which opamps are there for use?
Jul 14, 2007 at 1:27 AM Thread Starter Post #1 of 23

Soymilk

500+ Head-Fier
Joined
Jan 5, 2006
Posts
514
Likes
11
i'm trying to compile a list of opamps to consider for rolling opamps and to have in mind for other projects, but my knowledge is a little limited here. so far i only know of:

ad8610, opa627, opa637 (single channel)
ad8620 (dual channel)

what others are there? i know theres a couple other adi ones, but i forget their numbers. are there other popular burr-brown ones, too?

edit - ok, more than a couple others
tongue.gif


it'd also be helpful if you could say which amps they're used in, but if that's too much trouble then i'll look it up on my own when i have time. thanks
smily_headphones1.gif


compiling what's out there so far:

ADI:
AD8610/8620
AD823, AD843, AD744, AD8066, AD829, AD797
AD8397 (used unbuffered)
AD744 and AD829 ("comp pin out" to bypass the internal buffer)
AD8065 <- from another thread
AD825

TI/Burr-brown:
OPA627/637
OPA2132, OPA2134, OPA2107, OPA2227, OPA2228, RC4558, OPA604 <- from another thread
NE5534(A), NE5532(A)

National:
LM4562
LME49720 (dual) LME49710 (single)

New Japan Radio Co:
NJM4556
 
Jul 14, 2007 at 1:49 AM Post #2 of 23
biggrin.gif
ok let's start
AD823, AD843, AD744, AD8066, AD829, AD797
OPA2132, OPA2134, OPA2107, OPA2227, OPA2228
LM4562, NE5534(A), NE5532(A), NJM4556, RC4558
 
Jul 14, 2007 at 2:30 AM Post #4 of 23
well i'm thinking just any opamp that gets used in a popular diy amp, so i can look around at work and see what we have for use in future projects. i've already got some opa627 and 637's
biggrin.gif
which is nice since i'm about to build a ppa. (i dont want the 637's in a ppa though, right? i seem to remember reading somewhere that 637 + ppa = nono)

edit - you know, thinking a little more about it, i guess it'd make more sense to look specifically for ones useable in the ppa first
tongue.gif
but i'm open to any and all opamps.
 
Jul 14, 2007 at 2:57 AM Post #5 of 23
OPA637 is just a no-no with unity gain. It needs at least a gain of 5 to be stable

Of the ones I mentioned above, all are suitable for a Pimeta, except the AD823 (low-voltage, low current draw opamp), the RC4558 (ultra cheap opamp) and the NJM4556 (high current opamp). The AD8397 is also useless for use in a Pimeta (also high current). edit: they are for rather specific uses, and have some drawbacks if used in a buffered high quality amplifier like the pimeta
 
Jul 14, 2007 at 3:18 AM Post #6 of 23
hm, i guess if i find any of those i can build a pimeta for fun in the future and sell it later or something.

so what's the ad823 and rc4558 used in, a cmoy? and would the njm4556 work in a ppav2, or do you still need to go higher for that one? (b22?)
 
Jul 14, 2007 at 3:37 AM Post #7 of 23
yes, you could use the ad823 in a cmoy, but it has only 16mA output current, so it's best for low impedance, high efficiency 'phones. the rc4558 is used in ultra cheap consumer equipment. it's dirt cheap, so it's the opamp of choice if SQ doesn't matter.
all better amps have a separate buffer, the opamp is only used for voltage gain. so high current output is not something you need in a PPA or a Pimeta. The NJM4556 is famous as the opamp in the grado headphone amp, which is basically a cmoy. It delivers some of the high current the Grados need (but of course still worse than a buffered amp)
 
Jul 14, 2007 at 4:03 AM Post #8 of 23
ah, so i was thinking of it the wrong way around, if theres buffers you dont need high current output from the opamp. that puts things in perspective a little better.
 
Jul 14, 2007 at 4:11 AM Post #9 of 23
There is nothing wrong with the AD823 in a PIMETA. It has a problem in the CMoy because of the low current. The PIMETA adds buffers (BUF634) to all the opamps, so a low current opamp is not an issue.

Unity gain is only important in the Ground Channel in most of these amps. In fact, the OPA637 in L and R, with the OPA627 in Ground is a prime combo in the PIMETA, the PPA, and in the M3.

When making a list of suitable opamps for DIY headphone amps, the list should always start here:
Tangent's Notes On Audio Op-Amps
 
Jul 14, 2007 at 6:53 AM Post #11 of 23
There has been so much said about opamps here, do some searching. Opamps are the single most important part in a headphone amp (if not fully discrete). They seem to fall in and out of grace. There are some opamps of choice these days like:

LM4562 = LME49720 (dual) LME49710 (single)
AD8610/20
AD8397 (used unbuffered)
AD744 and AD829 ("comp pin out" to bypass the internal buffer)
OPA627/37

Just see what are used in the highly regarded amps (if it's not a secret) like Meier's Corda Move, LaRocco's PRII, LISAIII, MrX XP-amp, PPA, M3 etc.

What opamps we like seem to be a matter of personal preferences, synergy with phones and source etc. Buy a bunch of them you find promising and let your ears be the judge. It's not very expensive compared to the price of a complete PPA or PIMETA.

Some more things to consider are power supply, since some of the opamps need higher voltages to sound good, and class A biasing seem to improve more on some opamps than others.
 
Jul 14, 2007 at 9:42 AM Post #12 of 23
OPA627/637 are not worth their price which is justified in their intended applications like medical instrumentation. Use AD825 instead. If there is any true audio op-amp, that's the one. AD829 is nasty because it has got a huge input bias current which disregarded may cause high DC on the headphone output or unbalanced voltage division in the ground channel, depending on your application.
 
Jul 14, 2007 at 10:10 AM Post #13 of 23
Quote:

Originally Posted by balou /img/forum/go_quote.gif
yes, you could use the ad823 in a cmoy, but it has only 16mA output current, so it's best for low impedance, high efficiency 'phones.


You got the impedance part wrong. A low impedance headphone will draw more current from the opamp (if driven directly by the opamp) than a high impedance 'phone given a particular output voltage. It's just Ohm's Law -- I = V / R. If V is fixed and R is reduced, then I increases.

This is why an opamp with a robust amount of output current capability is needed to drive low impedance 'phones well.
 
Jul 14, 2007 at 4:26 PM Post #14 of 23
amb: yes of course... I somehow messed it up. high impedance, high efficiency headphones.

Quote:

Originally Posted by tomb
There is nothing wrong with the AD823 in a PIMETA


well, sound quality. it's not really regarded as high-end chip
 
Jul 14, 2007 at 5:06 PM Post #15 of 23
Quote:

Originally Posted by balou /img/forum/go_quote.gif
amb: yes of course... I somehow messed it up. high impedance, high efficiency headphones.
Quote:

Originally Posted by tomb
There is nothing wrong with the AD823 in a PIMETA


well, sound quality. it's not really regarded as high-end chip



Before the hoopla over the AD8397, the AD823 was highly regarded as one of the finest rail-to-rail opamps. However, that particular reference was about your statement:
Quote:

Originally Posted by balou
... all are suitable for a Pimeta, except the AD823 (low-voltage, low current draw opamp)


Low voltage, low current draw has nothing to prevent its application in a PIMETA. In fact, it can be a very good choice in a portable-version PIMETA because of its ideal operation in a battery-powered scheme.

It most certainly is regarded as a high-end chip - the AD823 was used for quite some time in Dr. Xin's SuperMini. He only replaced it, I believe, when the AD8397 came out. You can find this by searching for user "fixup" and "AD823" on many threads at Headwize:
http://headwize.com/ubb/showpage.php...47&srch=AD823;

Tangent gives it a pretty good review, too, in his Notes on Opamps referenced earlier.
 

Users who are viewing this thread

Back
Top