I am baffled. Built inverting amp with 1kΩ Rout and 1Ω Rin. Input from sig gen is 0.01V and increasing freq from 1 to 1M in decades. But I'm only getting ~20 amplification up until ~40kHz before it rolls off. I've checked the resistor values multiple times, directly measured the sig gen out with the scope, and changed the TL071CP several times also. Same thing...

 

Your signal generator will have output resistance equal to around 50Ω thus, only 20V/V.

 

hi Sam,
How is pin 3 connected to 0v, is it via that cap.? if yes the OPA has no DC path to 0V from the NonInv input.
E

Update: connect pin #3 directly to 0V

 

How is pin 3 connected to 0v, is it via that cap.? if yes the OPA has no DC path to 0V from the NonInv input.

The cap is across the - and gnd rails and pin 3 is connected to the gnd rail. Same on "upper" side has cap between + and gnd at the supply point to the chip.

Your signal generator will have output resistance equal to around 50Ω thus, only 20V/V.

Not sure I follow you. Used the same components to build 1000X non-inverting and it worked fine.

 

The cap is across the - and gnd rails and pin 3 is connected to the gnd rail. Same on "upper" side has cap between + and gnd at the supply point to the chip.

Hi Sam
You didn't show the full circuit, so I assumed a single supply.
E

 

The 50Ω impedance of the sig gen is added to the 1Ω of Rin?

 

The 50Ω impedance of the sig gen is added to the 1Ω of Rin?

The input impedance of the amp circuit is only 1 ohm. You are loading down the output of the generator.

 

Not sure I follow you. Used the same components to build 1000X non-inverting and it worked fine.

Sure, but the input resistance of an inverting amp in your case is equal to 1Ω.
So, the situation will look like this:




AV = Vout/V_sig = - R2/(R1 + Rsig) = 1kΩ/51Ω ≈ -20V/V

 

K got it, thx! Been beating my head over this for a couple of days. Didn't even consider the sig gen impedance and this is the first time I've encountered it. I'll certainly remember it in the future!

 

Non-inverting amp, by contrast, has very high input impedance regardless of the gain setting resistors.

Bob

 

So I need to change Rout to 50kΩ...

Edit: Yup exactly 10.0V out with 0.01V in @ 1Hz!

 

Your biggest mistake was to used such a low resistance in the amplifier circuit in the first place. You should never use resistance lower than 500Ω. If you do not fully understand what you are doing. Or you are unable to predict the consequences of such a decision.

 

So I need to change Rout to 50kΩ...

If the output of the signal generator is calibrated for a 50 ohm load then Rin =50 ohms and Rout = 50K.

 

Or you are unable to predict the consequences of such a decision.

Kinda surprised but then I guess Burr-Brown and then the TI rewrite did not intend for people like me to use their example diagrams to breadboard with to confirm the results.

 

You cannot achieve a gain of 1000 at 40kHz.

Here is the open-loop gain vs frequency.
Gain starts at 100dB (x100000) at DC and goes to 0dB (x1) at 5MHz.
Gain is 40dB (x100) at 40kHz.

 

My test circuit using the 50Ω impedance of the sig gen for the Rin and 50kΩ for Rout started rolling off ~900Ω and @10kΩ was 1.44V. I'll try and post the plot tomorrow.

 

started rolling off ~900Ω and @10kΩ was 1.44V.

What does that mean?