OP AMP EXPERIENCES and HOW TO

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
ok so ive read and looked and scuffed and scraped the forums for various information on 'HOW TO' with opamps (for the uneducated!) and ive learned a thing or two, but I know that there are people out there who can shed more light on these little buggers for us so I want to start a thread that people can write some up to date experiences in for descusion,
I am going to start with an experience !! and a question ??

I have implemented the new LM4562 into my ARCAM DV89 which used to have 2x OPA2134, to be honest I think they are far better than any I have heard before, tonally a lot more balanced its like someone has gone accross the sound stage with an iron! I found that the 2134 in this app was fatiguing in the upper mid, sometimes drowned out and sometimes aggressive in a small bandwith. iv'e also used the ad8620 and found that the low end was very fake sounding a bit cloaked like all basslines sounded the same, also I have fiddled with the ne5532's and found them to be very honest indeed but a little lathargic.

and so bringing me to....
what on earth is it that influences the sound and why and how in different curcumstances, ive read that I/V configuration requires a fast opamp, and that the datasheet only reveals data for a specific test case which more often is far from the actual reality and also that what one considers to be necessary for accuracy is only applying a perceived fix when is actually just the brain telling you i like it cos i did it!
I think what we need is like all the information whacked onto a table or something like that containing the underpinning variables like configurations loads capacitance and other variables
 
analog_sa said:


But why start with such a complicated device? Why not first ask how capacitors, resistors and wire influence the sound?

exactomondo!
I do beleive that consideration runs both ways so I mean in any 'system' resistors and capacitors and wire affect the functionality of any said connected device and here I wish to discuss the outline, exactly what does and crucially under what circumstances (system) they affect (ie does this affect any system in a positive manner)
its really about gaining prospective.
 
opamp impedance input output

tawn10 said:


exactomondo!
I do beleive that consideration runs both ways so I mean in any 'system' resistors and capacitors and wire affect the functionality of any said connected device and here I wish to discuss the outline, exactly what does and crucially under what circumstances (system) they affect (ie does this affect any system in a positive manner)
its really about gaining prospective.

I quote myself and start with a basic contribution to the thread please if you contribute then place a subject to make searching easier for those that are hunting

now OP AMP out impedance, in general driving a lower impedance affects the linearity of 'a' device a 600ohm load is usually the cutting point for an average opamp, (the standard set by old equipment using very low <by todays standard> impedance transformers) any lower than this and the opamp cant handle the AMPAGE to drive the load. most new kit has input impedance around 1-10k sometimes higher for solid state gear but watch it because the gain needed to drive the higher impedance is graeter, this can also contribute to distortion as some devices don't like driving high gain (like my bicycle wont go on the motorway) and some devices will drive low impedance loads (my bike does however go better than my car down a towpath)


most new kit has input impedance around 1-10k

you will see that between stages in an audio out or input are often 2k2 resistors or 1k or something similar
 
Hey Tawn 10,

I am sure this will start an uproar, but this needs to be addressed.

I feel for ya! I am new here and to electronics DIY. Every time I start a thread, I Post what I am looking to do (end result). Obviously there are many paths sometimes and sometimes just one path. What I get here and there is a "one liner" of about 5% of the information I need. And when I ask for more information and to elaborate..... Nothing. What is the purpose of the post, to make me feel stupid? It certainly didn't accomplish anything else. I don't understand what is really going on. This is supposed to be a DIY site that people help each other. Obviously the more experienced will do more of the helping but that is how it works and some day the beginner will be the teacher. One guy at least tried to nudge me along to the next step in the thought process about what I need to learn and I think that is acceptable if they don't just want to outright give exactly what you need.

I can build an entire house (except the foundation and don't ever want to) and if someone (a beginner) asked How can I put a big picture window in where I have a very small one? I wouldnt flip off, "well determine the R/O needed and get the materials required. Send pictures of the finished window." If someone cuts an area big enough for a huge window on a supporting wall without supporting everyting until the Header can be installed, it can do a lot of structural damage and sometimes bring down the roof and/or floor above. This can get expensive and do irrepairable damage and give someone a bad taste on DIY.

How do we get this back to what it is supposed to be? I am sure the Pros and true EE types get bored with us new guys but something has to give. It's like politics, Feed the rich at the expense of everyone else and bring down the middle class to low class. Now there only two classes and revolution is soon to follow. Hopefully there won't be just two types here, the knowledgeable/experienced and the beginners (me) and rarely will we be at the same party.

I hope everyone can understand this point of view. I am just explaining what is like down here at the bottom and I am sure I am not alone! It's just so dark I can't see anyone else! LOL

Regards//Keith
 
op amp audio quality

We should diversify between the advances of new circuitry inside op amps and the interaction with the outside world.
The 4562 indeed is a very nice one. The reason WHY is not to be found in the spec file, but you may find a kind of answer when looking at the article published in the EDN magazine:
http://www.hawkaudio.nl/news.htm. Just download the article (and my comments) there. It's remarkable that major manufacturers now are actually "listening"!
The matter of in- and output impedances is an entirely different thing. The output impedance is quite low because of the feed back. Without that feed back it would be quite high. Looking at the "open loop response" you'll see the amplification dropping down with some 6 dB/oct starting at a very low frequency. The newest op amp types now start falling down from around 1 kHz and that's a real improvement.
Now if you load an op amp with 600 Ohm the feedback factor will go down with frequency. This means that distortion goes up! And the output impedance goes up as well.
Considering the small transistor dies in the totem pole at the output of the op amp it cannot dissipate very much power, happily there's a current protection in it.
So whenever using an op amp consider to use a circuit thereafter with a relatively high input impedance, in that way avoiding the distortion going up.
This phenomen is one of the reasons tube preamps, when connected to a common CD-player or tuner, sound better then SS ones. The input impedance is high!
 
Re: op amp audio quality

johnrtd said:
We should diversify between the advances of new circuitry inside op amps and the interaction with the outside world.
The 4562 indeed is a very nice one. The reason WHY is not to be found in the spec file, but you may find a kind of answer when looking at the article published in the EDN magazine:
http://www.hawkaudio.nl/news.htm. Just download the article (and my comments) there. It's remarkable that major manufacturers now are actually "listening"!
The matter of in- and output impedances is an entirely different thing. The output impedance is quite low because of the feed back. Without that feed back it would be quite high. Looking at the "open loop response" you'll see the amplification dropping down with some 6 dB/oct starting at a very low frequency. The newest op amp types now start falling down from around 1 kHz and that's a real improvement.
Now if you load an op amp with 600 Ohm the feedback factor will go down with frequency. This means that distortion goes up! And the output impedance goes up as well.
Considering the small transistor dies in the totem pole at the output of the op amp it cannot dissipate very much power, happily there's a current protection in it.
So whenever using an op amp consider to use a circuit thereafter with a relatively high input impedance, in that way avoiding the distortion going up.
This phenomen is one of the reasons tube preamps, when connected to a common CD-player or tuner, sound better then SS ones. The input impedance is high!

so is it that as the feedback factor lowers (as with open loop) the circuit (source) impedance is seen by the load? and the reason that the feedback factor lowers is that the gain required to drive the 600ohm load is less?
 
AX tech editor
Joined 2002
Paid Member
Re: Re: op amp audio quality

tawn10 said:


so is it that as the feedback factor lowers (as with open loop) the circuit (source) impedance is seen by the load? and the reason that the feedback factor lowers is that the gain required to drive the 600ohm load is less?

The output impedance of the opamp is always 'seen' by the load. The feedback factor lowers the open loop output impedance with the same factor as it reduces the ol gain to the closed loop gain.
The feedback factor is the difference between ol gain and cl gain. As ol gain fall with frequency, the feedback factor falls with frequency, so the positive effects of feedback, like lowering the output impedance and distortion, also fall with gain.

The reason these opamps are designed with the particular fall of of ol gain with frequency is the phaseshift by internal caps. If the phase shift reaches 180 degrees, then the negative feedback turns into positive feedback and the opamp will oscillate. So, they roll of the gain such that before the 180 phaseshift frequency, the gain is reduced below 1. An opamp with gain below one will not oscillate at that frequency.

Jan Didden
 
tawn10 said:
what on earth is it that influences the sound and why and how in different curcumstances

I think what we need is like all the information whacked onto a table or something like that containing the underpinning variables like configurations loads capacitance and other variables

There are so many possible combinations of contributing factors that interact with the specific application (I/V converter, noninverting line stage, multiple-feedback active filter) that a comprehensive list correlating contributing factors with end results is not really available.

The two weak points in op amps that I believe can cause major interaction with an application are the thermal coupling of input and output stages and the imperfect electrical isolation of the input stage transistor pair from the chip substrate, these are well-known issues that Walt Jung has written about and proposed solutions for, but these solutions require additional components.
 
janneman: " ... If the phase shift reaches 180 degrees, then the negative feedback turns into positive feedback and the opamp will oscillate. ..."

I did not realize this ... very enlightening. I suppose to be hyper-critical of op-amp design one should actively evaluate the point (frequency?) at which any significant phase shift begins. Is there a spec that indicates such, or are we all stuck examining the sometimes crypitic graphs ...? (I know about slew rates, etc., but a specific spec that gives some indication of the phase shift curve "knee" would be nice ... maybe a percentage of unity gain bandwidth?)
 
OK so I guess that some people kind of missed the point then, when I started this thread, I intended it to be kind of a cookbook for the kitchen, I am very gratefull for all the feedback (negative and positive) pardon the pun it was just aimed at people like myself who have a little knowledge but not enough and a lack of background education,
you see I enjoy DIY electronics as my fathers before me so I didnt want to debate how you cant sum up all the possibilitys in one simple chart or read a debate out of the threads league by you just showing off (you know who you may be!) I just want help and to help others to learn something by means of some plain and simple logical teachings, think of it as a 'stuck in first gear' sucks corner.:confused:
 
peufeu said:
One should read "The Art of Electronics". Good style, interesting read.

> Is there a spec that indicates such

It's in the datasheets, but you need the knowledge to understand it... read the above book and you'll feel a lot better, not shooting in the dark anymore. It will save you a lot of time.

looks like just the ticket, I will endeavour to add this to the collection!
 
d-i-y hobby

tawn10 said:
OK so I guess that some people kind of missed the point then, when I started this thread, I intended it to be kind of a cookbook for the kitchen, I am very gratefull for all the feedback (negative and positive) pardon the pun it was just aimed at people like myself who have a little knowledge but not enough and a lack of background education,
you see I enjoy DIY electronics as my fathers before me so I didnt want to debate how you cant sum up all the possibilitys in one simple chart or read a debate out of the threads league by you just showing off (you know who you may be!) I just want help and to help others to learn something by means of some plain and simple logical teachings, think of it as a 'stuck in first gear' sucks corner.:confused:
Dear Tawn I understand your feelings. It's a good thing to have a hobby outside your professional job. Moreover it's nice being busy in some "musical" project. But, sorry to say, the times of the electronic audio hobby using a hammer and a screwdriver are (happily) long past. Nowadays one may enjoy modern electronics and modern circuitry in order to get a home-built audio system with perfect sound. The pitfall though is in the complexity of the circuit. There are two ways of doing this:
1. the simple way building something developed by someone else or offered by some company as a kit.
2. start studying electronics and when doing so try to understand the difference between "electronic" engineering and "audio electronic"engineering.
Mind you, most current (graduated) electronic engineers don't know about the working of a semiconductor or transistor or op amp. They learned using "black boxes".
Even a loudspeaker design, at first sight quite simple, is complex. There are still amateurs "hammering" together their loudspeaker systems and thereafter "tuning" them using their ears .....
Just go out and listen to some different types of equipment and you'll find that although a lot of labor and engineering is invested there still are things in the market not giving any musical emotion.
 
AX tech editor
Joined 2002
Paid Member
Re: Re: OP AMP EXPERIENCES and HOW TO

nuvistor said:


There are so many possible combinations of contributing factors that interact with the specific application (I/V converter, noninverting line stage, multiple-feedback active filter) that a comprehensive list correlating contributing factors with end results is not really available.

The two weak points in op amps that I believe can cause major interaction with an application are the thermal coupling of input and output stages and the imperfect electrical isolation of the input stage transistor pair from the chip substrate, these are well-known issues that Walt Jung has written about and proposed solutions for, but these solutions require additional components.

The thermal coupling and feedthrough from output to input is a non-issue. Opamps nowadays have routinely temp induced drifts of a few microvolt (10^-6V) per degree. In fact, due to the very smart designs, thermal effects in opamps is orders of magnitude less than in discrete systems.

Walt Jung wrote his article 20+ years ago (or was it 30?). Modern opamps don't display this effect.

Jan Didden
 
AX tech editor
Joined 2002
Paid Member
FastEddy said:
janneman: " ... If the phase shift reaches 180 degrees, then the negative feedback turns into positive feedback and the opamp will oscillate. ..."

I did not realize this ... very enlightening. I suppose to be hyper-critical of op-amp design one should actively evaluate the point (frequency?) at which any significant phase shift begins. Is there a spec that indicates such, or are we all stuck examining the sometimes crypitic graphs ...? (I know about slew rates, etc., but a specific spec that gives some indication of the phase shift curve "knee" would be nice ... maybe a percentage of unity gain bandwidth?)

Eddy, no negatives intended, but most here don't realize the most elementary stuff and yet proceed to pronounce that opamps are bad because bla bla bla.


The phase shift and roll-off are opamps 101. It is totally understood and routinely used in design. It IS in the datasheets, look at open loop gain vs freq and open loop gain vs phase. From these two graphs, a competent designer can design completely stable and flat amplifiers with significant distortion reduction and very low Zout. No magic.

Jan Didden
 
Hey, I might not be intelligent but I know that im not alone, sometimes we walk our own path and sometimes we follow others thats the only order that is for sure, I didnt think what I asked was such a tall order and so left I am with a question, Who exactly do you work for?

anyway some people are giving some 'sound input' (pardon another pun) thanks guys

Janneman you seem to have exactly what many of us in the dark side of the sphere are looking for, it would be great if you can break that down into "I do'nt understand" syntax (say the simplest example you can think of or a link to something)

Thanks again
 
AX tech editor
Joined 2002
Paid Member
tawn10 said:
Hey, I might not be intelligent but I know that im not alone, sometimes we walk our own path and sometimes we follow others thats the only order that is for sure, I didnt think what I asked was such a tall order and so left I am with a question, Who exactly do you work for?

anyway some people are giving some 'sound input' (pardon another pun) thanks guys

Janneman you seem to have exactly what many of us in the dark side of the sphere are looking for, it would be great if you can break that down into "I do'nt understand" syntax (say the simplest example you can think of or a link to something)

Thanks again

OK. Take a data sheet of an opamp, any opamp you fancy. Look at the open loop gain vs frequency graph. Say you want to built a gain-of-two amp with it, that's 6dB gain. Draw a horizontal line on the graph at 6dB. Note down the freq at which this line cuts into the gain-freq graph.

Now go to the phase vs freq graph, and draw a vertical line at the freq point noted above and check what the phase shift is at that freq. Is is close to 180 degrees? You will have an oscillator. Is it above 135 degrees? It will not oscillate but ring as hell after a pulse or squarewave.

If the pahse shift IS too high, what can you do? There are compensation techniques but I won't go into that. One simple cure is to increase the closed loop gain (if that fits your app of course). Repeat the whole thing with a closed loop gain of say 10x which is 20dB. You'll see that the higher the closed loop gain, the less your stability problems. That's logical: less closed loop gain is less feedback.

Jan Didden
 
Tawn,

1/4 of a century ago i had elementary electronics classes, also did amplifier design classes as a sidekick for free.
Even then, i was somewhat dissapointed about the far away from practical implementation level of the theory lectures and class notes.
In those years, the task given was to read the book and come back if you had questions after you finished it.

At the time i also participated in Johnrtd's diy audio mag/club/meetings.
Why the gifted F. Savelkoul, mentioned on the Hawk site, decided to flunk his electronics studies at Philips U. after a couple of years is something i'm able to comprehend.
I also tend to agree with Johnrtd's view towards the current generation of powerpoint presentation engineers, EE or otherwise.

If you desire to understand more than the average screwdriver diy hobbiest there's no way other than opamps 101 and reading datasheets.
You have the good fortune of being able to download complete books on opamps from Walt Jung for free and just about any datasheet as well. Fossil audio DIYs had to invest a lot of time and money to get info like that.
I love kitchen work too, but i've practiced cutting for years to avoid chopping my fingertips off before i bought the most expensive set of pro chef knives.(and still managed to separate the top of my right finger with a Chinese combat knife)

(I see John vdS is still wearing his Art Blakey groupy pants. :clown: )
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.