OPAMP feedback resistors

Thread Starter

Xispas

Joined Feb 22, 2014
1
Hi the am new here so first i'd like to say hello every one.
I am student of HNC in electronics and i have a questions for an assignment which am not sure what to answer in suming inverting amplifier, why the feedback resistor have to be not too big not too small?
Thank you very much for any help.
 

MrChips

Joined Oct 2, 2009
30,824
Welcome to AAC. I have moved your post to a new thread. The original thread belongs to the original poster (OP). To ask a new question on someone else's thread is called hijacking and is not allowed.

Since this is for an assignment the proper forum to ask this question is "Homework Help". On this forum you are expected to show what you already know about the subject matter.
 

ericgibbs

Joined Jan 29, 2010
18,874
hi X,
For starters.
Considering higher values of Rin and Rfb, look at the OPA datasheets for OPA input bias currents.
For low values consider the loading of the source that drives the OPA inputs.

There are other factors which govern the optimum choice of values, it depends upon the application.

Do you have a practical example of OPA circuit you could post.?

E
 

LvW

Joined Jun 13, 2013
1,761
It is the main advantage of opamp circuits with feedback that the gain primarily is determined by the external components only.
However, this is only possible if all non-ideal properties of the opamp can be neglected.
That means: During gain calculation we assume the input resistances to be infinite (no signal currents into the amplifier) and the output impedance to be zero.
This leads to the simple formulas (A,inv=-R2/R1 and A,noninv.=1+R2/R1.

As a consequence the resulting error can be kept in acceptable limits only if the external componenets allow such a simplification.
Hence, the resistors used should be large if compared with the real output impedance and small if compared with the real input impedance.
Thus, as a rule of thumb you should try to restrict the resistance range to values between app. (500...500k)Ohms.
 

ErnieM

Joined Apr 24, 2011
8,377
Well... the short answers is "it depends." Once while working on an amp requiring "as much gain as possible" (we settled on about 31) at 3MHz I learned all about input pin capacitance.

Seems each pin on a package actually has capacitance! And the input resistors to an op amp form a RC circuit with that cap limiting the bandwidth.

So... long story short, to get sufficient gain at 30MHz I needed to drop the input resistor to 50 ohms. My source had no trouble driving this so it all played together, but the voltage amp then turned into the major user of current in the device.
 

BytetoEat

Joined Mar 5, 2014
25
Well... the short answers is "it depends." Once while working on an amp requiring "as much gain as possible" (we settled on about 31) at 3MHz I learned all about input pin capacitance.

Seems each pin on a package actually has capacitance! And the input resistors to an op amp form a RC circuit with that cap limiting the bandwidth.

So... long story short, to get sufficient gain at 30MHz I needed to drop the input resistor to 50 ohms. My source had no trouble driving this so it all played together, but the voltage amp then turned into the major user of current in the device.
That is really interesting.. Crazy to think about the rc effects on those pins !
 

WBahn

Joined Mar 31, 2012
30,088
That is really interesting.. Crazy to think about the rc effects on those pins !
It's not just the RC effects. The bond wires within the package also have inductance and those can come into play at some point as well. For some pins we would specify multiple bond wires in order to reduce the inductance (it also reduces the resistance, so which is the driving factor depends on the situation). Since the same die mounted in different packages will result in different bond wire lengths, the choice of package can become very important (and for reasons other than this, of course).
 
Top