Can a summing amplifier work with a TL072 op amp

[Hackpenguin]

I'm having problems with using a summing amplifier. I don't know if it's the chip or needs a ground that is both positive and negative. I'm using a jfet input OP amp, I think it doesn't need to be a jfet input,but that is the chip I have. Please someone help me

 

[MikeMl]

If the input signals are ground (0V) referenced, then the amp will have to be powered with both positive and negative (dual) supplies.

 

[RadioRon]

The most common error in wiring summing amplifiers is in the power supply connections. First you must decide if you will use a single supply voltage or a dual supply (ie. positive and negative voltages). The dual supply allows a simpler op-amp circuit and so is often used. What power supply do you intend to use?

 

[atferrari]

Could you show the problematic circuit?

 

[AnalogKid]

I'm having problems with using a summing amplifier. I don't know if it's the chip or needs a ground that is both positive and negative. I'm using a jfet input OP amp, I think it doesn't need to be a jfet input,but that is the chip I have. Please someone help me

Please someone post a schematic. Seriously, I'm not trying to be grouchy or anything, but this is an electronics forum. We don't get a glass of water without a schematic. Round numbers, there are many different "summing" circuits, and *thousands* of different opamps. Which one is yours?

ak

 

[dr pepper]

You can usually get an op amp to work single supply, but you need to bias the i/ps correctly.
As just mentioned we need more info or a schematic.

 

[audioguru]

The TL072 is a dual audio opamp with Jfet inputs. Like any opamp, it works with a dual polarity supply if its (+) input is biased near 0V and it also works with a single positive supply if its (+) input is biased near half the supply voltage and the input, output and feedback grounded resistors have capacitor coupling.

A summing circuit is usually an inverting opamp circuit.

 

[spec]

I'm having problems with using a summing amplifier. I don't know if it's the chip or needs a ground that is both positive and negative. I'm using a jfet input OP amp, I think it doesn't need to be a jfet input, but that is the chip I have. Please someone help me

The TL072 is basically a low noise HiFi audio amplifier and would not be the optimum choice for a summing amplifier, but it would be adequate, unless high precision is required. A much more suitable opamp would be from the OPA192 family of opamps. https://www.ti.com/lit/ds/symlink/opa192.pdf or the related cheaper, but less accurate, OPA197 family. https://www.ti.com/lit/ds/symlink/opa197.pdf

The input voltage range of the TL072 is -Vcc +4V to Vcc -4V (-Vcc can range from -5V to -15V, while Vcc can range from 5V to 15V).

Similarly, the TL072 output voltage does not swing from -Vcc to Vcc

Because of this, it is difficult to use the TL072 as a traditional, DC coupled, inverting virtual earth summing amplifier with a single supply line.

The simplest solution is to generate a second supply line using an inverting charge pump power converter. **broken link removed**

The TL072 does have the major advantage of FET inputs which means, for practical considerations, the inputs do not take any bias current, thus eliminating voltage errors caused by the input bias current flowing through the input and feedback resistors. https://www.ti.com/lit/ds/symlink/tl074a.pdf

Some opamps, like the OPA192 family for example, have an input voltage range from -Vcc to Vcc and also an output voltage range from -Vcc to Vcc. This is known as Rail to Rail Input/Output (RRIO).

RRIO opamps can be used as inverting virtual earth summing amplifiers, with a single supply line, under the following conditions:

(1) -Vcc = 0V, Vcc = supply voltage. Input voltage = 0V to minus voltages
(2) -Vcc = supply voltage, Vcc= 0V. Input voltage = 0V to positive voltages

spec

**broken link removed**
**broken link removed**​

 

[ccurtis]

One common newbie mistake with op amp summing circuits is leaving an input open, thinking he is setting it to zero volts. It must be connected to zero volts (ground).

 

[spec]

One common newbie mistake with op amp summing circuits is leaving an input open, thinking he is setting it to zero volts. It must be connected to zero volts (ground).

Hmm, not to sure about that. From a theory point of view it is fine to leave a summing input open circuit but it depends on the circuit physical arrangement because an open input may be susceptible to pick up.

spec

 

[ChrisP58]

Hmm, not to sure about that. From a theory point of view it is fine to leave a summing input open circuit but it depends on the circuit physical arrangement because an open input may be susceptible to pick up.

spec

Even with the theoretical 'ideal' op-amp, there is no mechanism that will set an unconnected input pin to any particular DC value. And very unlikely that it would somehow end up at the midpoint of the two supply rails, which is where you usually want the non-inverting input of the classic summing amplifier to be.

Any real op-amp will have a non zero input bias current, which will tend to move an unconnected input to (or near) one of the supply rails.

 

[spec]

Even with the theoretical 'ideal' op-amp, there is no mechanism that will set an unconnected input pin to any particular DC value.

Any opamp, ideal or not, configured as an inverting summing amplifier will not be affected by an open circuit resistor connected to its inverting input node. The unconnected end of the resistor will be the same voltage as the inverting input of the opamp ie earthy.

And very unlikely that it would somehow end up at the midpoint of the two supply rails, which is where you usually want the non-inverting input of the classic summing amplifier to be.

See above

Any real op-amp will have a non zero input bias current, which will tend to move an unconnected input to (or near) one of the supply rails.

If you are describing a classic inverting summing amplifier, as shown in post #8, the inverting input of the opamp is connected to the opamp output and the non-inverting input of the opamp is connected to 0V, so the opamp has no unconnected inputs.

Just a general point: all nodes on any circuit are connected to air, unless the circuit is operating in a vacuum that is, and a resistor with one end open circuit is the same as air; it has no effect on any circuit including an opamp, ideal or not.

spec

 

[ccurtis]

Proper summing requires that current flows between the input voltages through the input resistors and the feedback resistor. Current cannot flow through an input resistor that is open at one end.

 

[spec]

Proper summing requires that current flows between the input voltages through the input resistors and the feedback resistor. Current cannot flow through an input resistor that is open at one end.

Not clear what proper summing means.

Not sure what current you are referring to as there is no current flowing through a resistor with one end connected to air.

Also not sure if you are agreeing or disagreeing with what I said in post 12.

Please clarify.

spec

 

[ChrisP58]

Sorry spec, I'm obviously not making my point clear. Let me try again.

My interpretation of ccurtis' statement in post #9 was that a common newbie mistake was leaving [the non-inverting] input open. ie. not connected to ground or to anything other than itself. Thereby creating a floating input that could, in an ideal(1) op-amp, be at any DC voltage between -∞ and +∞. But in common real world op-amp, will be somewhere between it's two supply rails. And will generally, due to a non-zero bias current, drift towards one of the supply rails. It will not, find it's way to any voltage 'earthy' unless there is some external DC path that encourages it to be at some user selected point.

I was not, and I don't think that ccurtis was, referring to an unconnected input on one of the resistors feeding the summing junction at the inverting pin. Which, as you correctly pointed out, is not floating, due to the gain setting resistor connecting to the output.

I made a similar mistake many years ago in an audio circuit when I AC coupled a signal into the non-inverting input of an op-amp with a capacitor. My power rails were +12 and -12. At power up, the amp worked, but within a few seconds, the ouput of the op-amp went to about 10 volts (I don't remember the polarity) with no whisper of the AC signal that I was feeding into the input capacitor. The problem that I had created was to have no DC path to ground to bleed the bias current off. So, the bias current from the op-amp was charging my input capacitor and creating a non-ground DC level on the op-amp input pin. The solution, of course, was a 10k resistor to ground connected to the net between the cap and the op-amp, thereby establishing a high impedance DC path that kept the net nominally at ground.

 

[spec]

Ah right, got it.

Yes, you can't leave any input of an opamp floating, although I'm sure we have all done that.

One problem you can encounter with a virtual earth inverting amplifier is that if the opamp is not unity gain compensated it will oscillate if the input is not referred to 0V.

spec

 

[ccurtis]

Spec, I agree with you for the special case of the summing circuit where the non-inverting input is grounded. In that case, leaving an input open is equivalent to applying a zero voltage to the input since the summing junction is held at zero volts anyway. For the general case of a summing circuit, where the non-inverting input is also fed with an input voltage(s) to be included with other inputs to be summed, leaving an input open is not equivalent to applying zero volts to the input.