The Opamp thread

[SpudHarris]

Any of you guys have experience with the TLE2141?? I've ordered a couple from Farnell, they are ''Jelly Bean'' OpAmps (real cheap) but are raved about on the D10 OpAmp thread as the L/R of the new ''Top Kit'' again biased to Class A.

 

[ROBSCIX]

There are many jelly bean opamps that can give you good soudn in the right circuits.

 

[leeperry]

cheap generic op-amps seem to work really well as buffers, combined w/ killer op-amps set as LPF or I/V...they tend to give a tuby sound to kill the harshness(553x/4580)

and can you beat the LME49xxx soundstage anyway? it's so amazing on headphones

 

[LuciferX]

Quote:

Originally Posted by leeperry /img/forum/go_quote.gif cheap generic op-amps seem to work really well as buffers, combined w/ killer op-amps set as LPF or I/V...they tend to give a tuby sound to kill the harshness(553x/4580) and can you beat the LME49xxx soundstage anyway? it's so amazing on headphones

What do you think of this?:

4580 (buffer) > OPA2132 (LPF or I/V)
4580 (buffer) > LME49720NA (LPF or I/V)
4580 (buffer) > LME49860NA (LPF or I/V)

(My HT OMEGA Claro Halo has 4 x 4580 by default)

 

[12Bass]

Generally speaking, I prefer using an AD8397 over an NJM4580 for driving headphones. I've found the 4580 to sound somewhat edgy and murky when compared with better op amps. The AD8397 can be a bit difficult to use in some circuits; however, it can provide six times the output current and sounds clearer and more punchy in the bass, IMO. I recently installed one in my RME Digi96/8 PST (with AD1852).

 

[majkel]

When I replaced NJM4580 for AD8022 as the output buffers in my CD player, overall sound resolution improved.

 

[murrays]

Quote:

Originally Posted by ecclesand /img/forum/go_quote.gif Interesting. So for a dual opamp...since there are two outputs, would you need 2 resistors...one from the +V to output A and one from the +V to output B?

Yes.

Quote:

Originally Posted by ecclesand /img/forum/go_quote.gif Or could you use one resistor and pigtail one end of it to both the A and B output?

Definitely Not! This would short the two outputs together.

 

[murrays]

Quote:

Originally Posted by ROBSCIX /img/forum/go_quote.gif The biasing only works with the AD744/NE5534 or the OBCA only works with these opamps? I have never seen the OBCA idea, can you explain what is going on with that unit? -Very interesting BTW. I appreciate your help and information.

have a look at this article ...
http://waltjung.org/PDFs/ADI_1992_Seminar_Audio.pdf

on this useful site ...
Services

 

[leeperry]

Quote:

Originally Posted by LuciferX /img/forum/go_quote.gif What do you think of this?: 4580 (buffer) > OPA2132 (LPF or I/V) 4580 (buffer) > LME49720NA (LPF or I/V) 4580 (buffer) > LME49860NA (LPF or I/V) (My HT OMEGA Claro Halo has 4 x 4580 by default)

yeah, it looks like you'd need to change the two first op-amps that are used for polarity stuff(AK4396 design), and then the two that are used as buffer.

I don't know how the HP amp goes into your soundcard design(it's prolly used as buffer, so only 2 op-amps are actually swappable for this output?), but the LME49720HA does wonders on my HD2(same DAC/caps as your card)....great bass response, very HQ mids and HUGE soundstage.

the high gain JRC4580 buffer(I've left it on +70%, but there's also a switch for +200%

) gives me a much better low end bass response and makes the sound very enjoyable as a whole on the HP out....I'd dare to talk about synergy w/ the 128X oversampling from the AK4396, plus the LME49720 serie is known to offer some very high class post-filtering behind DAC's.

I know DAC's are not supposed to have a "sound", but I'm more keen on the AK4396 design/sound(mandatory 128X OS) than on the PCM1792....many op-amps lack in I/V I was told(the AK4396 doesn't need it), like the burson's for instance...and the default OS on the BB DAC's is 64X, some manufacturers(like Asus) don't bother forcing it to 128X...the AK4396 sure does miracles

 

[ROBSCIX]

Quote:

Originally Posted by murrays /img/forum/go_quote.gif have a look at this article ... http://waltjung.org/PDFs/ADI_1992_Seminar_Audio.pdf on this useful site ... Services

Thank for the information.
I will add it to my nightly reading list.
There are some great documents out there, put out by the engineers
at various companies. Books dealing with opamps design, implemention
etc. Can always find something more to read. I just need to get myself a kindle
or other portable document reader so I can take my hobbies with me!

 

[LuciferX]

Quote:

Originally Posted by leeperry /img/forum/go_quote.gif yeah, it looks like you'd need to change the two first op-amps that are used for polarity stuff(AK4396 design), and then the two that are used as buffer. I don't know how the HP amp goes into your soundcard design(it's prolly used as buffer, so only 2 op-amps are actually swappable for this output?), but the LME49720HA does wonders on my HD2(same DAC/caps as your card)....great bass response, very HQ mids and HUGE soundstage. the high gain JRC4580 buffer(I've left it on +70%, but there's also a switch for +200% ) gives me a much better low end bass response and makes the sound very enjoyable as a whole on the HP out....I'd dare to talk about synergy w/ the 128X oversampling from the AK4396, plus the LME49720 serie is known to offer some very high class post-filtering behind DAC's. I know DAC's are not supposed to have a "sound", but I'm more keen on the AK4396 design/sound(mandatory 128X OS) than on the PCM1792....many op-amps lack in I/V I was told(the AK4396 doesn't need it), like the burson's for instance...and the default OS on the BB DAC's is 64X, some manufacturers(like Asus) don't bother forcing it to 128X...the AK4396 sure does miracles

The Claro Halo use this diagram:

[size=xx-small]The each polarity signal produced by AK4396 goes to 2 dual opamps (U8 for left, U10 for right) In those dual opamps, each opamp handles only one polarity, either positive signal or negative. Then both exclusively amplified positive and negative signals are re-assembled and goes to another opamps (U9 for left, U11 for right) for second amplification. In the second dual opamp, U9, The one opamp is for left headphone and the other is for left RCA output. In U11, one opamp is for right headphone and the other is for right RCA output.


This design maximizes the feature of AK4396DAC and it also dramatically improves stereo-talk.

It is recommended to swap all four opamps for your preferred analog sound
color. Factory installed opamps are 4580 (JRC).

If you choose to swap only two, they must be U8 & U10 ( They are marked on PCB of Claro halo)[/size]

 

[leeperry]

well, get a bunch of free samples(including the 49720HA/OPA2132P/LT1057/etc)...and see how it goes.

rolling op-amps is a trap, as none of these chips is perfect....only the burson's would be close to this apparently(as buffers, not I/V).

if you roll too much, you'll be expecting the impossible....I'd love the tamed down trebles of the 49722MA, the crazy low end bass/huuuuge soundstage of the 49720HA, and the super grainy mids of the OPA2132P. needless to say, this is not going to happen

the 49720HA is very easy to roll in the crappy STX pressed DIP8 sockets, but in proper machined sockets(where contacts are much tighter)...it's a major PITA:

so that's good! they'll stay there for a while...they're not perfect as the trebles are quite whiny(I have to EQ down my ear resonances anyway), but the bass/soundstage/mids are as good as it gets....the tuby high gain JRC4580 on top gives a very enjoyable SQ on HP, much higher than anything I had ever heard coming out of those Asus Xonar things....it sounds super-HQ HiFi, not digital harsh

 

[murrays]

Quote:

Originally Posted by leeperry /img/forum/go_quote.gif rolling op-amps is a trap, as none of these chips is perfect....only the burson's would be close to this apparently(as buffers, not I/V). if you roll too much, you'll be expecting the impossible...

Looking at the fact that you are putting them into a board that plugs into a PC, I would say that the op amps would be the least of the problem. A lot of electrical and RF interference may be polluting the power supplies. This may be why the metal can chips give you a noticeable difference? A lot of the difference you hear may not be the op amps per se, but the different types of noise fed into their power pins, and how they deal with it. It is always best to have clean power supplies - a difficult task in a PC environment. Have you considered/tried improving the capacitors on your board?

 

[leeperry]

indeed, a PC is a less than ideal environment...I had some EMI problems, but I appear to have fixed them all?! anyway, maybe I'll get one of these gigabyte mobo w/ 2oz of copper...I was told that they are fantastic for audio systems

whatever now w/ the HD2 or back then w/ the Asus Xonar's, I've never had anything polluting the audio...in a very audible way, that is.

I understand an external stabilized PSU would help, but quite honestly many external DAC/HP amp run from a wallwart...that is even worse than a quality PC ATX PSU? a proper design is supposed to really clean the incoming current?

I run a "deluxe" edition(released 1 year ago) of that HD2 board, w/ supposedly improved caps:

the sound sure sounds cleaner to me than on the Xonar..especially the stereo seperation, which doesn't appear "messy"...the bass is far better controlled, and the mids appear much more refined. I know I'll stay away from Nichicon caps in the future.

but yes, the metal cap 49720 is a God bless for noisy environments!

 

[ROBSCIX]

Except you are jaming them loosley into the sockets and leaving the leads exposed.
Weren't you complaining about oscialltion when you did this method on another card?
I suggest you get some adapters and install them the right way.