Quote:
TPA6120 is a ADLS driver repacked as a headphone amp. High current capability is often positive and means that you use only a part of it's maximum capacity => low distorion. 1.1 A, 15 volt peak at 1 MHz, not bad! |
It looks good on the surface until you actually pour through the detailed graphs.again not the whole story so it comes down to the listening which I did and for my personal taste,and it is only that guys,was not a "keeper".
Most of the ADSL chips look like seriously good audio drivers and when i ordered the ADI AD8000 driver was ready to be wowed but again was totally dissapointed in the final results.
there may well be a keeper in that bunch that someone has tried and is keeping a secret but so far i am batting 000 on it
the part about "audio chips" you are right.Most are designed as a general purpose "building block" which can be used where the specs say it is a good fit but the ones that move over to audio use are usually because some one some where took the time to do the heavy lifting and weed through all the "pretenders" to find the real deal.
The "pioneer" in this is Frank Van Alstine who even back in the early seventies used the best of what was available for his Dynaco Mods and AVA products and in FACT was the very first manufacturer to use a opamp/buffer as a headphone amp ! (AD743/BUF02)
The man listens and that means hours and hours of testing with music as the only way with the rest of us benefitting from the work.
I have boxes and boxes of chips that I thought maybe,just maybe this is THE ONE where I could find a combination that kicked a*s that no one else ever heard of but they are back in the box
The only "keepers" were the AD8397 which I got as a pre-production part with EVAL board,wired it up +6dB gain and the results so good with my Grados I popped in a box and is my portable ampo still until or if something better comes along.The other introduced at the same time is the AD8655 which runs at ridiculously low voltages and stil put out gobs of current but being a CMOS opamp it is not up to the sonics of the jfet/bipolar or all bipolar chips and i use it strictly for portable digital in AD converters and DA conversion analog stages where it kicks butt and even in a low power electret mic preamp running in at +30db the noise level is good.
two "finds" in thirty years is not much of a record
Dedicated audio chips other than power amps/digital DO crop up from time to time and that is the SSM series from ADI,a few from Maxim,The INA series from BB which could also be considered Medical field chips bast*rdised to audio,all the chips from THAT Corp. and a single opamp from National.
The rest ? Generic multi-use "plug it in and try it" chips.So many that no human has a shot at testing them all but the good news is most already tell you to not bother when you read the spec sheets.
On the buffer front my standard is the LH-0033 and LH-0066 (a BEAST !) with next in line being the "assisted" LT1010.
i say "assisted" because the input likes a low Z drive or you do not get the most out of the chip and why you see it mostly used with an opamp that has a low out Z either inside or outside the loop.The jfet front end gets around this.
Part ll is the class of operation.The LH series can be for all intent considered Class-a buffers and is why they sound so good even though on paper modern deices wipe the floor with them.Output current is lower than what a lot of modern opamps can deliver and that is the problem !
too many get hung up on the final delivery number and not the QUALITY of that delivery ! No normal opamp is getting even 40 mA in class-a output or it will burn up so when you see thse HUGE numbers you need to be aware they shift in class as the output power raises and in the end usually to class-B.
So the LT1010 looks mild in direct comparison to some of the "Super Op-amps" and when compared to many ADSL drivers plus every single other modern buffer but here is a difference.The 1010 can be driven class-a for almost its entire output range in the metal tab 5-Pin oackage by selecting a single resistor.This package also has terminals for DC blalncing for those who may need it unlike the DIP package which to me is a total waste of plastic.If i want limits then i would just as well use an opamp and a video CFB at that (AD811 still a goody
)
Many ways to skin the same cat but I am a "simple as I can do it" K.I.S.S. type of person who alwys chooses class-a operation when possible and depending on the end use just how much Class-a i can get.Obviously portable battery operation means reduction in requirement
rickster