The best sounding audio integrated opamps

Jump to Latest
A

audiolic

Member
Joined 2012
a1greatdane said:
Dear audiolic, pls try OPA1612 from TI/BB. it's a straight swop for the double 'over bass with bad top' BJT input 2068. I found it amazing having 1st tryied the AD8599. The AD was a definate improvement, but the BB made all a not more natural. I don't personally like the National chip series much, since I find them a bit musically 'lifeless' - very accurate and clean, but somewhat 'to good' - bit like some Luxurious Sino cars - probably 2 perfect... Matter of taste as well - indeed...

All the best of luck
DocO
Click to expand...

just bought 10 OPA1612 for 55$ shipped to my place :)

I'll try those first.. thx for your input

Anybody i was wondering.. i need to remove some caps and make some kind of bridge... what would you recommend cheap.. some kind of jumper wire
i tought something like this http://www.digikey.ca/scripts/DKSearch/dksus.dll?Detail&itemSeq=119017233&uq=634819544201584213

but it's a package of 200.. i just need 30 or so.. I know i could also buy some bypass ceramic cap but there to much choice to choose from hehe.. anyway idea on cheap bypass caps that would do the same as a solder bridge?

would this do http://www.ebay.com/itm/100pcs-0-1u...438?pt=LH_DefaultDomain_0&hash=item27ca82fc46
 
Last edited:
A

audiolic

Member
Joined 2012
here's the pic of the board and layout

hifimedy use only 2.2uf capacitor to remove DC out from CS4398 on is dac..
I guess i could also replace those 47uf 16v caps with 2.2 uf elna silmic II..
instead of removing those..

at the same time.. my guess is that the OPA will work perfectly anyway without any caps before it since DC from CS dac is quite small (ex. 2.2 uf cap to remove dc)?? what do you guys think. I did remove those caps without changing the JRC on an e-mu 1820m and wow did it clear up the sound :)

still what do you think i should do here.. bridge(no cap), 2.2uf elna silmic II, 2.2uf film cap, bypass cap?

i could buy this http://www.ebay.ca/itm/Solid-Core-0...mponents_Supplies_ET&hash=item1c24921786#shId
 
Last edited:
C

chris719

Member
Joined 2004
DC from CS4398 is 1/2 VCC if I remember correctly. You will get a large DC offset if you remove them and are using the balanced output as unbalanced.

I can't remember what the R value is there but I'm pretty sure 2.2uF is too small... you might roll off the low end. If you are going to have to use an electrolytic anyway it may as well be a big value.

Replace it with a 47uF Panasonic SU Bipolar (which actually fits the footprint in the 1820m anyway) or replace it with a 47uF Nichicon Muse ES Bipolar (which is too large to fit correctly but you can probably make it work).
 
Last edited:
abraxalito

abraxalito

Member
Joined 2007
So its an input buffer to a poweramp? In which case seems you don't actually need JFET inputs - my personal preference is for bipolar inputs. So assuming you haven't got very large biassing resistors (<100k say) then I'd go for a 5532 and make sure the grounding arrangement protects it from common-mode RF currents from the source component.

The classic faux-pas on such circuits as regards SQ is to implement an RC RF filter prior to the opamp but have the ground side of the C connect to the incoming signal ground.
 
A

audiolic

Member
Joined 2012
chris719 said:
DC from CS4398 is 1/2 VCC if I remember correctly. You will get a large DC offset if you remove them and are using the balanced output as unbalanced.

I can't remember what the R value is there but I'm pretty sure 2.2uF is too small... you might roll off the low end. If you are going to have to use an electrolytic anyway it may as well be a big value.

Replace it with a 47uF Panasonic SU Bipolar (which actually fits the footprint in the 1820m anyway) or replace it with a 47uF Nichicon Muse ES Bipolar (which is too large to fit correctly but you can probably make it work).
Click to expand...

Are you sure of this? so the OPA in the path will not fix the DC offset?
There is also 3 small stereo 1/8 mini trs on the back that mimic the output of the balanced outputs... (i could always use them instead of the balanced 1/4 jack. My system is unbalanced anyway at the moment..

I really wonder about putting electrolyc caps in the path.. I don't like it at all heheh..

here's what somebody said on another forum (On the CS4398, there is practically no dc offset to speak of. I measured a fraction of a mV. Maybe they vary from unit to unit. )

somebody else after (well you were spot on with the dac it has 2.4vdc my new agilent meter says and no dc after opamp so I soldered vishay ww 250 ohm from cap +(after removing) to rca + and left the rca pin connected to board. should I clip that?)

(somebody else said)
Quote:
Originally Posted by Ryssen
There´s 2v DC on the DAC´s output,the trafo or the caps blocks the DC so it wont reach the preamp or the amplifiers input.

Just make a virtual ground with two resistors from the plus voltage to ground. The midpoint are at the DAC output voltage.
50 or 100 ohm works OK.
A JFET buffer makes it even better, but the resistors alone sounds much better than capacitors or transformers.
----
there's is resistors in the path I wonder if they are use this way even without the caps..
----

dam op1612 is not JFET snif hehe
what would be the best JFET alternative to opa1612?
OPA2228? or op1612 will do even better than JFET buffer?
 
Last edited:
A

audiolic

Member
Joined 2012
Kindhornman said:
audiolic,
Are you quite sure that this is a balanced output device and that isn't just a stereo output. I have only seem balanced devices using XLR connectors, never saw anything with a 1/4" RCA connector.
Click to expand...

the e-mu 1616m has some kind of relay to 1/8 unbalanced outputs
and balanced 1/4" outputs not XLR or RCA
An externally hosted image should be here but it was not working when we last tested it.


e-mu drivers sucks but the card is one of the best ever made. Some mod could make it the best hihi CS4398 are great dac chip even the refined high of the bifrost dac with AK chip didn't compete with the space and micro/macro details of the cs4398.. in my opinion.. the caps at the output suck tho... once removed it's like night and day.. didn't change the JRC2068 tho so can't wait to replace them with OPA1612

what is sad a little is that clock on it are small smd clock (3mmx7mm) .. wonder if i can find good clocks in this size or if i could fit a Crystek anyway.. the best would be have iancanada fifo reclocker + low phase noise clocks between my internal RME card with good drivers and the modded Emu 1616m.. but still the e-mu would work on external clock .. but i would accept that hehe :)

the internal clock is said to have 500pico second jitter and 750pico seconds on toslink input.. i guess it's acceptable for now..
 
Last edited:
A

audiolic

Member
Joined 2012
hehe sorry just found out what a jfet buffer was :)
http://www.diyaudio.com/forums/pass-labs/124889-b1-buffer-preamp-357.html

kinda too complicated for my setup.. but i guess i'm stuck with the caps and opamp or i use this-> (Just make a virtual ground with two resistors from the plus voltage to ground. The midpoint are at the DAC output voltage.
50 or 100 ohm works OK.

One resistor from the DAC PSU +voltage to virtual GND.
One resistor from the virtual GND to the real GND.
The virtual GND (midpoint between the resistors)
will now have the same voltage DC potential as the DAC outputs.

The "real" GND must not be used as it is the virtual GND you connect to your amplifier.
)

Why would i connect anything to my amplifier other than audio cable?

anybody would like to draw me a plan to make something like this but that would work with balanced out?
 
Last edited:
A

audiolic

Member
Joined 2012
ohh after some research
NPCAP? PSA Series - United Chemi-Con - Polymer | Dynamic Catalog | DigiKey Corporation

anybody?

one guy wrote this
--------------
blackgate?? meh, i totally dont know what the fuss is about, even with hundreds of hours the sound isnt right around 250-350hz,

i just dont like them and would take silmic II or nichicon kz over them any day in signal, especially if the silmic are bypassed with a small pps, polypropylene, or relcap polystyrene film and foil cap. in digital power supply i'll take oscon, or nichicon solid polymer for leaded, for smd the panasonic SP/SX series special polymer caps are in another class @ 0.005ohms, kemet polytantalum smd if very small size is needed such as the newish 1206 size, al bypassed when appropriate with c0g, or ps film, bulk decoupling in power supply the nichicon, silmic II for smaller positions, then mundorf or F&T for 2200 up, or for ultrareliable bomb proof high current, the Evox/Rifa PEH169 are impossible to beat for me and if higher performance again is needed and you have the space, adding a 50-200uf unlytic polypropylene shunt to the rifas is unbelievable. blackgate do not even figure in all of that, if cheaper bulk caps with decent performance are needed the pana FC are indeed a great value.

i wont go into films, there are plenty of threads on that already and at that level and application i would strive for no cap
--------------------


interesting :).. i guess elna silmic II bypassed with http://www.partsconnexion.com/capacitor_film_reliablecap.html


would be interesting.. what do you think? Which relcap would you recommend for bypassing a 47uf 16v elna silmic II

this is an interesting read to
http://canada.newark.com/pdfs/techarticles/oct2005.pdf
 
Last edited:
A

AndrewT

R.I.P.
Joined 2004
audiolic said:
hehe sorry just found out what a jfet buffer was :)
http://www.diyaudio.com/forums/pass-labs/124889-b1-buffer-preamp-357.html

anybody would like to draw me a plan to make something like this but that would work with balanced out?
Click to expand...
Since you have found the Pass site, go and look at some of his balanced input stages.
BTW,
the B1 is one of the simplest circuits ever.
It is two jFETs stacked one above the other.
Of the 6 legs, two are connected at the bottom. Two are connected in the middle. That leaves just 4 external connections: Plus power, minus power, input, output.

All the rest is simply an RC PSU run off a wall wart power plug.
 
A

audiolic

Member
Joined 2012
AndrewT said:
Since you have found the Pass site, go and look at some of his balanced input stages.
BTW,
the B1 is one of the simplest circuits ever.
It is two jFETs stacked one above the other.
Of the 6 legs, two are connected at the bottom. Two are connected in the middle. That leaves just 4 external connections: Plus power, minus power, input, output.

All the rest is simply an RC PSU run off a wall wart power plug.
Click to expand...

i wonder if i could fit 6channel balanced inside my e-mu 1616m card enclosure.. I would need a mini board hehe
 
You must log in or register to reply here.
Top Bottom