OPA-Earth & OPA-Sun, Discrete Opamp diy kits

[lineup]

What do you people think of
OPA-Earth & OPA-Sun discrete opamps
by member audio-gd

The test results published looks alright.
Are they anything to consider, in place of Audio IC Opamps.
For example OPA2134, AD797, LM4562, AD827 and LME49710, LME49870.
Schematics, Info & Prices. Link.
http://www.audio-gd.com/enweb/pro/diy/OPA.htm

We have some very good alternative Discrete Op-Amps suggestions
by The Master Nelson Pass here:
http://www.passdiy.com/pdf/diyopamp.pdf

I have too, published a number of similar discrete Class A OpAmps.
For voltages between +-15 to +-50 Volt ( 30-100 Volt total Supply).


Lineup - Discrete Op-Amp projects

... ...

 

[KSTR]

What I don't like about them (Sun & Earth) is that one of them seems to be 1:1 copy of another commercial discrete opamp replacement, the Burson amp (or vice versa, which appears to be rather unlikely, but I just don't know). There is a thread about the Burson amp here where schematics are shown (and some measurement), from a french forum.
http://www.diyaudio.com/forums/showthread.php?threadid=127559
(see first dozen of replies).

- Klaus

 

[scott wurcer]

Op-amps don't have a ground pin 😱.

At least one of those schematics has a "ground" connection that is definately not optional.

 

[Franz Gysi]

Hello

Last weekend, another diyaudio member told me the first time about this opamps.

I was really impressed.

But now, knowing this are some copies, makes me unhappy

Next weekend, the diyaudio member will visit me, bringing in some of this opamps.

And some of the mysterious DY2000 (AD823 in another package? Really an Analog Device Product?).

I have the opportunity, to use a professional audio analyzer, the PrismSound DScope III http://www.prismsound.com/test_measure/products_subs/dscope/dscope_home.php

And I have the intention, to do some tests, in comparison with another good opamps.

I am really open for every suggestion, what measurement in wich configuration (gain, input impedance, inverting/noninverting) would be of interest.

Please post your suggestions/wishlist here.

And I will post the results.

Kind regards
Franz

 

[analog_sa]

I am really open for every suggestion, what measurement in wich configuration (gain, input impedance, inverting/noninverting) would be of interest.

How 'bout just listening? None of the known measurable parameters bear any relation to perceived sound. Which you probably know anyway.

 

[syn08]

None of the known measurable parameters bear any relation to perceived sound.

Yeah, right 😀
The grapes are sour anyway! 😀

 

[KSTR]

None of the known measurable parameters bear any relation to perceived sound.

Probably you should make yourself familiar with the work of Earl Geddes in this field, and especially make sure to read this paper by de Santis / Henin to update your position:
http://projekter.aau.dk/projekter/fbspretrieve/9852082/07gr1061_Thesis.pdf

- Klaus

 

[KSTR]

I am really open for every suggestion, what measurement in wich configuration (gain, input impedance, inverting/noninverting) would be of interest.

Maybe the recent excellent work of your fellow countryman Samuel Groner is worth to look at in this context:
http://www.sg-acoustics.ch/analogue_audio/ic_opamps/pdf/opamp_distortion.pdf
Local thread here:
http://www.diyaudio.com/forums/showthread.php?threadid=129202

- Klaus

 

[analog_sa]

Probably you should make yourself familiar with the [/url]

Dear Klaus

Having been interested in audio for a very long time i am certainly aware of your references. And of other similarly bold claims. Pardon my scepticism but none of this is very conclusive. Opamps are relatively easy to test in a bypass test by ear. These specific products may be very nice but their execution and specifically wiring harness is very suspect.

 

[Gordy]

I am really open for every suggestion, what measurement in wich configuration (gain, input impedance, inverting/noninverting) would be of interest.

Please post your suggestions/wishlist here.

And I will post the results.

Kind regards
Franz

My suggestion:

Driving impedance 1... 600 Ohms (to represent typical output Z of a previous stage)

Driving impedance 2... 5000 Ohms (to represent typical output Z of a volume potentiometer)

Load impedance... 10000 Ohms in parallel with 300 pF (to represent typical input Z of an amplifier, plus cable capacitance)

Input level... variable amplitude sine, from 775 mV to 2000 mV (RMS)

Test for...
...harmonic distortion spectrum at 20 Hz, 100 Hz, 1000 Hz and 10000 Hz
...how the distortion spectrum changes with changes in input amplitude
...intermodulation of twin tones (but I am not sure of the frequencies? Many use 19 and 20 kHz)
...phase response (especially of the higher frequencies)
...noise (I think it is best to test this with the input shorted?)
...cross-over distortion (because even just a little seems to make a difference to the sound quality)

Good luck.

 

[Wavebourn]

I would like to see an open loop gain and phase shift VS frequency curves.

 

[Bonsai]

I can buy the fact that going 'discrete' can allow a designer optimize a circuit for specific performance characteristics that cannot be accomplished by integrated designs e.g. PMA mentioned he likes to run his input and amplifier stages at higher currents than in integrated circuit designs. You can also make a really rugged output stage.

What I dont buy is this crap about discrete designs giveing more 'air' and 'fluidity' to the sound. This is crap that's been dug up from audiophile magazines by reviewers that know squat about electronics engineering and then perpetrated by people who refuse to acknowledge the gains made in IC op-amps over the last 20 years - or worse still, they claim that people who use oip-amps don't know how to design 'proper' audio circuits. There are a few notable exceptions in the reviewer fraternity (e.g. JA), but they are out numbered unfortunately. Just like I've saif before, and someone repeated early, the signal on the average recording has already been through about 10 or 20 op-amp stages - the last stage, whether IC or discrete won't make a jot of difference.

 

[lineup]

Yeah, Bonsai.
I can not do anything but agree with you.

Why I like to design/build my discrete opamps
is much beause of the fun of it.
The fun of being creative.

I do not for one moment fool myself to think,
that my discrete designs will outperform even one simple OPA134.
Hardly even in one single parameter

Now custom designs, can be made WITH opamps, too.

Like high output curent, high voltage circuits, lowest noise.
Minimal offset, minimal bias current, precision.
Ultra high speed, slewrate, gainbandwidth.
For transmitting digital signals in the GigaHertz.

Because there are great opamps for anything!
Made to be Specialized.
-------------------------------------------------------------------------------------

All Opamps manufacturers has got some Design Selection guides.
Here is the Operational Amplifiers Selector Wizard at www.national.com
http://www.national.com/analog/amplifiers
See my Attachment Screenshot!

Lineup Opamps Advice for Free

 

[QSerraTico_Tico]

Dy2000

Hello

Last weekend, another diyaudio member told me the first time about this opamps.

I was really impressed.

But now, knowing this are some copies, makes me unhappy

Next weekend, the diyaudio member will visit me, bringing in some of this opamps.

And some of the mysterious DY2000 (AD823 in another package? Really an Analog Device Product?).

I have the opportunity, to use a professional audio analyzer, the PrismSound DScope III http://www.prismsound.com/test_measure/products_subs/dscope/dscope_home.php

And I have the intention, to do some tests, in comparison with another good opamps.

I am really open for every suggestion, what measurement in wich configuration (gain, input impedance, inverting/noninverting) would be of interest.

Please post your suggestions/wishlist here.

And I will post the results.

Kind regards
Franz

Hi, I think that DY2000 is a rebadged part.
The AD logo is reproduced poorly and an other part# is still vaguely visible.
The date code suggests week 33 of the year 1997, pretty old.
Absolutely nothing to find on the AD site
www.analog.com
I am not surprised as the Zhaolu with this chip comes from.........China.

 

[Bonsai]

I agree discrete is fun . . . that's why I'm also making a discrete based general purpose audio 'OPA'. But, I dont think National or ADI will need to worry too much!

 

[SY]

Square waves are fine, but what happens with reactive loading and what does the distortion spectrum look like? Do they even have the capability to measure that? I'll be interested in Franz's results.

Yes, if they're just copies of someone else's design without any development or verification, there's a question of ethics, too.

 

[Franz Gysi]

Maybe the recent excellent work of your fellow countryman Samuel Groner is worth to look at in this context:

Very interesting! Thanks for the input.

I will try to do some comparable measurements to the results from Samuel Groner. Maybe, I will include one of the chips he also tested, for comparison.

Square waves are fine, but what happens with reactive loading and what does the distortion spectrum look like? Do they even have the capability to measure that? I'll be interested in Franz's results.

What is reactive loading? Unlinear like a speaker?

Yes, if they're just copies of someone else's design without any development or verification, there's a question of ethics, too.

I don't know, who is copying from who! When I look in web.archive.org, I cannot say who was the first, Burson or Audio-gd.

But: audio-gd seems to be more innovative, as he announced the OPA-Moon, Class-A output! BTW: I ordered some moon's and will measure it later.

Why are there no audio opamps with Class-A output by design from the big semiconductor manufacturers available?

Thermal issues?

Franz

 

[nelsonvandal]

the signal on the average recording has already been through about 10 or 20 op-amp stages - the last stage, whether IC or discrete won't make a jot of difference.

I think you know, like all of us do, that amplifiers do sound different. Every single transistor has it's own sound, and some claim cables, PCB's and resistors has a sound of their own.

I think it has to do with monotony - some call it listening fatigue - when you hear the same coloration over and over it's bugging you.

When listening to different recordings, all with their own sound, it doesn't matter how many synths, opamps, samplers or effect racks the sound has passed through the recording process. As long as every record has it's own sound, those colorations aren't annoying in the same way as coloration produced by your HiFi equipment, and opamps in HiFi equipment do color the sound. If monolithic opamps are better or worse than discrete ones is another story.

 

[nelsonvandal]

Why are there no audio opamps with Class-A output by design from the big semiconductor manufacturers available?

Franz

Aren't most of them class A up to a certain load? It has been popular to force monolithic opamps to class A operation by applying a resistor or active current source from output to negative (or sometimes positive) rail. I thought the world of doing this until I did this tweak live, and I heard no difference at all. Placebo is a miracle drug.

 

[Franz Gysi]

... reading some pages out of the paper from Samuel Groner, I am realizing, such measurements are big work.

On page 209, the chapter "Personal Conclusions" he is asking three questions.

One of them:

How does the distortion performance of discrete opamps compare with good IC amplifiers?

Later:

This question must be unanswered for the time being as not enough discrete amplifiers have been tested yet

I try to get in contact with him...

Franz