BB vs TI

[living sounds]

TLE207x is the newer "improved" version of TL07x often suitable for drop in replacement.  When comparing the two,  differences were both measurable and audible.

Yes, and I usually put the TL072 back in.  ;D  The latter is one stable little op amp, the TLE2072 is not.

 

[JohnRoberts]

TLE207x is the newer "improved" version of TL07x often suitable for drop in replacement.  When comparing the two,  differences were both measurable and audible.

I hope it is better or why bother?

About 15 years ago working on my first generation drum tuner design I needed to run a couple op amps sections from the 9V battery. I figured it would be nice to use one of the new "improved" modern low current op amps. I don't recall the exact series (and don't want to remember). I forget exactly why it didn't work but I wasn't going to do multiple SMD prototypes to find out why (back then I wasn't building my own prototypes). To get the job done and move forward I punted and spec'd in a TL06x that worked. Not a great spec sheet performer, but a known entity and it got the job done. (My next generation tuner used a class D audio chip). 
=====
I am impressed by the modern technology that claims distortion so low that it can't even be measured with conventional test benches without using tricks (like running at high noise gain and then dividing the result by that Nx). This is not apples to apples but parts that good can't easily be measured directly.

I feel sorry for the guys trying to design the high end test equipment these days, not trivial (not that it ever was). 

JR 

 

[user 37518]

Which is why no decent design engineer would use a TL07x for a summing amp.  :

I actually used an NE5534 as mic preamp in my DJ mixer kit (1978) but that was intended for high Z mics. The NE5534 calculates out to a 9dB NF (noise figure) with a low Z mic, so again a poor choice for low Z mic preamps.

Phono preamps are less obvious (I have design too many phono preamps to list them all). If you inspect the Bode plot (open loop gain) of TL07x op amps, we will see a falling gain plot caused by the internal compensation pole. Then compare that to the RIAA EQ curve, RIAA is ballpark 40dB gain at 1kHz (60dB at 20Hz) but falls off to only 20dB gain at 20kHz, so actually usable (while I never have).  If you inspect how noise sources combine to influence noise floors in gear you will find it dominated by early gain stages. Of course high noise gain circuits like VE sum amps with a lot of inputs benefits from modern low ein devices. I can criticize weaknesses in both but back in the 70s they were a godsend for audio designers. These days we have more options. I am trying to offer perspective... adding a 3 uV noise stage in an audio path with existing noise floor already tens of dB higher noise will be barely measurable, and unlikely to be audible. I am too lazy to do the math but hint it involves square root of sum of squares.

JR

I agree that uncorrelated noise sources add in RMS fashion, and if both are the same level they sum to 3dB not 6dB, I am also aware that the first stage with the highest gain is the determining noise stage, however, my point was that the TL07x only works well in a very restricted set of conditions, i.e. do not exceed 20dB of gain, do not use it as a mic pre or summing amp, be careful when using it in non-inverting mode since it will display more common mode distortion than a 553x, when using it as a buffer it is easy that the input common mode range will be exceeded and the dreaded output inversion will ocurr, do not drive low loads or distortion will rise, so a lot of precautions need to be taken care of and I think that the benefit is not worth it, which is why I mentioned that nowadays they are mostly relegated to a DC servo or meter buffer stage. That being said, back in the day, good designers such as yourself, managed the technical difficulties and made great sounding equipment with less than ideal parts like the TL07x.

Now onto phono preamps, the TL072 may have a convenient gain response but the noise will be around 10dB higher than if using a 5534, is that relevant to the ear? I dont know, but I don't like how it looks on paper...

 

[user 37518]

TLE207x is the newer "improved" version of TL07x often suitable for drop in replacement.  When comparing the two,  differences were both measurable and audible.

I don't see the real improvement, noise is not stellar (18nV/rtHz of the TL072 VS 12nV/rtHz typ and 17nV/rtHz max for the TLE2072), slew rate is higher, but in my opinion, slew rate was never the issue with the TL072. The only real improvement I see is a higher unity gain bandwidth. Is that worth paying 5 times the price of a TL072? I don't think so... I much rather go an entirely different way and get something like an OPA1678.

 

[bluebird]

another brewery thread off in many directions when all I was doing was making a comment about BB opamps on evilly for about 15 dollars VS brand new BB/TI opamps of the same model for about 5 dollars and in the same quantities....

Lol, yeah but at least its not about politics and we get to have some good tech nuggets thrown our way! whats cool about this forum, is even when someone misunderstands a question, there's always something more to learn.

P.S.
I always liked TL072's in distortion pedals...

 

[JohnRoberts]

I agree that uncorrelated noise sources add in RMS fashion, and if both are the same level they sum to 3dB not 6dB,

but who is talking about two equal noise sources? How about 3uV added to 30-50uV noise floor, or several 3uV sources to an even larger noise floor. Same result... insignificant.

I am also aware that the first stage with the highest gain is the determining noise stage, however, my point was that the TL07x only works well in a very restricted set of conditions, i.e. do not exceed 20dB of gain, do not use it as a mic pre or summing amp, be careful when using it in non-inverting mode since it will display more common mode distortion than a 553x, when using it as a buffer it is easy that the input common mode range will be exceeded and the dreaded output inversion will ocurr, do not drive low loads or distortion will rise, so a lot of precautions need to be taken care of and I think that the benefit is not worth it, which is why I mentioned that nowadays they are mostly relegated to a DC servo or meter buffer stage. That being said, back in the day, good designers such as yourself, managed the technical difficulties and made great sounding equipment with less than ideal parts like the TL07x.

designing electronics was not easy 40-50 years ago... today you guys have it much easier.

Now onto phono preamps, the TL072 may have a convenient gain response but the noise will be around 10dB higher than if using a 5534, is that relevant to the ear? I dont know, but I don't like how it looks on paper...

I haven't done the math on that, in general I wouldn't use a 553x for a phono preamp because of it's higher input noise current. While a phono cart may be only a few K ohms resistive, the inductive components means higher resistance at HF, for even more IxR noise. Bifets have vanishingly small current noise.

BUT TO REPEAT I never used a TL07x or 553x for a phono preamp, but I have used worse.  The old National Semi LM381/LM387 was bipolar and not even unity gain stable so when I used it back in 1978 for my DJ mixer kit I tricked the op amp into thinking it was operating at higher closed loop gain gain to make the RIAA curve more accurate. IMO my design didn't suck for 40+ years ago but I have done multiple much better phono preamps since then.

I am not saying to use the TL07x or NE553x in every socket today, but if used properly they probably still wouldn't suck today.

JR 

 

[john12ax7]

Yes, and I usually put the TL072 back in.  ;D  The latter is one stable little op amp, the TLE2072 is not.

Was it just stability or did you also prefer the sound of the TL072? I didn't see any stability issues with the TLE2072, noise and distortion were both lower,  and subjectively it sounded clearer / closer to the source. But I can see that for some instruments and applications TL072 might be a better match.

Though If current isn't an issue Ild take OPAx134 over either.

 

[living sounds]

Was it just stability or did you also prefer the sound of the TL072? I didn't see any stability issues with the TLE2072, noise and distortion were both lower,  and subjectively it sounded clearer / closer to the source. But I can see that for some instruments and applications TL072 might be a better match.

Though If current isn't an issue Ild take OPAx134 over either.

Didn't like the sound no matter where I tried it. But I've seen it used in commercial designs that sound great, synths mostly though.

 

[JohnRoberts]

Didn't like the sound no matter where I tried it. But I've seen it used in commercial designs that sound great, synths mostly though.

You appear to be very critical of how devices sound. Have you ever tried to identify what you are hearing?

Back in the 70s I rolled some of my own test equipment (two-tome IMD @ 19-20kHz) to help make objective measurements to discern subtle differences.

JR

 

[user 37518]

Didn't like the sound no matter where I tried it. But I've seen it used in commercial designs that sound great, synths mostly though.

What exactly you did not like? I find it hard to distinguish between low distortion amps, most of the time is missuse of the part like unwanted oscillation that makes the user believe there is a difference.