Stoopid Q About Inverting Opamps

[Paul Marossy]

I'm fiddling around with a circuit someone else designed that uses an inverting opamp. When I go online and read about inverting opamps, it always says that the positive pin is connected to ground. However, in the circuit I am toying around with those inputs are connected to Vref instead. What is the purpose of that? Is that what has to be done because it's operating on a non-bipolar power supply? Seems like that would be the case as that would be the midway point between 9V and ground.

So I guess all these tutorials and what are not assume that a bipolar power supply is being used?

I know this is probably a very elementary question, but I never really thought about it before... 

[Gurner]

When I go online and read about inverting opamps, it always says that the positive pin is connected to ground. However, in the circuit I am toying around with those inputs are connected to Vref instead. What is the purpose of that? Is that what has to be done because it's operating on a non-bipolar power supply?

Yes. (whatever voltage you connect the +ve pin, will be DC level your AC signal sits on     ...for a single supply you'd want that to be halfway between your power supply rail & ground, (so 4.5V for a 9V supply)

So I guess all these tutorials and what are not assume that a bipolar power supply is being used?

Yes - by & large - they do.

[Paul Marossy]

Thanks for the clarification on that. My realm of experience with opamps is kinda small, and even smaller with inverting opamps. I've never even used an opamp with a bipolar power supply! Stompboxes don't need them... 

[amptramp]

When an inverting op amp is operating in the linear mode, it will drive the inverting input voltage to nearly the same as the non-inverting voltage.  There is an input common-mode range for op amps which represents the input voltages where the op amp can be made to operate in the linear mode.  Usually, this is a few volts below the positive supply to a few volts above the negative supply.  If you have bipolar supplies,, zero volts is usually within the common-mode range.  If you have a unipolar supply, the non-inverting input is usually referenced to a voltage between ground and the other supply, usually  Vref halfway between.  There are rail-to-rail input op amps which will operate anywhere between both power voltages, but you should be aware that there is a difference between rail-to-rail input and rail-to-rail output.  Some amplifiers can still operate with inputs beyond one rail but usually not by much and usually not in both directions and some go to one rail but not the other.

The common mode range exists because it is necessary to drive an input device in the op amp which will be a transistor or FET.  A bipolar transistor needs at least a diode drop beyond the emitter voltage to turn the transistor on.  FETs may be different - a depletion-mode FET (such as a JFET) can operate with the gate voltage beyond the source voltage - lower for n-channel and higher for p-channel.  A MOSFET depends on the doping to provide the voltage range at which the input operates.

Some op amps can go beyond the input range and recover without incident when the voltage returns to within the range.  Some, like the common TL071 series will invert the output due to complicated and unintended interactions with the substrate, but will revert to normal once the inputs go back to normal.  A few will destroy the amplifier if they go beyond certain limits within the power supply range.

[Paul Marossy]

Some op amps can go beyond the input range and recover without incident when the voltage returns to within the range.  Some, like the common TL071 series will invert the output due to complicated and unintended interactions with the substrate, but will revert to normal once the inputs go back to normal.  A few will destroy the amplifier if they go beyond certain limits within the power supply range.

I recently had that pleasure with a TL072. 
But it was a learning experience too. 

So is an inverting opamp scenario any more susceptible to noise (like hum/buzz) than a non-inverting opamp?

[R.G.]

So is an inverting opamp scenario any more susceptible to noise (like hum/buzz) than a non-inverting opamp?

No. In fact, the opposite is true. There is (was, I guess; lots of those guys are retired or dead now) a maxim in the heavy opamp-using community that went "Always invert - unless you just can't."

[Electron Tornado]

There is (was, I guess; lots of those guys are retired or dead now) a maxim in the heavy opamp-using community that went "Always invert - unless you just can't."

What are the pros and cons of using inverted vs non-inverted?

[amptramp]

Non-inverted gives you a high input impedance and independence of the input from feedback characteristics.

Inverted gives the possibility of a gain less than one, ability to set the input common mode voltage to a predetermined value, the ability of connecting multiple inputs as a summing junction and having them behave independently and feedback from stray capacitance that stabilizes rather than destabilizes the stage.  A lot of filter stages only work in the inverted connection.

[Paul Marossy]

Thanks for the info guys!