The Opamp thread

[raoultrifan]

Thanks for sharing SpudHarris!

By looking to the specs (especially speed, unity gain stable and input bias current) I can see that OPA602 could be a good candidate to I/V stage of a DAC, though it's noise is a bit higher than OPA604, way above the noise of the OPA8597 and above the AD797. It's speed is not very high, so it will probably not oscillate in most configurations, even with a badly designed PCB.

In what stages have you been using it, please? I/V? LowPass? Voltage Gain? Output buffer?

Thank you!

 

[SpudHarris]

What do you mean? Are they not both considered good?

Sorry, this was in response to reply from imran27

 

[SpudHarris]

Thanks for sharing SpudHarris!

By looking to the specs (especially speed, unity gain stable and input bias current) I can see that OPA602 could be a good candidate to I/V stage of a DAC, though it's noise is a bit higher than OPA604, way above the noise of the OPA8597 and above the AD797. It's speed is not very high, so it will probably not oscillate in most configurations, even with a badly designed PCB.

In what stages have you been using it, please? I/V? LowPass? Voltage Gain? Output buffer?

Thank you!

Hi raoultrifan, thanks for the reply, I have only been testing on the Lycan of late but ideally I will be also using in my Fi-Quest. Really I just wondered if there are any new Opamps worth trying.

 

[imran27]

What do you mean? Are they not both considered good?

Sorry, this was in response to reply from imran27

I don't know anything about OPA602BP. I want to know how does it compare to OPA1612?

As of now my favorite is the OPA2209

 

[SpudHarris]

Opa602BP is a great Opamp. The TO99 version is very rare (Opa602CM). I'm sure Leeperry will chip in regarding these. I think he hooked me up with a seller about 5 years ago of the TO99 version. Managed to grab a couple plus a spare.

This is my go to at present but can't compare against 1612 until they arrive. Like I said, I seem to have got rid of my1612's by including them in portable amp sales.

 

[raoultrifan]

[...] Really I just wondered if there are any new Opamps worth trying.

Hi! Define new.
AFAIK the OPA1688 is kinda a new, but really new would be SS V6 from Burson and MUSES03 from NJM/JRC. The first one was already tested with success in CMoy and also in Obejctive2, the second was tested in many devices and the last one should be very good in I/V stages (not sure who tested yet, sorry).

 

[imran27]

Hi raoultrifan, thanks for the reply, I have only been testing on the Lycan of late but ideally I will be also using in my Fi-Quest. Really I just wondered if there are any new Opamps worth trying.

OPA1692, OPA2189.

I've pre-ordered my OPA1692 and eagerly waiting for OPA2189 to go into production. The single version OPA189 is already at my desk, just waiting for adaptors.

 

[SpudHarris]

Farnell has OPA189 as a dual Opamp... Will buy a few anyway!

 

[SpudHarris]

Muses 8832? Anyone tried?

Have got OPA2209, OPA189 and OPA1612 on order will report back when I've tried them.

 

[raoultrifan]

Don’t have any devices to test 8832, also needs to be soldered on the PCB. It’s a low-voltage opamp (up to 14V), so it’s mostly for USB powered devices, although seems to have very good specs.

 

[endia]

Don’t have any devices to test 8832, also needs to be soldered on the PCB. It’s a low-voltage opamp (up to 14V), so it’s mostly for USB powered devices, although seems to have very good specs.

is that "1V/µs" slew rate enough?

 

[raoultrifan]

It should be fine, I tend to trust Japanese companies usually.

http://www.radio-electronics.com/info/circuits/opamp_basics/operational-amplifier-slew-rate.php

L.E.: 8832 is designed for low-power devices (probably USB-based power), that means that ouptut RMS is somewhere around 1.3V, hence peak-to-peak should be around 3V.

 

[Rroff]

is that "1V/µs" slew rate enough?

Personally I'd say no - several well regarded sources recommend at least ~1.7V/µs for audio or that 1V/µs is "plenty" for audio but that assumes its bang on when quite a lot of op amps it is actually the "typical" rating (including most NJM/JRC devices) and can drop down significantly below.

An old school audio and electronics engineer I know IRL when I brought the subject up was pretty adamant you "should" aim for 10V/µs with audio heh though he is used to working with like 1970s or earlier electronics which might change the game a bit (he isn't a big fan of op amps over discretes either for audio).

Back on topic a bit - nwavguy makes a case for 1V/µs to justify use of the 4556 and while I've not noticed it with lower impedance, high current headphones with some high impedance headphones that don't take a lot of power I've noticed some minor issues with articulation which while I can't prove it I'm like 99% sure are down to the slew rate - when I drop in a couple of RC4580 or LMH6643 (low voltage part so don't drop it into a +/- 12V amp, etc.) I don't notice it. (One tune I use for testing is Swedish House Mafia - One which has a number of instruments used which are close to pure wave forms).

End of the day it is more my subjective impression than something backed up with science but I've always found better results using op amps that can do 5-10V/µs (there is no real benefit for shooting for the stars with slew rates and it isn't uncommon with op amps designed for things like ADSL drivers that the slew rates come from the use of "boosting" systems that don't always produce the best results for audio).

EDIT: This is what Doug Self has to say in this context (while talking about an op amp with a slew of 9V/µs):

Assume you are running your op-amp from±18V rails, and that it can give a±17V swing on its output. For most op-amps this is distinctly optimistic, but never mind. To produce a full-amplitude 20 kHz sine wave you only need 2.1 V/µs, so even in the worst case there is a safety margin of at least four times. Such signals do not of course occur in actual use, as opposed to testing. More information on slew limiting is given in the section on op-amp distortion.

 

[endia]

Personally I'd say no - several well regarded sources recommend at least ~1.7V/µs for audio or that 1V/µs is "plenty" for audio but that assumes its bang on when quite a lot of op amps it is actually the "typical" rating (including most NJM/JRC devices) and can drop down significantly below.

An old school audio and electronics engineer I know IRL when I brought the subject up was pretty adamant you "should" aim for 10V/µs with audio heh though he is used to working with like 1970s or earlier electronics which might change the game a bit (he isn't a big fan of op amps over discretes either for audio).

Back on topic a bit - nwavguy makes a case for 1V/µs to justify use of the 4556 and while I've not noticed it with lower impedance, high current headphones with some high impedance headphones that don't take a lot of power I've noticed some minor issues with articulation which while I can't prove it I'm like 99% sure are down to the slew rate - when I drop in a couple of RC4580 or LMH6643 (low voltage part so don't drop it into a +/- 12V amp, etc.) I don't notice it. (One tune I use for testing is Swedish House Mafia - One which has a number of instruments used which are close to pure wave forms).

End of the day it is more my subjective impression than something backed up with science but I've always found better results using op amps that can do 5-10V/µs (there is no real benefit for shooting for the stars with slew rates and it isn't uncommon with op amps designed for things like ADSL drivers that the slew rates come from the use of "boosting" systems that don't always produce the best results for audio).

EDIT: This is what Doug Self has to say in this context (while talking about an op amp with a slew is enoughof 9V/µs):

that is why I asked..
just checked nwavguy's site, he is referring Doug Self and his claim is 3 V/µs is "fast enough" for many audio applications although @raoultrifan's "scenario" needs less than 0.5 V/µs at a frequency of 25kHz in calculation (http://www.radio-electronics.com/info/circuits/opamp_basics/operational-amplifier-slew-rate.php)..

 

[Rroff]

that is why I asked..
just checked nwavguy's site, he is referring Doug Self and his claim is 3 V/µs is "fast enough" for many audio applications although @raoultrifan's "scenario" needs less than 0.5 V/µs at a frequency of 25kHz in calculation (http://www.radio-electronics.com/info/circuits/opamp_basics/operational-amplifier-slew-rate.php)..

Fair enough if you've a ~3V peak from something more designed for USB application then in theory you barely need 1V/µs - but that won't get the best out of some headphones and even then I'm not entirely convinced. Plugging the kind of numbers from my variation of the O2 comes at around 1.5V/µs.

Dunno I'm not an engineer but 100s of hours of messing about with DIY amps doesn't leave me with the impression lower slew rates are ideal.

EDIT: One thing to bare in mind as well is that a lot of my impressions come from using the Sennheiser HD600 which seem to be a fan of all things voltage.