TI parts are the problem lacking quality. Since TI took over National Semi I ran out and purchased a supply of new old stock original National Semi.
Op amps that sound good? LF351 LF353, LF347 NJM2068 in certain applications the AD711 series. There are a few more but do not recall right off. In circuits optimized for the part a lot of people find good results with the TL072 series but, I could never ever get them to sound decent no matter what I did. I do know everyone that got them sounding good first graded them by running on ±20 volt supplies and used only the ones still working after a day.
Things to look for in ANY amplifier circuit for audio. Symmetric slew rates. Good high frequency CMRR like the LF353 is one of the best at 1MHZ. Not to fast. High speed means lots of susceptibility to RF injection and ofter stability problems. Small signal audio does not need 200V/µS slew rate. Unity gain stable. Plenty of bias current for the opamp. Micro power is for micro power devices which might only be portable audio devices which always sacrifice sound quality.
CHIENMORT- I get it. You are deaf to and cannot hear the gross distortion. That op am is from the early 1970's belonging in a museum. Not to be ashamed you cannot hear this gross distortion. Many cannot which is why this garbage op amp and several others like the 5534 are still used. Some simply cannot hear. I can easily identify that series where ever it is used by is signature bad sound.
Thank you for the suggestions.
I've never tried the LF353, but I have tried the LF356 single as another member here described its sound as "neutral".
Some may describe the 356's sound as "neutral", but I found it somewhat harsh in most circuits I tried it in and just not able to live with it.
I'll look for some NOS National LF353's on eBay and order a couple to try.
"Op amps that sound good? LF351 LF353, LF347 "
I find the LF353 is almost unlistenable, ditto the LF347.
Marchand used to ship their crossovers with LF353, replacing with OPA2604 makes a not-subtle improvement in sound.
DBX uses the LF347 in its crossovers, except for the ones handling the HF, where they use the TL074.
"You are deaf to and cannot hear the gross distortion"
No comment.
I find the LF353 is almost unlistenable, ditto the LF347.
Marchand used to ship their crossovers with LF353, replacing with OPA2604 makes a not-subtle improvement in sound.
DBX uses the LF347 in its crossovers, except for the ones handling the HF, where they use the TL074.
"You are deaf to and cannot hear the gross distortion"
No comment.
djk- Am so glad I do not have to listen to your gear or system. My good luck!
djk- What is this "unlistenable" character of the LF353 if you would please describe? How about the NJM2068?
To do the same for the 5532, it has really sloppy muddy bass sounding like wet mud and a rubber boot. Even closer would be like all bass kind of sounds like, and not trying to be rude or crude, the pop flap of a single large flatulent bubble exiting the place those come from. The bass too is not awful. It is very unsatisfying lacking punch being floopy. The mids and highs are not awful either however, they have a sound which grates at all frequencies and levels sounding a bit like individual rain drops on a tin roof mixed in with the music. Hundreds or thousands in a row in with the music. This sound is constant and follows the music. There are awful sounding op amps like the 833 and I would choose the 5532 over that screech and howl op amp any day.
356 is single ended output is it not? Not zesty.
LF411 is not really bad either though I prefer the LF351. The NJM2068 is as good as the best I have heard.
A lot of listeners like 1" dome tweeters without a lens or anything to control directivity. An open dome. That is fine. 1" dome like that all suddenly flip phase on axis at or near 5500Hz and thereby mask out distortions above that frequency as incoherent sound. Does anyone here use a 1" dome? My tweeters do no flip phase until above 22,000Hz. They stay in a single phase from resonance to above the mentioned frequency. This is required if you wish to know what the highs are actually sounding like.
I know this is not speaker land however, here is a tweeter which stays in a single phase to above 20,000Hz. This one requires EQ and that is easy and correct.
https://www.madisoundspeakerstore.com/horn-tweeters/seas-prestige-27tbcd/gb-dxt-h1499-tweeter/
if anyone is interested...
djk- What is this "unlistenable" character of the LF353 if you would please describe? How about the NJM2068?
To do the same for the 5532, it has really sloppy muddy bass sounding like wet mud and a rubber boot. Even closer would be like all bass kind of sounds like, and not trying to be rude or crude, the pop flap of a single large flatulent bubble exiting the place those come from. The bass too is not awful. It is very unsatisfying lacking punch being floopy. The mids and highs are not awful either however, they have a sound which grates at all frequencies and levels sounding a bit like individual rain drops on a tin roof mixed in with the music. Hundreds or thousands in a row in with the music. This sound is constant and follows the music. There are awful sounding op amps like the 833 and I would choose the 5532 over that screech and howl op amp any day.
356 is single ended output is it not? Not zesty.
LF411 is not really bad either though I prefer the LF351. The NJM2068 is as good as the best I have heard.
A lot of listeners like 1" dome tweeters without a lens or anything to control directivity. An open dome. That is fine. 1" dome like that all suddenly flip phase on axis at or near 5500Hz and thereby mask out distortions above that frequency as incoherent sound. Does anyone here use a 1" dome? My tweeters do no flip phase until above 22,000Hz. They stay in a single phase from resonance to above the mentioned frequency. This is required if you wish to know what the highs are actually sounding like.
I know this is not speaker land however, here is a tweeter which stays in a single phase to above 20,000Hz. This one requires EQ and that is easy and correct.
https://www.madisoundspeakerstore.com/horn-tweeters/seas-prestige-27tbcd/gb-dxt-h1499-tweeter/
if anyone is interested...
5532s are ubiquitous in expensive studio consoles, it s about impossible that several of those werent in the sound road that ended in you testing room..
So, how is it that your test records didnt sound muddy since they went through 5532s before entering your LF411 and other biFet op amps..?..
Or perhaps that they compensated the 5532s alleged muddy sounds, that must be the explanation after all....
So 1000 people wont hear the distorsion of a 5534 but you pretending that it s audible should be the most probable outcome..?.
You realize what you are claiming.?..
Last edited:
"5532s are ubiquitous in expensive studio consoles, it s about impossible that several of those werent in the sound road that ended in you testing room."
This is not true at all in the last 15 years. While it is true many still cling to this junk part it is by no means the choice in many mixing consoles. More modern and better sounding parts abound.
No one here has taken the simple challenge of placing 10 op amps of a particular kind in a row and listen as I have many times. djk has not written why he dislikes the op amps I find most useful for audio. This thread looks more like a religious ******* match with belief in the god part 5532 over all other saviors. Just because something once was used often or the best at the time in no way validates that part forever. Try a little real world testing.
As far as those listening test and "my system," the only thing which is common to the new audiophile on their first experience in listening to my system is catatonic jaw drop. In short the comments run along the line, "I have never heard anything even close to this" or something equivalent. The head of Nissan audio for cars came by and could not tear himself away. To me, what most call high end is like AM radio therefore, listening and evaluating parts using AM radio standards, everything supporting the use of the 5532 in the thread is inarguable. It is a matter of experience and perspective, or more likely lack thereof. My best recommendation is try these various parts yourself and use the one you find most satisfying in your system. That is all which really matters.
This is not true at all in the last 15 years. While it is true many still cling to this junk part it is by no means the choice in many mixing consoles. More modern and better sounding parts abound.
No one here has taken the simple challenge of placing 10 op amps of a particular kind in a row and listen as I have many times. djk has not written why he dislikes the op amps I find most useful for audio. This thread looks more like a religious ******* match with belief in the god part 5532 over all other saviors. Just because something once was used often or the best at the time in no way validates that part forever. Try a little real world testing.
As far as those listening test and "my system," the only thing which is common to the new audiophile on their first experience in listening to my system is catatonic jaw drop. In short the comments run along the line, "I have never heard anything even close to this" or something equivalent. The head of Nissan audio for cars came by and could not tear himself away. To me, what most call high end is like AM radio therefore, listening and evaluating parts using AM radio standards, everything supporting the use of the 5532 in the thread is inarguable. It is a matter of experience and perspective, or more likely lack thereof. My best recommendation is try these various parts yourself and use the one you find most satisfying in your system. That is all which really matters.
Hi Guys,
I was under the same curse, what was better,I bought 20 op amps in pairs and built one circuit for them and played each one until I could say what was good or not good for each op amp.
A active system takes advantage of this as one op amp might sound better as in tweeters or high end as another might be best for bass and another for mids,so buy a few and try them, I was surprised at the difference in the scale and tone !
NS
I was under the same curse, what was better,I bought 20 op amps in pairs and built one circuit for them and played each one until I could say what was good or not good for each op amp.
A active system takes advantage of this as one op amp might sound better as in tweeters or high end as another might be best for bass and another for mids,so buy a few and try them, I was surprised at the difference in the scale and tone !
NS
I once built a precision rectifier and lowpass filter to convert MAD Vac to DC.
I had 3 opamps to hand: TL072, NE5532 and LF353 ( well actually I had LM324 and NE5534 too but forgot I did). The rectifier section requires an opamp with a fairly good GBW, so I chose the 5532 rather than LF353. But these things aren't always straightforward it seems...
Initially I used 5532 on both rectifier and lowpass circuits, but there was some odd behaviour. Just as a quick experiment I changed the rectifier IC to LF353, and the circuit worked as intended.
I never tried the TL072 or others, perhaps ill revisit the circuit again on a spare minute.
I ended up with little reason why the NE5532 should work oddly in what was a textbook (read basic) circuit.
In the end I assumed the input overvoltage protection diode had a part to play, or something internal to the IC which didn't react well to the circuit.
I had 3 opamps to hand: TL072, NE5532 and LF353 ( well actually I had LM324 and NE5534 too but forgot I did). The rectifier section requires an opamp with a fairly good GBW, so I chose the 5532 rather than LF353. But these things aren't always straightforward it seems...
Initially I used 5532 on both rectifier and lowpass circuits, but there was some odd behaviour. Just as a quick experiment I changed the rectifier IC to LF353, and the circuit worked as intended.
I never tried the TL072 or others, perhaps ill revisit the circuit again on a spare minute.
I ended up with little reason why the NE5532 should work oddly in what was a textbook (read basic) circuit.
In the end I assumed the input overvoltage protection diode had a part to play, or something internal to the IC which didn't react well to the circuit.
If an opamp you have tried costs ten times as much as a 5532 then OF COURSE you will think it sounds a lot better. As has now been said often the desks are full of 5532. What I have found impressive is 24 bit remastered analog mixed on a tube mixer. There is a vendor in England who still makes all-tube mixers to order. High voltage jFETs have also been tried with good results in minimal complexity mixers.
Its not often Bipolar types like 5532, 833 etc. are drop-in replacements for a Bifet opamp like LF353, TLO72 etc. For starters, the input impedances and offset conditions are quite different. Even assuming the design had allowances, you would still expect some anomalies in operation.
Good job noSmoking! You actually did something rather than relying on rhetoric. As myself and likewise the designer of the famous Euphonics mix board, "try it and find out." You did what everyone should do.
While it may not be a conclusive result for some guys , it would be interesting to see what you thought of your opamp pairs ! It would be pointless for others to comment on what you think about them. They are at liberty to have different opinions !
ashok- Not exactly certain what you mean. Maybe this?
These days use the NJM2068 quite a bit it being very neutral in my circuits. The LF353 is pretty good also just behind the NJM2068 with the chance to have jFet inputs all that can bring. The LF411 is good for low drift and the AD711 sounds just like the LF353 only brighter. If one needs a brighter LF353 the AD711 is a good choice. This pretty well covers all I ever use. High speed op amps seem to never sound great with tizzy highs and often thin bass. Kind of what is good for the purpose is the purpose. The LF353 has incredible CMRR frequency response so it is very good at rejecting out of band garbage in the right circuit. It is quite a bit better than most op amps for that CMRR frequency response. For power op amps that TDA7293 seems to sound far better than it has any right to. I tend to look for neutral. Described another way, if the music seems to play more slowly, it is better more neutral. Exciting and attention grabbing gets really old causing ear fatigue very soon.
Would love to hear from NoSmoking to learn what he has found!!!
These days use the NJM2068 quite a bit it being very neutral in my circuits. The LF353 is pretty good also just behind the NJM2068 with the chance to have jFet inputs all that can bring. The LF411 is good for low drift and the AD711 sounds just like the LF353 only brighter. If one needs a brighter LF353 the AD711 is a good choice. This pretty well covers all I ever use. High speed op amps seem to never sound great with tizzy highs and often thin bass. Kind of what is good for the purpose is the purpose. The LF353 has incredible CMRR frequency response so it is very good at rejecting out of band garbage in the right circuit. It is quite a bit better than most op amps for that CMRR frequency response. For power op amps that TDA7293 seems to sound far better than it has any right to. I tend to look for neutral. Described another way, if the music seems to play more slowly, it is better more neutral. Exciting and attention grabbing gets really old causing ear fatigue very soon.
Would love to hear from NoSmoking to learn what he has found!!!
That's interesting. By contrast, the '5532 in particular seems to have a bit of a high-frequency common-mode problem, TI more so than Signetics/Philips. You can see that in Samuel Groner's measurements of common-mode distortion and high-frequency nonlinearity (non-inverting). TI's MC33078 actually did a fair bit better in that regard (too bad its output stage is badly in need of some Class A bias).
It's too bad that high-frequency CMRR isn't commonly documented for opamps. It could explain a thing or two.
The '5532's protection diodes have apparently upset the odd application. Apparently they also like being adequately decoupled, and directly driving capacitive loads at gains near unity is also rumored to be a bit of a stretch due to low gain margin.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.