operational amplifiers

[keatescarn]

Hi I need help with a question. does anyone have any ideas.

any help would be great
thanks

Scenario
With the development of cheap and reliable operational amplifiers the need to design amplifier circuits has disappeared. However it is still necessary to utilise these devices in a range of standard configurations. When doing this it is important to use the manufacturers’ data sheets to identify important parameters so that an appropriate device may be selected from the huge range available.

Task 1: P2.1, P2.4
a. An inverting amplifier is required to amplify the ac signal from a transducer before it is applied to an ADC (Analogue-to-Digital Converter). The specification is given below:

Signal Source Vrms = 700 mV @ 0 Hz < frequency < 20 kHz. Rout = 100 Ω
The maximum acceptable error in the signal voltage must be less than 1%.

ADC Voltage input range  5 V. Rin = 8 kΩ

Amplifier 741 connected in an inverting configuration.

Analyse the circuit to determine an acceptable design. You should use the design methods and equivalent circuits, in conjunction with the manufacturers’ data sheets. The design should include the voltage gain, input resistance and bandwidth of the circuit. The various errors due to bias current etc should also be considered.


Modify the above circuit so the voltage gain is 10.

 

[audioguru]

A lousy old 741 opamp is 44 years old. It is noisy and has trouble with frequencies above 9kHz because its slew rate is very low.
It is spec'd to use only a plus and minus 15V supply.
There are many better opamps available.

 

[Grossel]

A lousy old 741 opamp is 44 years old. It is noisy and has trouble with frequencies above 9kHz because its slew rate is very low.
It is spec'd to use only a plus and minus 15V supply.
There are many better opamps available.

I always hears how bad the 741 is. Is there some worse known opamps then?

@keatescarn
Hint: what is the highest possible slew rate that you expect from the amplifier output?

 

[audioguru]

I always hears how bad the 741 is. Is there some worse known opamps then?

A lousy old LM324 quad and LM358 dual opamps are also pretty bad. They are low power so their output transistors do not have enough bias current which produces crossover distortion.
The low current also reduces the slew rate so they have trouble above only 2kHz.
They are very noisy.

 

[simonbramble]

i actually made a random number generator out of an LM324 op amp. the original circuit had a Zener at the front, to generate the noise... until I realised the noise of the op amp was just as bad as the Zener.... It worked extremely well and the numbers were completely random

 

[simonbramble]

I think the person who set you this question does not know what they are talking about. Calling up the datasheet of the 741, it has a worst case input offset voltage at ambient of 6mV. This adds a 6mV error to your 700mV signal, which is 0.86% before you start considering errors in resistor values plus input bias current x input resistor errors and quantisation errors of the ADC. As with the posts above, I would choose an op amp with lower input offset voltage. It will be lower in cost than choosing 0.1% resistor values to make up for the loss in accuracy incurred by using a crappy op amp in the first place

 

[unclejed613]

741's also have high input bias currents which can cause additional offset errors. this looks like a homework assignment, so the OP is probably stuck with using a 741. being that it's a homework assignment, compensating for all the errors is probably part of the "fun".

 

[WTP Pepper]

Although I agree there are better op amps out there for this application, the 741 will happily handle signals above 9KHz depending on the gain and that theory applies to all op amps. They also can use less than +/-15V and can even use a virtual grounded single supply if you know what you are doing.
I would choose a LMV651 over a 741.

 

[MrAl]

I always hears how bad the 741 is. Is there some worse known opamps then?

@keatescarn
Hint: what is the highest possible slew rate that you expect from the amplifier output?

Back in its day the 741 was a much sought after op amp and made it into some very high quality designs. Things have changed exponentially since that era however, with better and more specialized designs hitting the market. People forget about the past because most of the stuff we have today is so much better. If you take a quick look at some of the more recent op amps you'll see much better specs and that will at some point make you feel like putting down the oldies too.
People still put down the LM358 too, but it comes in handy sometimes and is quite cheap. Of course the specs have to fit the application, and sometimes the LM358 does and sometimes it doesnt. For example, i would not recommend an LM358 for low level audio applications.

 

[Jugurtha]

Whenever you see 741, you know it's academia.

Teachers learn from books where the 741 is mentioned, they teach the 741, they get the 741 in the lab. And people who write books do write about stuff that is still going on in universities. It's a cycle.

For teaching purposes (academia, basic lab assignements) I think it's a good chip. But for practical stuff...

I think it's a grat chip to highlight the differences between an ideal op-amp and a real op-amp. I mean, offset voltage , slew rate, limited bandwidth, finite gain.. I think it's good to show look how messy the world is.

 

[audioguru]

the 741 will happily handle signals above 9KHz depending on the gain and that theory applies to all op amps.

What about slew rate? With a 30V supply, it is the very low slew rate of the output of a lousy old 741 opamp that limits its frequency to 9kHz, not just its gain.
With a gain of 1, its bandwidth is 1MHz!
But its output slew rate limits its output to 9kHz at high levels. When its frequency is higher than 9kHz then its output cannot move fast enough. Above 9kHz at 28V p-p its output becomes slowly slewing triangle waves.

They also can use less than +/-15V

741 opamps are spec'd ONLY with a 30V supply. Most opamps have a supply voltage RANGE listed and are spec'd at a few supply voltages. Some 741 opamps do not work when the total supply is 10V or less.

 

[unclejed613]

looking at the data sheet, a gain of 10 at 20khz is well within the GBW curve. however with 700mV of input signal, slew rate limiting will be causing some distortion,
maybe the spec on a 741 data sheet needs to be re-named "Slow Rate"

 

[ADWSystems]

So if the LM741 and 358 are so blah. What is your go to general purpose Op Amp? (and why)

 

[audioguru]

I use TL071 single, TL072 dual and TL074 quad low noise, low distortion and wideband opamps for most things. Their full output bandwidth is 100kHz. Their power supply is 7V to 36V.

I use MC33171 single, MC33172 dual and MC33174 quad opamps for low voltage and low power and replacements for the lousy old LM358 and LM324. Their full output bandwidth is 35kHz. Their power supply is 3V to 44V.

I use MC34071 single, MC34072 dual and MC34074 quad opamps for low voltage, high voltage (up to a 44V supply) and normal power. Their full output bandwidth is 100kHz. Their power supply is 3V to 44V.

 

[unclejed613]

i use TL071, TL072, or TL074 (single,dual, quad) for most applications. the National Semiconductor LF353 is almost identical in performance to the TL072. i use TL062 and TL064 for battery power applications like guitar effects boxes. for applications requiring higher bandwidth TI has come up with a new higher performance series TLE2072, which is essentially an upgraded version of the TL072. the TL07x series of op amps are low noise, have a GBW of 3Mhz, have very low input offset voltages, and very low input bias currents, as AG mentioned, they have a full output bandwidth of 100khz. and they can operate at up to +/-18V rail voltages. and previously in this thread, we were talking about slew rate.... the slew rate of the TL07x series is 13V/μS compared to the 741's 0.5V/μS. and the supply current is 1.4mA/amplifier which is a lot lower than the 8mA for NE5532 op amps (which means you can operate a lot more TL072's on a 1A supply than you can NE5532's).

 

[ADWSystems]

Most of the designs I'm working are slow so bandwidth is not generally an issue. But the designs tend to run from a single power supply and close to the rails. What is the available input voltage range for your "go to" OpAmps? They don't appear to be rail to rail. What are your thoughts on the LM2904?

 

[audioguru]

The TL07x and TL08x opamps have Jfet inputs. They have a problem called "phase inversion" where the output suddenly goes high if an input voltage gets within a few volts from the negative supply voltage.

An LM2904 is a dual opamp the same as a lousy old LM358 except it can withstand -40 degrees C to +85 degrees C.
Its inputs work at ground and its output goes down almost to ground so a negative supply is usually not needed.
It uses a low supply current, is very slow, has crossover distortion and is very noisy.

 

[ADWSystems]

I also found a source for the MC33172 which you reference. Is that rail-to-rail?

But if I'm already dealing with a slow signal (and I mean slow), why wouldn't the LM358/2904 would be a good fit?

Maybe I should just ask it this way. If you needed an opamp to run from 0-5V or 0-12V an be rail-to-rail to amplify (x10) very slow signal and was in expensive. What chips would you grab for first?

P.S. This is that part of the job I hate most. There are so many chips with so many nuianses, it's hard to find the right (and cost effective aka. cheap) chip.

 

[audioguru]

An MC33172 dual opamp has the same low power of an LM358 dual opamp but does not have crossover distortion and has a much higher slew rate so it works up to 35kHz. They both work from a supply as low as 3.0V and both have inputs that work at 0V without a negative supply. The output low of the MC33172 goes as low as about 0.08V and as high as 0.7V less than its supply voltage with no load current. The output high of an LM358 goes as high as 1.2V less than its supply voltage with no load current.

I have never needed and never used a rail-to-rail opamp so I can't recommend one.

 

[ADWSystems]

Sorry to dredge upan old thread but my question is on the same topic. (Single supply preferred, nearly rail-to-rail input and output)


I was able to succesfully use the MC 33072 and everyone is happy. Next project similar design. So I went to order up some more MC33072 and our discount suppliers are out of stock, but they do have the MC34072. I scanned through the datasheets but cannot see any significant differences. I see a wider output range +/-14.2 vs +/-14.7 but not much else. What differences between the MC33172, MC33072, and MC34072 should I be noting.