Discrete 5534 5532 opa604 replacement in diy projects like gssl anyone tries ?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

garp

Well-known member
Joined
Jan 15, 2012
Messages
49
You can find a lot of discrete direct replacement boards on ebay for single or dual opamps like this one :

https://www.ebay.com/itm/DUAL-Channels-Full-Discrete-Op-Amp-Module-replace-NE5532-MUS02-OPA2604-LME49720/272265794397?hash=item3f644e535d:g:-58AAOSwOVpXVYx4

or these (much more expensive) :

https://www.ebay.com/itm/Burson-V5i-D-HiFi-Discrete-Dual-Operational-Amplifier-OP-Amp-Chip-replace-OPA627/253318135406?epid=19011575292&hash=item3afaeffe6e:g:9nsAAOSwv-ZaNzKU

Does anyone tried them in diy projects ? i am planning to mod my 1178 input stage (ne5532) with one of them ... good idea ?


 
A lot of folks, myself included, do not believe op amps have differences that are actually audible. These folks don't usually speak up because generally we want people to have fun here and it's a bit of a drag to call people out. But I also think that, depending on the situation, it can be important to speak up if you see something a little off. So when people start talking about how op amps sound different, I get twitchy.

The differences in op amps that matter are impedance (bipolar vs fet), output offset, drive capability, noise, voltage swing, and other such things that depend entirely on the circuit in which they are being used.
 
I just swapped one toshiba 7136 for a opa604 in my pm180 mixer and results are more than audible ... believe me or not but when i make a simple difference betwween two identical signals passed thru a moded and non moded channel ... it leaves some "residues" more at low and high frequencies, proof that the result is far from identical. For my 1178 upgrade i followed Jim Williams advices with great results so far ... so the 5532 are already gone with amazing enhancement in the overall sound (the midrange tendency of this device is almost gone and the spectrum seems to be extended) i just asked myself about employing such opamps at no risk  in a gssl (power consuption ... current etc ... ). I am a long time defender of the discrete vs opamp designs (but also tube vs transistors, point to point vs pcb tape vs daw etc ... ) so you can see me as an old school  old bear maybe but i am also living in 2017 planet earth and when i read hifi consumers comments about how these opamps can dramaticaly improve cheap mp3 players or headphones amplifiers ... i am teased to try it in a ssl compressor ( by far i prefer the 202dbx gold can than  the that's chip ... go figure)
 
Lousy op amps could be audible, there are diminishingly smaller and smaller audible differences between better op amps, ignoring major technology differences like bipolar or fet input that suggest which applications they are better for..

op amps that didn't suck became widely available and reasonably priced since the 1980s.

JR
 
Unless extremely low quality OpAmps are used (709, 741), I wouldn't go through all this trouble.
You may end up with higher current consumption (causing problems) and oscillation, if 'faster' OpAmps are used.
Especially in commercial equipment, the designer usually knows what he is doing, and changing things may cause more problems than quality improvement!
 
analogguru said:
Now let's have a look what is inside....

Looks similar to a rail-to-rail opamp e.g. the CA3260:
but opa604 uses p-channel input JFETs and 553x uses NPN bipolar input devices, among numerous other internal differences.

JR
 
Op amps need some work to perform as intended.

Differences between op amps of sufficient quality (the 5534/32 is definitely among these) are small to inaudible once the following points have been taken care of:


  • a high quality PSU is used
  • layout and grounding optimized
  • local decoupling added/optimized

Don't bother switching op amps before dealing with all of the above.
 
Thanks living sounds for advice ;)  in the case of the 1178 i followed jim's instructions (bigger caps on power supply, multiple bypass capacitors, and results are amazing with the simple  5532 swap for the lm's ( i kept the "S" stamped philips in case of) . I didn't replaced the tl084 ... the noise floor has lowered headroom increased and spectrum widened enough to not go further now ... and use it ! I used it on the 2 buss maybe 10,5 seconds years ago (not joking ... its was so ... gritty !) but found it cool on the guitar's or drum buss enough to use it quite often at those places ( even back vocals are sometimes surprising with the 1178 ). But now ... i will make extensive tests to see if i can find more tasks to offer to it ! And ... the character is not gone, strangely the mods reveals interesting artefacts (pleasant !) of the urei design and i found it immediately more pleasant and gentle with the signal than before ... but wait ... it is still good rock'n'roll stuff  when cranked !

In fact my fears are more on the power consumption of the final design if i replace two of the 5532 with discrete ones ... also how to ovoid or minimize oscillations of the psu in this case ? I think that jim's mods will help but quite unsure ... i don't want to burn the power transformer of my 1178 stupidly ! i Have ordered a led strip to get rid of the bulbs and save some micro-amperes but anyway ... my skill stops here :(
 
the only ics in my desk were sitting in the summig amps. i replaced them with burson v5 amps and yes, i can hear a difference. an improvement to my ears. i did lots of a/bing
 
Ahhhhhhhhhhhh !!! Thanks salomonander ! What was the previous ic ? 5532 / 5534 or opa familly ?  did you add bypass cap between pins 1 and 2 ? Make some adjustments ?
 
garp said:
Ahhhhhhhhhhhh !!! Thanks salomonander ! What was the previous ic ? 5532 / 5534 or opa familly ?  did you add bypass cap between pins 1 and 2 ? Make some adjustments ?

The problem with questions like this are that context is important. I think most of us agree that a ne5532 in a properly designed circuit is about as good as it's going to get.

In the pm180 mixer example....an lm741 is the automobile equivalent of an AMC Gremlin.  8)
 
This is exactly why i asked ... he told me he had improvements so my  question is "what was the context ?" previous IC etc ... I am sorry but the problem with your answer is that i really have no clue of what a GMC is ... i am living in france and here everything above 3000 cm3 is considered as a truck ;)
 
garp said:
This is exactly why i asked ... he told me he had improvements so my  question is "what was the context ?" previous IC etc ... I am sorry but the problem with your answer is that i really have no clue of what a GMC is ... i am living in france and here everything above 3000 cm3 is considered as a truck ;)
an amc gremlin was an unpopular car, but not an audio term so feel free to ignore.

JR
 
i was attempting to joke also ;) . Seriously , i have ordered some just to be sure and judge by my humble ears ... some from germany and some from china .  i just wanted to say to all chinese disclaimers that their memory has some failure ...  many years ago europe and usa were invaded with crap products coming from a far east country called ... japan. A lot of stuff were crap but during the 70's decade, things begun to better and better manufactured to the point were no factory was able to follow the dance in terms of quality price AND reliability . You want an anecdote ? When i recaped my PM180 ALL capacitors were still in the specs and even sometimes higher ... The 1000 uf i removed from my 1178 urei (same era !) exhibits 450 uf on my meter ... the 220 uf : 110 ! in fact no cap in the urei was still even near the specs. Same for my prophet 5 v2 and my oberheim ob8 ... they always have a disease even when recaped , my mks80  mks70  stills work like when i bought them (new) ... and i never even opened them (what for ?).  End of story.
 

Latest posts

Back
Top