTP to ground has one side to ground, its nice and isolated from RFI, but the one on the input has both terminals floating, so I found that shielding the cap case to ground removes a source of RFI input.

Also, what I described without the input blocking cap, will perfectly match what you said you see with the position of the pot on the input changing the DC at the output of the amp because it will be applying a varying parallel resistance across the input 47k resistor so DC current will vary, and DC voltage on the output will track.

Hope that make some sort of sense.
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.

#22 Re: NFB capacitor

Post by Cressy Snr »

Cheers Nick,
Things are starting to make sense now. :)
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.

#23 Re: NFB capacitor

Post by Cressy Snr »

I’ve also seen other circuits where the feedback resistor matches the input resistor. They all have input DC blocking caps I can see why the resistors match now.
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#24 Re: NFB capacitor

Post by Nick »

Yep, it seems to be another thing that's more often left unsaid. I may well be wrong but its what I think is going on.
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: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#25 Re: NFB capacitor

Post by Nick »

Yep, the actual value of the feedback resistor doesn't set the feedback level, its the ratio of the PD that its a part of.
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.

#26 Re: NFB capacitor

Post by Cressy Snr »

Aye, its something else that seems logical, now I’ve realised with your help what is going on.
As The Korgis said, 'everybody’s got to learn sometime.' :)
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

#27 Re: NFB capacitor

Post by IslandPink »

Now there's a thing - I never knew that the Queen had taught them to speak.
"Once you find out ... the Circumstances ; then you can go out"
Post Reply