New Posts New Posts RSS Feed - Inside op-amps
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Welcome to the Graham Slee Audio Products Owners Forum

 

Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here)

This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd


Inside op-amps

 Post Reply Post Reply Page  <1 89101112 14>
Author
BAK View Drop Down
Senior Member
Senior Member
Avatar

Joined: 14 Mar 2010
Location: Kentucky, USA
Status: Offline
Points: 1744
Post Options Post Options   Thanks (0) Thanks(0)   Quote BAK Quote  Post ReplyReply Direct Link To This Post Posted: 01 Aug 2022 at 1:41pm
"We can use the 15nF + 4n7 capacitors in the NFB, which makes 3n6 approx."

That is in series.
Bruce
AT-14SA, Pickering XV-15, Hana EL, Technics SL-1600MK2, Lautus, Majestic DAC, Technics SH-8055 spectrum analyzer, Eminence Beta8A custom cabs; Proprius & Reflex M or C, Enjoy Life your way!
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 01 Aug 2022 at 2:47pm
Yes, and here is a typical application,

Typical Single Rail RIAA Configuration

That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 01 Aug 2022 at 3:56pm
Capacitive loading of op-amps is often identified as 100pf.

This is the capacitance which if placed at the op-amp output, will form an RC filter with the op-amp's output resistance Ro.

Its corner frequency cannot be prior to its GBW crossing, or it will use up its phase margin and oscillate.

It is also a handy way of estimating Ro if the datasheet doesn't say.

The highest Fc for an NE5534 is 10MHz, where it crosses below unity gain.

Here there's a bit of a cheat because GBW is specified at 10MHz with both 100pf load and a 600 ohm load. The 600 ohm load effectively parallels Ro, and so the calculated Ro will be smaller.

From 1/2π f c, 1/2π 10,000,000Hz * 0.000,000,000,1F = 159 ohms

Subtract 600 ohms and Ro maximum is 216 ohms. The schematic shows 5 ohm emitter resistors, so this higher resistance must be the intrinsic emitter resistance, but at 16mA it will only add 1.6 ohms.

This is a catch-me-out and that's because it's a quasi-complementary, and the collector driven side will be at a far larger resistance. So believe the maths.

The reason the NFB capacitance can be so much bigger is because, at the same time as its phase margin is being narrowed, it compensates for that narrowing. It is in effect a large compensation capacitor.
Cap load effect
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (2) Thanks(2)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 01 Aug 2022 at 5:36pm
For a deeper understanding of op-amp stability search for "Operational Amplifier Stability by Tim Green"

WARNING: The 15 part papers are scattered all over the web and not maintained by TI. You might hit some Trojan sites by accident so ensure your computer and browser are well protected!
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (2) Thanks(2)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 02 Aug 2022 at 9:00am
If op-amps are so good then why is the internet full of instability cures?

(I mean engineering ones)

The audio nutter has a tendency to over indulge. I indulge just enough these days...

The op-amp with the biggest spec is NOT the best. I understand the word "boast" but does it really impress your friends? 

(or do they fall asleep?)

The top two most grotesque op-amps in my life are the NE5534 and the OP37. Below them are the modern variants that often have a number ending in 34 or 37. Have we not learned of one such "advancement" - LF411 to AD711?

Audio amplifiers once had economic importance. In today's world they no longer do. The great advancements in op-amps today are in low voltage 5V single rail types for handheld devices. Those are not Hi-Fi!

I doubt we're going to see a cure-all true high fidelity op-amp this side of history. So we have to put up with what we have.

In discrete design (transistors) the sensible designer would balk at trying to do what op-amp manufacturers suggest.

They all seem to think that we want to drive capacitive loads. The thing is, there have been capacitive loads since "the big bang". You make an emitter follower and it oscillates if connected to a capacitive load, so you isolate it!

With what do we isolate it? A resistor! But stating that is treason on some of those audio nutcase forums. Look! loonies exist. They're best ignored (so why am I writing this?)

One of the BIG DISCOVERIES in op-amp stability is RISO. RISO is the resistor I just eluded to. Why does common sense fly out of the window when dealing with op-amps? Probably because of the ground swell of nutcases?

RISO is hated! An op-amp will drive an 8 ohm headphone direct (NOT), take a look at all the "mint" HPA's. EN 61938 shows how it's done, and it works, but there is no "damping factor", and you must have damping factor! (NOT)

(ever taken apart a headphone diaphragm? Has it got any mass worth losing sleep over?)

(One exception is the AD823 which is collector output, but there's a catch in there too)

So what is the capacitive load? The cable of course! It depends on how long is a piece of cable.

The designer doesn't know what you'll use, so he should use an isolating resistor - R iso.

EN 61938 gives hints. It can be up to 120 ohms on a headphone output, and up to 2k2 on "gadgets" like phono stages. Only when it comes to the output of a control preamplifier does it get really tough... at 1k.

So why not use RISO?

This post will be continued after this break
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 02 Aug 2022 at 10:34am
As with all teaching aids, the entire picture is going to be shown up by a resulting plot on a simulator, and you can bet it's going to be only slightly similar to what's shown below. And that's because each point (of inflection) should be a decade away from the next, and good luck with that one!

Stabilising Decompensated Op-amp


The point, however, is to stabilise the awkward NE5534 or OP37 for signal gain that hits unity (0dB crossing), because unity gain isn't stable!

It isn't stable for the simple reason these op-amps are decompensated and only have stable gains of 3 or above (NE5534) or 5 or above (OP37).

So, as Cfb causes the amp to eventually reach a gain of 1, then it collides with the open-loop gain profile where it is -40dB/decade, and goes into oscillation.

So, we need to increase the gain so it collides with the open-loop gain profile where it's stable at -20dB/decade.

We don't want to interfere with the signal gain, so we do it in noise gain.

The provider of the noise gain boosts the gain back up beyond the wanted signal frequency range and somewhere way beyond our hearing, so it doesn't matter.

(hint: we can always filter that after Riso)

This is why we understand noise gain! It's nothing to do with naughty noise!

Now, me not being a teacher (I'm supposed to be a doer), my explanations are crude, and probably don't fully explain, but what the heck! I know what I'm doing, which used to be good enough.

But, you know, I've been plagued by clever arses for the past 24 years, who want to get it about that I know nothing, and so I do these technical posts to show that I do know. And maybe at the same time, it might teach the clever arses a thing or two.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (2) Thanks(2)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 02 Aug 2022 at 11:28am
(You know? I think I now understand where all this started. It was the day, the hour, the very minute, that a no mark managing director told me I knew nothing about phono preamplifiers!

Within seconds my brain computed all the phono preamplifiers I had designed for different purposes, including the understanding of ceramics! Yes, useless ceramic cartridges, which were only used by the top nightclubs before MMs became robust, and before the Technics SL's. Woah! What a sound!

OK, I was once told to design a phono preamp for broadcast work. Yes, and that managing director prevented me doing it my way - and yes, the BBC rejected it - but that wasn't my fault.

And neither was it my fault that this other managing director had mucked about with the phono preamplifier overnight, and changed things that he didn't understand, only to blame me for it soon after I got in the next morning.

So, what do you do? I picked up all my things and left, never to return.

I must have known my stuff because within hours he was on the phone making excuses for his behaviour. But the remark had cut deep - really deep - so I told him to get stuffed!

Perhaps with all his dealers, he might have thought that one day I may be selling my stuff to them? I don't know, but what I do know, is that all and sundry gave me the biggest negative welcome to the hi-fi industry. Other new kids on the block were welcomed with open arms by the hi-fi press - even if their fruits were second-rate.

And so, it's continued these last 24 years, and now the little children are at it too. The guy who started it all is now dead. I shall make no further comment about him.)
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
 Post Reply Post Reply Page  <1 89101112 14>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.234 seconds.