Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Discrete vs opamps...
New Posts  All Forums:Forum Nav:

Discrete vs opamps... - Page 11

post #151 of 176
I think I should add that both single-ended and push-pull designs can both be class-A biased. I personally believe that it is the single-ended, class-A designs that really deliver the ultimate high-end eargasm.
post #152 of 176
Quote:
Originally Posted by Tyll Hertsens
I still don't think it's as easy as saying a descrete design is always better. At this level, personal listening preference is very important
Good enough for me.
post #153 of 176
Quote:
Lastly, I want to say that Nelson Pass is one of my personal design heros. In understanding his designs you'll find that simply saying class A is the only way to go only presents a very small part of the picture.
I have about five Nelson Pass designed amps banging around the house but they are winter amps only ! The basic "zens" are about as inefficient as amps can get and it takes the same amount of heat sinking to cool 10 watts (mine is actually 8 ) as a conventionally biased 200W amplifier.All heat

I have another that uses the Aleph biasing scheme that has less demand on the heat sinking because of the way the bias "slides" from the main N-CH main mosfet to the P-CH active dynamic current source part.Less heat,pretty cool topology but still a mini-heater.

Quote:
I think I should add that both single-ended and push-pull designs can both be class-A biased. I personally believe that it is the single-ended, class-A designs that really deliver the ultimate high-end eargasm.
I agree.Unfortunately not suited to all situations and in my case in the loudspeaker arena I have to move up to PPT because the dynamics are just too blunted with my curent speaker setup.I love the sound of my 300B SET amp and beleive it has the most musically satisfying performance I have ever heard in my home...just not enough of it so I must choose a lesser (? Going to 12B4 IT PP) amp to satisfy my power demands.With headphone amps yes.Mine are now all triode amps,single ended,all class-A gain stage/follower output stage amps

And this is where it gets back to topic.I am not the one that took this thread all over the map but in trying to explain to closed minds why some statements were just not true or so off base as to be misinformation the effort to bring things back in line took it to the confusion level.

the original answer by me on the actual topic before it strayed and became whatever this is was that it was my personal belief that pure class-A operation done right is superior to class A/B and since all opamps are in fact class-A/B on the output after a certain point.
This brought in the "well what about the class-A biased opamps ?" to which I explained that they too were only class-A up to a point due to physical limitations built in to the package and substrate and that after that point the output needs to be "handed off" for the rest of the power curve.

This brought every conceivable whacko method of trying to claim an opamp either IS or COULD BE a true class-A driving amp for headphones from persons who refuse to ever read any of the volumes of available information and to which once they get hold of a topic and convince themselves they are right,even if the evidence says otherwise,continue to demand reasons why they are wrong so they can then say "I beleive it is...." without one shred of any evidence to back up a single claim made.Not even a minimal head-to-head listening test which in the end is the ultimiate and only valid test.

I could stray the topic even more and go into excrutiating detail ( ) on why I beleive class-A operation to be the preferred method whenever and where-ever possible but I will spare everyone that at least but there is a serious amount of mis-statements and misrepresentations going on here and it is not a new thing.

I never knocked a single topology and not a single amp known to humans as is my usual policy but only stated what I preferred and this lead to me being challenged to put up or shut up.
Well folks I put up and proved 99% of my points while not a whole lot has come from the other side other than "maybe we culd get Burr-brown to make us a special opamp so rick can be wrong"

I should know better than to try and have a discussion on the merits on anything even remotely technical here at head-fi,I damn sure have had enough crap for previous attempts but this is one of those deals where sometmes I just need to be beat over the head with the information before it sinks but sink it has and it only took nine pages........................

cheers all

Rick is outta this thread for good
post #154 of 176
Time for me to add my 2 cents, or in this case $440.

There are in fact monolithic pure class A amplifiers out there.
Some of them in fact work great for headphones.

Here is one of them
http://gilmore.chem.northwestern.edu/amonolithic.jpg

Runs on +/-15 volts, 850ma current consumption. DC to 400mhz
Works down to 10 ohms, no sweat. Now rick may claim a foul ball
but this meets every requirement for a monolithic pure class A
opamp. Cheap too at $220 each. If you can still get them.

Fact is that there is only 2 standard manufacturing lines for making
monolithic circuits. One that makes vertical npn's and lateral pnp's
and the other that makes it the other way around. Unless you are going
to do silicon on saphire or GaAs you cannot ever have matched pnp
and npn's on the same substrate.



quoting tyll
"For example, we (at HeadRoom) simply can't do a descrete design and still have a crossfeed circuit, so if you want crossfeed you'll have to live with some percentage of the electronics being monolithic."

I now completely disagree with this statement. I offer up the following
picture as proof that an all discrete module the same exact size as the
headroom module with crosfeed can in fact be built. In fact i've put
it on my todo list.

http://gilmore.chem.northwestern.edu/g9f22318.jpg (warning huge)
blow up to full size the upper right corner.

Whether or not tyll's in house people can assemble such a device
economically is an entirely different matter.
post #155 of 176
Yeah, OK, kevin you're right: it is "possible" to build a crossfeed amp with dicretes, but the number of componants and matching requirement would make it economically unviable as a product. Nice pix though! What is it?
post #156 of 176
i only read half the thread and i barely understand anything you guys are talking about, but my understanding is that an 100% class A integrated op-amp capable of driving 'normal' headphones would have to dissipate up to 10W continiously from a surface as small as your pinky.

i can't help but to compare this to cooling a CPU, your Pentium 4, your AMD Athlon or whatever your running. some of these have thermal densities (watts dissipated per square inch) that sound awefully close to the 10W per pinky, yet they stay somewhat cool thanks to big copper coolers with fans on them. some are even watercooled

now, assuming that the cooling capacity needed to keep said class A opamp cool exists, would it be far fetched to build a class-A opamp-based headphone amplifier?
post #157 of 176
Quote:
Originally Posted by Firestarter
i can't help but to compare this to cooling a CPU, your Pentium 4, your AMD Athlon or whatever your running. some of these have thermal densities (watts dissipated per square inch) that sound awefully close to the 10W per pinky, yet they stay somewhat cool thanks to big copper coolers with fans on them. some are even watercooled
This was discussed. Long story short, the CPUs are mounted with conductive material all the way to the core for cooling purposes while the opamps do not. The industry is moving towards lower power designs so heat sinking really isn't in the cards.

Oh Yeah, You're welcome Rick, I thought I'd let you sit this one out...
post #158 of 176
Quote:
There are in fact monolithic pure class A amplifiers out there.
Some of them in fact work great for headphones.
cool part but is it an opamp or an actual "amp" as in powewr stage ?

That package screams power and has the required contact area for heat sinking but looking at the power taps I don't think high speed op-amp and those puppies belong in the same sentence without the oscillation bugaboo creeping in.Just too much "metal" on the differential input if but just conjecture having no clue what that part is.

The closest to a class-a monolithic chip I have is the LH-0063 and that took some work just to get it into the 80ma range and that is a buffer,not opamp wich means feedback amp.

Quote:
I now completely disagree with this statement. I offer up the following
picture as proof that an all discrete module the same exact size as the
headroom module with crosfeed can in fact be built. In fact i've put
it on my todo list.
not wanting to pick a fight with Tyll I let that one go but since it has been brought up.....

Anything an opamp can do you can do with discretes.Many times better and for sure with less actual circuit complexity if that is what is desired.
KG likes a DC coupled path so his designs go to PP/balanced but even a simple single-device single-stage class-A JFET source-follower/mixer can perform the duties of adding a crossfeed in a pinch.It will have a big blocking cap on the output but doable just the same.

Folks can choose whatever they want and it means squat but these "what is the best" type questions as a topic for discussion at Head-fi lead to crap 100% of the time and anyone with a brain knows there is only personal best anyway so why bother ?

better yet i wonder still why i did
post #159 of 176
Just my tuppence happeny worth but a well designed discrete can sound much better than a poorly designed opamp based amp and, vice versa, a well designed opamp based amp can sound a truckload better than a poorly designed discrete type.

Don't automatically assume that "every" amp is well designed and sounds great, some of them are designed by gibbons and sound like turd, it all comes down to design at the end of the day and just because it's "discrete" means jack **** if it's been slung together rather than "designed"

So, the answer to the original question is, it all depends on the design. Use your ears to judge.

My tuppence happeny worth.
post #160 of 176
rickcr42>
I have another that uses the Aleph biasing scheme that has less demand on the heat sinking because of the way the bias "slides" from the main N-CH main mosfet to the P-CH active dynamic current source part. Less heat, pretty cool topology but still a mini-heater.

Ah I see you also have found the virtues of dynamic biasing. Yes Mr. Pass is the man shall we now call Him Dr. Pass (Just kidding)

Tyll Hertsens>
I think I should add that both single-ended and push-pull designs can both be class-A biased. I personally believe that it is the single-ended, class-A designs that really deliver the ultimate high-end eargasm.

Not my cup of tea however Good luck with that

Tyll Hertsens>
I read about five pages of this thread and then just gave up and jumped to the end because I thought I'd like to comment on an aspect of this discussion. So please excuse me if I cover something that was said late in the thread:

Understandably so Me also

Tyll Hertsens>
So let me sum up what I think the novice might want to know about descrete vs. monolitic designs:

1) In a perfect world, descrete designs can achieve higher levels of performance.

2) In the real world, IMHO and at a given price point, with low to medium cost headphone amps (up to, say, $800), good op-amp designs are just as likely to deliver the sonic goods as good descrete designs, and are more likely to be reliable.

3) In cost no object products, I still don't think it's as easy as saying a descrete design is always better. At this level, personal listening preference is very important, so your pick should be based more on your ears than what the design topology is. For example, we (at HeadRoom) simply can't do a descrete design and still have a crossfeed circuit, so if you want crossfeed you'll have to live with some percentage of the electronics being monolithic. Also the tube vs. transistor question rears its head.

Lastly, I want to say that Nelson Pass is one of my personal design heros. In understanding his designs you'll find that simply saying class A is the only way to go only presents a very small part of the picture. The devil is in the details with descrete designs, and a lot of homework (in terms of listening tests and confirming reliability) is needed for someone to make a long-lived, satisfactory product selection.

I agree with all the above and also would like to remind you of my position as if you haven’t already forgot! I think any op-amp IC based design is to be divided into a voltage gain and a power stage so as to keep thermal distortion down, such as an OP Amp Buffer arrangement. I completely agree with tyll that under $1000 USD that an IC designs is the best bang for the Buck and above that a fully discreet design is the best. The matching and set up are just one of the Reasons a fully discrete design is more expensive. As for how thay hold up just look at all thouse 1970’s receivers still working after 30 years and most had discrete component amps and IC tuners and Preamp so I am not one to believe an IC is going to be more service free than an all transistor design However monolithic output stages leave a lot to be desired and even at a low price point a good old Diamond buffer driven from an op-amp is about as good as it gets at the under 1k price point. There are about only two IC buffers in wide use in consumer electronics the BUF-634 and the HA-5002 family, this includes HA-5002 clones like the HA5033 and OPA-633. Same circuit different specs for the HA-5002,HA-5033 and OPA633.

The primary determination upon the sound of an Audio component is the Designer. It takes quite a bit of talent to get any high performance circuit working properly regardless of what generic active devices are used. I have seen both great and not so great versions of IC use, Transistor implementation, and Tube abuse as well as tube’s used in proper.

Since we all love Nelson pass I would highly suggest you his article on DIY op-amps. You may want to book mark this link as the Pass site no longer hosts this article why not a mystery to me is. http://cygnus.ipal.org/mirror/www.pa...m/diyopamp.htm
post #161 of 176
hate to say "told you so" but..........OK I don't hate to say it."Told ya"

Quote:
Ah I see you also have found the virtues of dynamic biasing. Yes Mr. Pass is the man shall we now call Him Dr. Pass (Just kidding)
Nelson Pass IS the man but dynamic biasing is more means to an end than it is a better way on the sonic merits.i will take my "Zen Lite" over my "Aleph 3" knockoff hands down on the sonic merits.
The Zen series is mostly not a viably commercial topology being as inefficient and non PC as it gets (HUGE DC blocking cap) unless you go balanced and even DIY versions get pricey so consider how expensive an "commercially viable" Zen would be that satisfied not only actual listening tests but conventional "I want my PC audio" and pretty graphs as well ?

DIYers can get away with selling specialised gear,major manufacturers need to produce products that are plug an play hence the Aleph Series
post #162 of 176
ppl : the article on diy opamps is still on passdiy, just on the "legacy" page : http://www.passdiy.com/legacy.htm
post #163 of 176

Opamp tuners

I have a MAC 67 tube tuner, an Accuphase T-100 non-opamp tuner and a modified Kenwood 7500 with a low DC offset BB 2604 opamp. They all sound world class. The Kenwood has deep bass(but not as warm and the Mac 67).

The Kenwood with BB2604 has the best focus of all three. It also has very good dynamic range for an opamp audio stage which I previosly thought was only a strength with discrete outputs.
post #164 of 176
Quote:
The Kenwood with BB2604 has the best focus of all three. It also has very good dynamic range for an opamp audio stage which I previosly thought was only a strength with discrete outputs.
the discussion is on driving headphones,not an interconnect cable which has entirely different requirements.Also never said opamps can't sound good,just which I beleive top be a better way
post #165 of 176
Quote:
Originally Posted by rickcr42
hate to say "told you so" but..........OK I don't hate to say it."Told ya"



Nelson Pass IS the man but dynamic biasing is more means to an end than it is a better way on the sonic merits.i will take my "Zen Lite" over my "Aleph 3" knockoff hands down on the sonic merits.
The Zen series is mostly not a viably commercial topology being as inefficient and non PC as it gets (HUGE DC blocking cap) unless you go balanced and even DIY versions get pricey so consider how expensive an "commercially viable" Zen would be that satisfied not only actual listening tests but conventional "I want my PC audio" and pretty graphs as well ?

DIYers can get away with selling specialised gear,major manufacturers need to produce products that are plug an play hence the Aleph Series
I have used the same output stage of my own design with Dynamic Biasing and another class A and thats 3 amps ICq for a headphone amp is that class A enought for ya, Moreover if your output current is greator than the Icq then you have a class B output stage will all its Drawbacks like the non conducting transistor becoming rev. Bised where in Dynamic Biasing it never dose evean with a serious overload. plus High heat, However like you say its farly hard to mess up a class A design IMHO Class A is a one trick pony that is often highly inflated in ratings by consumer marketing. BTW thanks for the link 00940
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Headphone Amps (full-size)
Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Discrete vs opamps...