[SOLVED] non inverting amplifier input current

[brain-dead]

Hi Everyone,

I am very confused with op amp input current. I have been looking over various op amp datasheets and I have noticed in the absolute ratings a requirement for input current.

I have built a simple non inverting mic pre amp using a feedback / gain ratio of 10K / 1K then capacitively coupled into a non inverting integrator with a gain of 2, 1K / 1K in parallell with 47nF to remove noise from the MIC.

Mic Signal ---> 1uf / 22K high pass ---> Non inverting x 11 gain --> 1uf / 22K high pass --> Integrator

My confusion is that if the input impedance of the op amp is so high, then I have always assumed no current will flow into the input terminal. So in my case, the current would flow through the 1uf capacitor and develop the input voltage over the 22K resistor, which would be a small current.

I read something about the op amp sourcing current if the ESD circuitry is reverse biased, however I cannot find out anywhere why this is and how to possibly avoid it

As an experiment I bought a range of op amps and decided to see which one sounded the best, I decided to go for the MCP624 as it sounded the cleanest and most responsive.

I have no series input resistors directly proceeding the input non inverting terminal, therefore I have op amp -> high pass -> op amp. Would this cause any current to be sunk or sourced by the input terminals? I am finding it very confusing as if the input impedance is so high then the current should be negligible???

I have Vdd at 5 volts. So for my calculation I am making the following assumptions:

The high pass 1uF cap is considered a short, the max input voltage to the integrator is 5v, therefore this voltage will develop across the 22K resistor:

5 / 22000 = 0.2mA

So it seems to me that I do not need a series input resistor???

Any help would greatly appreciated, I have built the circuit and it works great, I just don't understand how the input would sink or source current.

 

[chuckey]

From the MCP624 data sheet, the input current is less then 5 pico amps, so the current flowing throught he leakage of the input capacitor, would be enough to provide an earth. (**broken link removed** ). Did you use the auto zero function?
Frank

 

[crutschow]

From the MCP624 data sheet, the input current is less then 5 pico amps, so the current flowing throught he leakage of the input capacitor, would be enough to provide an earth.
.............................

Never depend upon the leakage through a capacitor to provide input bias to an op amp. A good non-electrolytic capacitor can (and likely will) have lower leakage than the op amp and it would be biased incorrectly. You should always use a bias resistor if coupling with a capacitor.

 

[brain-dead]

Hi Frank,

I am not using the model with the autozero function. It is the quad version.

- - - Updated - - -

Thanks zapper, I will post a diagram of how the circuit looks at the moment. How should the bias resistor be arranged in my circuit?

 

[crutschow]

Post the circuit and I'll look at it.

 

[brain-dead]

Hi Guys,

Thanks for all your help. After your input and researching online, I have found out what they mean by maximum input current. For anyone who wants to know it only seems to matter when your circuit board has an ESD charge applied to it. The internal diodes then clamp to stop damage to the op amp, the result of this is that current can flow into or out of the inputs through the diodes, therefore you must limit this current can flow, basically by sticking a resistor infront of the input terminal. I have left it out and all seems fine.

: )

Thanks