I just looked at the TL071 datasheet. Keep in mind common mode voltage is limited by VCC and -VCC and it's ability to get close to either. If you need to get closer you need a rail to rail opamp. You can exceed the limits but it will not be linear output. In some cases damage may occur, depends on the internals.

The ability for opamps to get to the rails are limited by use of current mirrors and diode junctions that have vbe drops associated with them. Sometimes they are not symmetric

 

Hello,

Perhaps this opamp handbook might give you some more insight in to opamp parameters.

Bertus

 

Great resource. I know Sam has the Floyd book which has all that in it. I think the best way to learn about it is to build one. When things don’t work, which is challenging when we have 20+ transistors, it pushes you into understanding the circuit. It’s basically a differential going into a gain stage and output amp. Overly simplified but it’s what it is. Different front ends have different parameters. Compensation is important... etc.

 

'm not seeing that on the datasheet for the 380

Well that explains a lot... I just double-checked the National Semiconductor PDF. With my poor eyesight I didn't see the decimal in front of the 5!

 

Here's a good source explaining common mode voltage and the issues related to it... beyond our beloved Floyd... pehaps you've seen something like this on the scope.
https://www.planetanalog.com/are-you-violating-your-op-amps-input-common-mode-range/#

Cntl + scroll wheel zooms things in and it's my friend as my reading glasses get stronger...

 

Perhaps this opamp handbook might give you some more insight in to opamp parameters.

I actually have that one and the Analog Devices Handbook by Walt Jung and a few others that I haven't gotten to yet. Thx.

 

the Walt Jung one is awesome.

 

pehaps you've seen something like this

I quickly scanned it and that is exactly what I'm seeing. I looked over Floyd's Electronic Devices and it is pretty much a redo of his Electronic Circuit Fundamentals albeit a bit more in depth. I liked Floyd's hands on approach a bit more than Grob's but he missed the boat leaving out simultaneous linear equations and complex numbers that Grob used. So, I'm going next to Malvino's Electronic Principles which covers the same materials but with his spin on it for a fresh look. I also have the workbook and lab manual to go with it. For some reason, Pearson Publishing quit putting the answers to the exercises in the Floyd book and had them online but they are gone now as the book was released again in a new edition even though there is very little change from edition to edition. Really disappointing.

 

I can't find the link anymore, luckily I saved it... . PM me your email... its just over 10M, should be ok with g-mail.. his electronic fundamentals is used for an introductory analysis...electronic devices is all about semiconductors and practical uses. So just enough to be useful without going in depth into the physics. If you need a physics book, you need a physics book. Art of Electronics seem to go a little deeper at times with the caveat that it can be skipped, but I really enjoyed some of their labs.

Grob's chapter in AC analysis is gold.

 

Be aware that a TL07x, TL08x or some other opamps have the Opamp Phase Inversion problem where the output suddenly does to a volage as positive as it can if an input voltage becomes within a few volts from the negative supply (or within a few volts above ground when there is a single positive supply).

 

Opamp Phase Inversion problem

Thanks for the heads up. I noticed that in the PDF so I'll have to explore that a bit on the breadboard. Once I get some coffee in me and my eyes open...

 

It phase inversion is a problem (that is, you are driving the amp-op too hard), you might want to consider using the LM358. The TL071 shows much better noise figures, although in an audio application with the LM380, this wouldn't show. The LM358 is dirt cheap, and the THD (0.001%) and input noise figures (3uVpp) are good enough for your application.

Frankly, even if the output stage was the best discrete system available, I couldn't tell the difference between the TL071 and the LM358. You won't need the ofset compensation pins on the TL071, so you would probably be better off using both stages of the TL072 for distributed gain and improved bandwidth (although the bandwidth for a single stage is more than enough, even if the GWP is divided by 10).

 

The LM358 is dirt cheap, and the THD (0.001%) and input noise figures (3uVpp) are good enough for your application.

The lousy old LM358 is never used for audio because it has severe crossover distortion (its distortion is not spec'd), is so noisy that the noise level is not spec'd and it has a horrible slew rate that cuts frequencies above only 5kHz.

I makes crossover distortion because it was designed for low current therefore its output transistors are not biased into class-AB.
It sounds like a buzzer:

 

Aha! So that is what the second wave is. I had seen that and wondered what was causing it. Looked like something I had seen studying amps and using diodes to eliminate it. Didn't connect the dots on that one... So far, playing around with the TL071 it has been fine. One question about it though. On the TI PDF, it says for the input voltage

(3) Input pins are diode-clamped to the power-supply rails. Input signals that may swing more than 0.5 V beyond the supply rails must be current limited to 10 mA or less

36V input!?!?

also for and I do realize this is a variant but...

0.5V which is what the 741 is rated for, which the TL071 is a pin for pin replacement for.

I can only assume the 36V input is correct but...

 

Does it say 36V or above VCC- (minus)?

Notice on top they are talking about two sets:
1. All except the TL07xH
2. Only for the TL07xH

 

Does it say 36V or above VCC- (minus)?

Notice on top they are talking about two sets:
1. All except the TL07xH
2. Only for the TL07xH

I'm using the TL071CP DIP8 so it's the top one and it says VCC-, even if that is -Vcc that would be a +18V input using +/-18Vs. That sure sounds high is why I asked. I'm using +/-12Vcc playing around with them and pushed the Vin to +3V and it followed fine and didn't die so...
Just hoping for some clarification here... And it won't even swing from rail to rail so that sure seems high.

 

from VCC- to VCC+ it's 36 Vpp... that's how I read it. It's the absolute max... not really a practical number but for the opamp's survival rating so to speak. If you were doing a voltage follower, you can input something close to that and make it work. But as others pointed out becareful because I had forgotten about them flipping polarity which would throw it all out of phase.

Just burn one out for the heck of it... what's a lab without burning some components.

 

That's what it says. I just find it hard to believe. Well, I just pushed one do or fry to 14Vpp on +/-12Vs and it lived and followed without distortion so I guess I'm a believer although I'm still shaking my head... Although that was without a load...

 

My discrete opamp would fry due to max reverse emitter base voltage. I don't recall exact value and adding diodes would help. Not sure how they design silicon transistors that can take so much voltage and survive but they do.

 

The lousy old LM358 is never used for audio because it has severe crossover distortion (its distortion is not spec'd), is so noisy that the noise level is not spec'd and it has a horrible slew rate that cuts frequencies above only 5kHz.

I makes crossover distortion because it was designed for low current therefore its output transistors are not biased into class-AB.
It sounds like a buzzer:

That is a nasty distortion. I would guess that was taken under quite a bit of load. Is that correct? Anyway, how much load does the input of an LM380 represent?