Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

opamps and differential amplifiers

Status
Not open for further replies.

samcheetah

New Member
why are single-ended op amps more famous than fully-differential amplifiers. fully differential amplifiers have double the gain as compared to single-ended op amps. then why are all the books filled with single-ended opamp explanations and applications. what was the reason for dropping out that double-ended output "creature" from the scene. what does the op amp offer that a differential amplifier doesnt offer????
 
If the input signal is single-ended, a differential amplifier circuit offers little in the way of performance (and more in parts and complexity) for simple one or two stage amplifier circuits. If the input is differential (e.g., a typical professional dynamic microphone), the diffamp is the only way to go, offering all kinds of common-mode rejection.

Dean
 
If I understand Samcheetah, he is asking about differential output amplifiers. Differential output op amps are in fact available, but they have only become available in IC form in the last few years.
Op amps have differential inputs because of the feedback configurations that become available, and because they allow one to do addition and subtraction, inversion, etc. Differential outputs are not generally needed, but are very useful for driving some of the newer A/D converters, for driving speakers, and for driving differential signal lines such as twisted pairs.
Differential amplifier designs have been available forever (almost), but true differential op amps are relatively new developments.
 
yes Ron_H you understood me. im sorry i should have said "single ended output opamps" instead of just "single ended opamps".

i know that they have only recently become available in the IC form. and thats mainly because the newer high speed ADCs use differential inputs. for that u need a differential output opamp before it. or if u have a single ended output opamp u need a transformer (i dont remember the configuration).

but my question is that why were differential output op amps abandoned from being developed in the IC form in the first place. differential output opamps have double the gain of single ended output opamps.

i think ive made my question clear now (i hope :lol: )
 
samcheetah said:
but my question is that why were differential output op amps abandoned from being developed in the IC form in the first place. differential output opamps have double the gain of single ended output opamps.

Probably because there wasn't any use for them?, the gain is irrelevent really, it simply has to be large enough to ignore in your feedback calculations - as all opamps are.

You've mentioned one potential use, and that's for a very new specific application, apart from that they don't seem much use. I suppose they could be useful for balanced outputs?, but that's very easily done using a simple double opamp circuit.

Operational amplifiers were designed as the active elements for analogue computers (before transistors), you actually built a circuit to do the calculations you required. I don't think a differential output would have been any use?.
 
Once the input signal hits the first FET, the vertical amplifiers of oscilloscopes are differential all the way to the deflection plates.

Another excellent application for these little puppies would be in mixers, PA amplifiers, etc. where there is a differential input. Since most of them require a phase-splitter somewhere before the output amplifier, keeping the signal differential all the way to the end would have lots of advantages.

The output of ECL is differential but not vary darned operational!

Dean
 
Dean Huster said:
Another excellent application for these little puppies would be in mixers, PA amplifiers, etc. where there is a differential input. Since most of them require a phase-splitter somewhere before the output amplifier, keeping the signal differential all the way to the end would have lots of advantages.

And lots of disadvantages - every pot and switch would need to be doubled!.
 
samcheetah said:
i know that they have only recently become available in the IC form. and thats mainly because the newer high speed ADCs use differential inputs. for that u need a differential output opamp before it. or if u have a single ended output opamp u need a transformer (i dont remember the configuration).

but my question is that why were differential output op amps abandoned from being developed in the IC form in the first place. differential output opamps have double the gain of single ended output opamps.

i think ive made my question clear now (i hope :lol: )

Cheap stereos use output stages with opposing swings on the outputs. Inverters often switch both wires of a transformer. But this creates an output which doesn't reference to ground. Unless you're driving ground-isolated components like a speaker or transformer, it is difficult to make circuits to handle differential signals and utterly pointless since you have no real need for a higher signal voltage. It's not really lower signal-to-noise ratio either, each wire gets the noise of the various components so generally you'd get twice the noise.

High speed digital logic often uses differential signals, although for somewhat different reasons than the justifications for analog.

High speed diff ADCs are useful when you've got a diff signal. It happens. I did a design for a thermocouple which had the thermocouple junction itself touching ground, so you can't just attach one terminal to ground and read the voltage off the other. I'm also working with strain gauges which internally resemble a Whetstone Bridge of variable resistors. The 2 output terminals are a differential voltage, the voltage of either one to ground is meaningless. Thus you NEED a differential ADC, otherwise you'd need to take an ADC reading for each one and mathematically subtract to find the difference. This is not only expensive and complicated, it compounds the small errors in the ADC.
 
I'm also working with strain gauges which internally resemble a Whetstone Bridge of variable resistors. The 2 output terminals are a differential voltage, the voltage of either one to ground is meaningless. Thus you NEED a differential ADC, otherwise you'd need to take an ADC reading for each one and mathematically subtract to find the difference. This is not only expensive and complicated, it compounds the small errors in the ADC.
Or you could use an instrumentation amplifier, which has differential inputs, single-ended output, and either fixed or programmable gain.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top