Two Words, "WALT JUNG", forgive my caps but for a man who works for analog devices, he seems to recommend the use of video op amps as buffers in his composite op amps "best of both worlds", my ears do nothing but credit his research into op amps as being quite worthy of exploration.. Most video op amps are not difficult beasts to tame, just follow the datasheets on best layout, choice of feedback resistor, decoupling you are home free, sometimes i think some folks fail to read the instructions and just delve in and then wonder why it all blew up..Datasheets can be your best friend or worst undiscovered enemy.
Colin
Colin
Horses for courses
Beppe, it might help to understand the following:
An "op-amp" is just a particular configuration of an amplifer consisting of: A differential input, voltage gain, bipolar output within the supply rails.
Beyond this there are op-amps designed for
a) DC or low frequency(like the OP07 I mentioned at the beginning)
b) Chopper stabilised with incredible low input offset current and voltages
c) Very fast video use where speed is the main requirement and noise and Dc accuracy are secondary
d) Audio use where linearity and low distortion are paramount (requiring an appropriately wide bandwidth)
Mr Curl's condescending remark is no help - would he really recommend that a beginner experiment with a video amp for his first audio op-amp project?
Beppe, it might help to understand the following:
An "op-amp" is just a particular configuration of an amplifer consisting of: A differential input, voltage gain, bipolar output within the supply rails.
Beyond this there are op-amps designed for
a) DC or low frequency(like the OP07 I mentioned at the beginning)
b) Chopper stabilised with incredible low input offset current and voltages
c) Very fast video use where speed is the main requirement and noise and Dc accuracy are secondary
d) Audio use where linearity and low distortion are paramount (requiring an appropriately wide bandwidth)
Mr Curl's condescending remark is no help - would he really recommend that a beginner experiment with a video amp for his first audio op-amp project?
vynuhl.addict said:
Colin, I can not deny any of this.
Especially this I find is true:
... sometimes i think some folks fail to read the instructions
and just delve in and then wonder why it all blew up..
Datasheets can be your best friend
Datasheets have Application Hints
where you can find the Best Way to use each Op-Amp.
In Reality:
Most DIY people are not readers & audio philosophers with Engineering Degree
This can make it a bit more difficult for 'normal' or 'newbiees'
to understand even 40% of such instructions, found in Datasheets.
we take for granted that people knows electronic terms
- impedance?
- feedback?
- open-loop?
- bias current?
I am NOT an Engineer ( and when see the abstract 'DIY' in some topics, I am most happy I am NOT )
What I know, has taken me YEARS and YEARS and 1.000 of schematics views
and reading and reading of DIY articles and projects.
I think it took me like 5 years, before I started to understand discrete transistor circuits
We should not judge other people, by our own status.
People are different, with knowledge in different subjects.
My list is a list of Op-Amps,
1. that are good for audio
2. do not require extreme or exceptional power supply
3. are forgiving and tolerant to use with normal audio resistance
Some of Analog Devices AD8xx & AD80xx op-amps, surely can enter this list.
I know this is a fact.
Other AD op-amps will not be good for someone, like beppe61,
that is starting to explore the 'nice world of op-amps for audio'
Regards
lineup
john curl said:
Thank you very much Mr Curl for your kind and valuable reply.
I would like to take this occasion to thank you sincerely for disclosing your knowledge and expertise through this forum in a lot of occasions.
And please excuse me if I mentioned some of your statements without your permission and maybe in a wrong way.
Thank you very much indeed.
Kind regards,
beppe
P.S. There are high open loop op amps that sound pretty darn good for audio
Interesting
lineup said:
Very good information, cliffforrest
This is what I tell people here, too.
Because I have seen so many, even experienced Audio Enginners at our forum - having long discussions about problems with Video Speed Op-amps they are trying to use for Audio.
.....
Regards,
lineup
Lineup Audio Lab
Thank you very much Sir for these very interesting directions.
I have to read for the next 6 months !
By the way it is fortunate to know that some cheap opamps are very good for audio purposes.
I will start there I think.
Let's say a nice NE5534 buffer ...
Thank you very much indeed.
Kind regards,
beppe
I hear you lineup, like yourself I am also on the path of discovery but what I have learned over the years since I was bitten by the audio bug
is that you can never discredit and must give serious look into the work that has been done before you, no matter which interest it has been done in "low distortion"ala self, "voicing" ala hugh dean whom i hold great respect for, and walt jung , Nelson Pass, Bob Cordell , John Curl along with so many others. As you point out it is probably true, some dont know what to make of datasheets , I was in this boat a few years back, I used to throw caution to the wind and buy the diy mods hyperbole, that is before i realized and well designed circuit is just that "well designed}" and the signal path makes the biggest difference. In the world of op amps, you really have it easy, here a company has spent many reasearch dollars to give you apllication notes, something that has been tested and tried to give the best average result with graphs to go along with it. 100% of the time when a manufacturer states bandwidth it is unity gain gbw only, not open loop, all op amps have poor open loop bandwidth even the 1d825, which does happen to have moderately better closed loop bandwidth that the opa627, but worse than opa637 at an average gain of 10.. I personally dont have a status unless that includes the quest for better sound..We are all here to teach eachother..
Colin
is that you can never discredit and must give serious look into the work that has been done before you, no matter which interest it has been done in "low distortion"ala self, "voicing" ala hugh dean whom i hold great respect for, and walt jung , Nelson Pass, Bob Cordell , John Curl along with so many others. As you point out it is probably true, some dont know what to make of datasheets , I was in this boat a few years back, I used to throw caution to the wind and buy the diy mods hyperbole, that is before i realized and well designed circuit is just that "well designed}" and the signal path makes the biggest difference. In the world of op amps, you really have it easy, here a company has spent many reasearch dollars to give you apllication notes, something that has been tested and tried to give the best average result with graphs to go along with it. 100% of the time when a manufacturer states bandwidth it is unity gain gbw only, not open loop, all op amps have poor open loop bandwidth even the 1d825, which does happen to have moderately better closed loop bandwidth that the opa627, but worse than opa637 at an average gain of 10.. I personally dont have a status unless that includes the quest for better sound
..We are all here to teach eachother..Colin
vynuhl.addict said:I hear you lineup, like yourself I am also on the path of discovery but what I have learned over the years since I was bitten by the audio bug
is that you can never discredit and must give serious look into the work that has been done before you, no matter which interest it has been done in "low distortion"ala self, "voicing" ala hugh dean whom i hold great respect for, and walt jung , Nelson Pass, Bob Cordell , John Curl along with so many others.
As you point out it is probably true, some dont know what to make of datasheets , I was in this boat a few years back, I used to throw caution to the wind and buy the diy mods hyperbole, that is before i realized and well designed circuit is just that "well designed}" and the signal path makes the biggest difference.
In the world of op amps, you really have it easy, here a company has spent many reasearch dollars to give you apllication notes, something that has been tested and tried to give the best average result with graphs to go along with it. 100% of the time when a manufacturer states bandwidth it is unity gain gbw only, not open loop, all op amps have poor open loop bandwidth even the 1d825, which does happen to have moderately better closed loop bandwidth that the opa627, but worse than opa637 at an average gain of 10..
I personally dont have a status unless that includes the quest for better sound
Colin
Dear Mr Colin,
I would like to say that I feel the same as you.
I can see you state all op amps have poor open loop bandwidth
What is your opinion about this fact:
1) is bad
2) is good
3) means nothing ?
This is my very doubt after all.
On the basis of what I have read I would say it is bad.
If the answer is 1) I would go for a discrete solution right away for my line buffer.
This is my question.
Thank you very much indeed.
Kind regards,
beppe
All three, its reality, in all circuits, not just op amps. With discreet circuits we still have to bring the open loop gain down with local and global feedback for better linearity and open loop bandwidth. To me it means a tool for determining which op amp to use in which gain block senario, unless you are building an intercom you will always need to use feedback to bring down olg to linearize over the desired 20-20khz spectrum and slightly beyond to keep phase anomalies as far out of the audibe spectrum as possible.beppe61 said:
Op amps are easy to implement and if you can read are almost brainlessly easy to use. designing a discreet you will still run into open loop gain issues which you will have to deal with with degeneration , compensation and feedback. To me discreets are the absolute king in high power ,current situations, but in line stage, and even some headphone scenarios you have a much broader choice of configurations and many worlds can work well if not stellar based on implementation..beppe61 said:
Colin
i think you might have a slight misunderstanding here. op amps are never used open loop for audio. they are usually used at much less than their open loop gain. using an op amp with an open loop bandwidth of 20khz or more isn't a bad idea, but it's not neccesary. in fact, with the slew rates of some of these RF/video op amps, you might actually end up with instability problems. power amps generally have a total voltage gain of between 10 and 100, and any op amp in the amplifier (input stage for instance) is only going to contribute a small portion of that gain. let's assume for instance an amp with a voltage gain of 30 (about right for a 100-150 watt amp). the whole amp is going to have a feedback loop with a 30:1 resistance ratio, and the VAS and input stage will both contribute to the gain. if the VAS has a gain of 10, then the op amp will only be operating with a gain of 3. if the GBW of the op amp is 10Mhz, then the bandwidth of the op amp will be 3.33Mhz, plenty more than enough for an audio amp, and actually a designer might need to limit the slew rate of the input stage to stabilize the amp. the open loop bandwidth is never an issue here since the op amp is never operating open loop, unless the phase shift through the rest of the amp reaches 180 degrees before the amp gain goes down to unity gain. if that happens, then the op amp could exhibit open loop gain at whatever frequency this condition exists at (since at this frequency there is no negative feedback), and you create a huge oscillator.
there are only very few uses for an op amp operating open loop at audio frequencies, the most common being a guitar fuzz box, and even with that particular application, the gain is usually limited to some manageable amount, since running at full open loop makes for a device prone to latching up, etc...
there are only very few uses for an op amp operating open loop at audio frequencies, the most common being a guitar fuzz box, and even with that particular application, the gain is usually limited to some manageable amount, since running at full open loop makes for a device prone to latching up, etc...
unclejed613 said:
Thank you very much Sir for your kind and valuable reply.
By the way my fundamental question remains: are opamps good for audio purpose ?
If yes, why are so often subject of discussion and debate ?
We should be all listening to op-amp preamps.
But from what I understand it is not the real situation.
The more technical question is then: why op-amps fail to convince everyone when they amplify a musical signal ?
Of course I know that op-amps are neve operated open loop.
Nevertheless there are more than one famous audio designer who thinks that the "open loop bandwidth" should be as wide as possible, maybe at least wide as the audio bandwidth.
And this is not the case with the vast majority of op-amps.
I believe that is an important drawback of opamps, of course when used for audio purpose.
In other applications this should not matter.
But I think we all have more interest in audio.
Anyway I will be pleased to know your opinion.
Thank you and kind regards,
beppe
Hi beppe,
Why don´t we listen all to tube amps or discrete amplifiers if they are so superior? Besides that, a lot of discrete circuits are of "opamp-design".
Operational amplifiers are versatile general building blocks, easy to implement and cheap.
Therefore they play a major role in professional and consumer audio. The "acoustical" performance seems good enough.
Regards
Jürgen
convincing everyone is an impossible mission.
Why don´t we listen all to tube amps or discrete amplifiers if they are so superior? Besides that, a lot of discrete circuits are of "opamp-design".
Operational amplifiers are versatile general building blocks, easy to implement and cheap.
Therefore they play a major role in professional and consumer audio. The "acoustical" performance seems good enough.
Regards
Jürgen
juergenk said:Hi beppe,
convincing everyone is an impossible mission.
Why don´t we listen all to tube amps or discrete amplifiers if they are so superior?
Besides that, a lot of discrete circuits are of "opamp-design".
Operational amplifiers are versatile general building blocks, easy to implement and cheap.
Therefore they play a major role in professional and consumer audio.
The "acoustical" performance seems good enough.
Regards
Jürgen
Dear Mr Jurgen, I understand.
Everyone has his/her own taste.
You say "a lot of discrete circuits are of "opamp-design"
Actually I was seeing the circuit of the old Mark Levinson JC2 line stage.
Maybe I am wrong but it seems to me that is indeed a "discrete op-amp" but with very few active components.
I am pretty sure that its open loop gain is very lower than that of a monolithic opamp and, I think by consequence, the open loop bandwidth much wider.
To end I do not think that a op-amp is a bad thing by itself.
For instance, those DIY discrete opamps proposed kindly by Mr Pass are extremely intriguing to me.
As are His projects that usually sport very few active components.
I will be experimenting in the next days (job permitting) with a line preamp prototype madeout of just two BC300 (I like its metallic can, is an aesthetic opinion of course).
In the first stage it amplifies and in the second is working as a buffer.
I am extremely curious about what kind of performance can be obtained by such minimalist circuits.
By the way I tend towards bjt devices after all.
Very cheap and convenient to use.
Thank you very much indeed.
Kind regards,
beppe
beppe61 said:
vynuhl.addict said:
Some recommend Op-Amps, like everybody
had unlimited cash
Other publish 'golden' projects, as if we all were made out Dollars.
We have many people in forum from 'poor' country.
We have people from: Africa, Asia and poor European countries.
USA have a lot, lot , lot of poor people, sleeping on street even.
These poor americans, are not much seen ON The NightShows in TV.
At best there might be a 10 min covering on homeless people,
in CBS 60 Minutes ... a program that try to be honest in reports.
* We also, I know have many young students as forum members.
They have no income .. maybe live on parents money or school-loanes.
They wont rush to buy 10 pieces of OPA637 ..this I can tell ..Not even my absolute FAVORITE: AD797
Because for one such OP
you can get 30, yes thirty NE5534 .. or more!
**************************************************
Besides .....
NE5534, a bit old, but
which really is a bargain, and very useful in many different circuits
and
OPA134, the King of useful JFET OP-Amps for reasonable Price
and
OPA227, OPA228
My personal favourites right now... there are others
AD825 like mentioned by the vinyl crazy man = addicted
will make good audio. Has been used in really HIFI circits.
Regards
lineup
There are a few op amp chips whose open loop bandwidth is greater than 20 KHz. E.g. the Linear Technology LT1360/1361/1362 or the National Semiconductor LM6171/6172. The National chips are cheaper and faster.
These are video op amps with a modest open loop gain (by op amp standards), very high gain-bandwidth product (abbreviated GBW) in excess of 50 MHz, and slew rates in excess of 1000 V/us. You must be very careful with such devices, because they will oscillate at radio frequencies if the circuit design or layout are wrong. Good quality power supply bypass capacitors are essential. You should have an oscilloscope with a bandwidth of at least 50 MHz to debug your circuit.
Such op amps tend to have high input bias currents. This means you must use low source and feedback impedances if you wish to DC couple them. They are also moderately noisy. The good news is that they will drive low impedance loads (> 500 ohms) with ease.
IIRC, the reason for the wide open loop bandwidth is to reduce frequency dependent phase distortion. But because the open loop gain at lower frequencies is much lower than most "audio" op amps, low frequency distortion will be higher. There's no free lunch.
Good luck!
These are video op amps with a modest open loop gain (by op amp standards), very high gain-bandwidth product (abbreviated GBW) in excess of 50 MHz, and slew rates in excess of 1000 V/us. You must be very careful with such devices, because they will oscillate at radio frequencies if the circuit design or layout are wrong. Good quality power supply bypass capacitors are essential. You should have an oscilloscope with a bandwidth of at least 50 MHz to debug your circuit.
Such op amps tend to have high input bias currents. This means you must use low source and feedback impedances if you wish to DC couple them. They are also moderately noisy. The good news is that they will drive low impedance loads (> 500 ohms) with ease.
IIRC, the reason for the wide open loop bandwidth is to reduce frequency dependent phase distortion. But because the open loop gain at lower frequencies is much lower than most "audio" op amps, low frequency distortion will be higher. There's no free lunch.
Good luck!
lineup said:
Some recommend Op-Amps, like everybody
had unlimited cash
Other publish 'golden' projects, as if we all were made out Dollars.
We have many people in forum from 'poor' country.
We have people from: Africa, Asia and poor European countries.
USA have a lot, lot , lot of poor people, sleeping on street even.
These poor americans, are not much seen ON The NightShows in TV.
At best there might be a 10 min covering on homeless people,
in CBS 60 Minutes ... a program that try to be honest in reports.
* We also, I know have many young students as forum members.
They have no income .. maybe live on parents money or school-loanes.
Not even my absolute FAVORITE: AD797
Because for one such OP
you can get 30, yes thirty NE5534 .. or more!
**************************************************
Besides .....
NE5534, a bit old, but
which really is a bargain, and very useful in many different circuits
and
OPA134, the King of useful JFET OP-Amps for reasonable Price
and
OPA227, OPA228
... there are others
AD825 like mentioned by the vinyl crazy man = addicted
will make good audio. Has been used in really HIFI circits.
Regards
lineup
You know, I have to admit I sometimes fail to think about these technicalities when reccomending op amps. Also I was fortunate enough to have obtained some samples from TI to play with and test in my circuits a couple years back. Students do have access to samples and are very easy to get if you are a student. But at 18 dollars a pop retail it is ridiculous, and to pay real world prices may not be worth it in the long run. Opa2134 is excellent for a low price chip, same with AD823(in my opinion better for I/V stage in cdplayers than AD825). I have not played around so much with the NE55xx series.
Colin
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.