OPA627
I di search with google using words:
opa627ap opa627bp
This guy says he couldn't hear diff between ap and bp
also about OPA2134 and other OPamps:
http://tangentsoft.net/audio/opamps.html
Here you can download OPA627 PDF datasheet
and read about the differnces yourself:
http://www-s.ti.com/sc/ds/opa627.pdf
It is just a matter of KnowHow - how to search! with google
/halo -the great searcher!
I di search with google using words:
opa627ap opa627bp
This guy says he couldn't hear diff between ap and bp
also about OPA2134 and other OPamps:
http://tangentsoft.net/audio/opamps.html
Here you can download OPA627 PDF datasheet
and read about the differnces yourself:
http://www-s.ti.com/sc/ds/opa627.pdf
It is just a matter of KnowHow - how to search! with google
/halo -the great searcher!
I swapped out all eight 'A' grade OPA 627's with 'B' grades (about a year ago) in my well-modded Assemblage 'Platinum' DAC 3.1, and was pleased with the result.
It was one of the most expensive 'single' mods I have ever done at $160 US plus carriage to UK and Customs etc., from Digikey.
I wouldn't wish to go back now to the earlier ones, but at this cost, it is hard to say it was necessarily good value as far as a particular mod goes.
However, this DAC is excellent (and, in my opinion, justifies the extravagence) and I have done almost everything else I know of to 'gild the lily' as it were, and I was after the last possible fraction of performance from it.
It was one of the most expensive 'single' mods I have ever done at $160 US plus carriage to UK and Customs etc., from Digikey.
I wouldn't wish to go back now to the earlier ones, but at this cost, it is hard to say it was necessarily good value as far as a particular mod goes.
However, this DAC is excellent (and, in my opinion, justifies the extravagence) and I have done almost everything else I know of to 'gild the lily' as it were, and I was after the last possible fraction of performance from it.
I wonder what the benefits really were B vs A type?
B type has slightly better DC-specs (which are rather unimportant in the audio band. It's only important at DC.
The noise is lower but was with the A type lower than the digital noise (98 dB S/N), unimportant
The open-loop gain was slightly higher in a application with heavy feedback, unimportant
I can't see that changing from A to B would really improve the sound qualilty....and....be worth the money. Was it for free I would also do this tweak.
B type has slightly better DC-specs (which are rather unimportant in the audio band. It's only important at DC.
The noise is lower but was with the A type lower than the digital noise (98 dB S/N), unimportant
The open-loop gain was slightly higher in a application with heavy feedback, unimportant
I can't see that changing from A to B would really improve the sound qualilty....and....be worth the money. Was it for free I would also do this tweak.
CraigBuckingham said:
You are entitled to your opinion, but I have done the job, and there *is* an improvement which I am pleased with, as I said.
As I also honestly reported, I am not so sure that for many people it would be considered a cost-effective mod., especially if the rest of their system was not of extremely high quality.
That kind of outlay would probably be better spent elsewhere.
There have been many other mods I have caried out to my systems over the years (some of them being extremely time-consuming and also costly) which have been a complete waste of time, but luckily, this was not one of them, and it was a five minute 'plug in' job.
I won't bother you with the perceived results from replacing the four PCM1704 DACs with 'K' grade versions in the same DAC which was nearly as expensive, but which is a nightmare to carry out as they are 20 lead *small* SOICs, as you seem to know better, anyway!
Regards,
CraigBuckingham said:
I have added this class A thing into my RIAA amp and my headphone amp and I can't hear a thing, maybe it's possible to measure a difference.
I gather that the greatest benefit is reached when the load (current) is slightly above the internal current of the op's outputstage. For small signals this tweak doesn't harm but how much good does it make? Don't know. The distortion for the OPA627 is 0.000008%, is it neccessary with lower values...?
I have used both the OPA-627AP and OPA-627BP grades in a Headphone Amp with a closed loop gain of 20 dB and the AP grades produced DC offset voltages on the Output from less than 1 Mv to over 10 MV by going to the BP grade this DC voltage on the output went down to Less than 1 Mv and typicaly 0.002 MV consistent on 12 different devices. so if this is of importence to you and your Circuit is Direct coupled then to save having to hand select the Op Amp i would use the BP grade.
For those who are interested I have a great article on class A biasing of audio chips from Audio Electronic some years ago.
The article refers to the merits of various chips according to application, however the glory goes to the Borbley Fet line stages for sound qulaity over all other options.... understandably so.
The advantages of chip biasing into class A are there for mainly pro audio where you are really pumping towards 20+ dba input 600 ohms, such as a mixing desk.
Into relatively high impediances the noise/disrtion cancelled mechanisms in these advanced chips is very good.
However the gross feedback appliedk for unity gain of these devices is what kills the music compared to a nice simple discrete fte buffer aka Pass.
regards
ian
The article refers to the merits of various chips according to application, however the glory goes to the Borbley Fet line stages for sound qulaity over all other options.... understandably so.
The advantages of chip biasing into class A are there for mainly pro audio where you are really pumping towards 20+ dba input 600 ohms, such as a mixing desk.
Into relatively high impediances the noise/disrtion cancelled mechanisms in these advanced chips is very good.
However the gross feedback appliedk for unity gain of these devices is what kills the music compared to a nice simple discrete fte buffer aka Pass.
regards
ian
Wow!
Wow - I thought most op-amps were limited to much lower supply V's than this, must have a multiplier stage built-in
Tee hee!
Andy.
P.S. I've also found that with good modern op-amps (e.g. AD86x0) the class A thing seems to degrade rather than improve. It works better with a current source than a resistor, but doesn't make things sound better in the way I want them to.
Wow - I thought most op-amps were limited to much lower supply V's than this, must have a multiplier stage built-in
Tee hee!
Andy.
P.S. I've also found that with good modern op-amps (e.g. AD86x0) the class A thing seems to degrade rather than improve. It works better with a current source than a resistor, but doesn't make things sound better in the way I want them to.
It is standard to write mV for milliVolt instead of MV which means MegaVolt. Same for mA for milliAmpere etc.
Concerning the subject: I never remarked any sonical difference between AP and BP versions of the OPA627. Only offset voltages differ IMO. Nevertheless I prefer to use BP's just in case.
CraigBuckingham said:
Hi Craig,
I was not being "sensitive", but, with respect, the matter is defintely *not* "that simple" as you suggest. Had it been, I wouldn't have made any further comment, nor would I have wasted further time in writing this.
Having attempted to help Costiss by relating some relevant *factual* experiences of my own, your views virtually directly contradicted what I said, without apparently you even having tried this out for yourself.
If you had made any such comparisons, why didn't you say so, together with the location you had tried them in - which would be relevant, rather than merely expressing an opinion?
How helpful (as opposed to being misleading) to Costiss in reaching a conclusion as to what to do in his case, is that?
Also, you then went on to categorically state (with apparent authority, which might similarly have mislead Costiss) that operating these devices in class 'A' was the way to go, possibly because you had read about it somewhere else!
The reason I am guessing this is, if like me you had already tried that mod with the 'A' grade OPAs for yourself, you would have discovered (like me, and Peranders & Wombat, it appears!) that this particular mod is certainly not universally beneficial, if it is any improvement at all!
I have no wish to cross swords with you, and my only motive is to *help* Costiss, which gave rise to my original caveat over costs etc.
It was also for a similar reason that I didn't comment on the AD825 he asked about, which I have also used (and like), but as I have not directly A/B'd it with an OPA, any guesses on my part would not have been of any real value to him.
Regards,
Ok Now that we have the Mv and mV thing over. My thoughts on Current sourcing an Opamps output stage to obtain class A Bias. A resistor as the Current source or current sink for those Critics. i tride in the late 1970's and while it smoothed the sound out in the opamps i tried them with LM-353,LM-356,NE-5532, and a host of others. The resistor in addition striped all the Dynamics out and in some devices created DC offset voltages that were Higher than Without the resistor this was apperently because of using too much Current so i backed down the Current from 10 mA to 4. this improved the DC offset. The Circuits this method was used on rainged from Line Level gain stages to RIAA phono preamps to I/V converters. Since the resistor was giving what i considered less than perfect results i went to an Active current source Via the CRD diodes these are a J-fet and resistor inside a Glass diode case. and range in Current from about 0.33 mA to about 5 mA. These devices improved over the resistor method alot.These Current reg Diodes can get spendy so i next used a jfet and an external resistor that cost less and did the same thing. the only apperent advantage in the CRD was the High Voltage Complience of about 100 Volts vs the 20-30 Volts of most J-Fets. With Opamps this is not a issue.
These were my findings in the Late 1970's to the Mid 1980's after that Opamps were introduced that were IMHO accurate enough to reveal the Limitations of just a Current source while these newer Op Amps still smoothed out with a CRD, the CRD seemed to strip away the Detail. Investigation and conversations with others showed that placing a small resistor between the Current source and its load in this case an Op Amps output. would Buffer the Non linear capacitence of the Current source and this did improve the sound. I later used a jfet cascode current source and an isolating resistor. this is the only circuit i have used so far that smoothes out the Sound while not also degrading other as pects of the Performance. so far this method has improved all Opamp circuits i have used it on and with all Opamps including the New AD-8610. I should add that most of the Opamp circuits i use the method on are buffered using an Active open loop Buffer IC Like the BB BUF-634 and other's.
These were my findings in the Late 1970's to the Mid 1980's after that Opamps were introduced that were IMHO accurate enough to reveal the Limitations of just a Current source while these newer Op Amps still smoothed out with a CRD, the CRD seemed to strip away the Detail. Investigation and conversations with others showed that placing a small resistor between the Current source and its load in this case an Op Amps output. would Buffer the Non linear capacitence of the Current source and this did improve the sound. I later used a jfet cascode current source and an isolating resistor. this is the only circuit i have used so far that smoothes out the Sound while not also degrading other as pects of the Performance. so far this method has improved all Opamp circuits i have used it on and with all Opamps including the New AD-8610. I should add that most of the Opamp circuits i use the method on are buffered using an Active open loop Buffer IC Like the BB BUF-634 and other's.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.