what's wrong with this non inverting opamp configuration?

[xchip]

Its supposed to be a non-inverting op amp with a gain of 100 running on a single power supply.

The virtual ground is set at 2.5, and the input I want to amplify is an VAC of 0.1 volts, however i see a gain of 1 :-(

I'm sure I am doing something reeeally stupid :-)

Thanks!

[jimmc]

Add a resistor between output of U1A and C4/Pin 3 U1B. At present voltage source (Rout = 0 ohm) is driving the output of U1A... try monitoring it.

Jim

[xchip]

Same thing, I tried with a 1k and 100K resistor. Thanks thoguh!

[Andy Watson]

Did you put it here?

[skipjackrc4]

I agree with the other two posts.  You should have a resistor there, probably on the order of 100k.  I use a very similar design as an input buffer to effect pedals. 

Also, is your AC coupling capacitor really 10 millifarads?  That is...high.

[xchip]

Yes it worked!! The pic helped

So it was not such a dumb question! I dont really understand why a resistor there helps get the op-amp working, any explanations or links are welcome! I am so surprised i never came across any document explaining the need of such a resistor!

As for the cap if I was making it 10uf I was getting something that looked like a damped oscillator. Now with the resistor in the place you guys suggested a 10uf is working nicely.

Thanks!

[IanB]

Seems like you got there, but see below also.

An op amp amplifies the difference between its two inputs. Do you not see that you had made a short circuit between the inputs before? If you make a short circuit there can be no voltage difference to amplify.

Also a gain of 100 is not going to work if your input signal has an amplitude of 0.1 V. That means the output would need an amplitude of 10 V, which is not going to happen on a +/- 2.5 V supply...

Another thing to note is that the LM324 output may go down close to 0 V, but it will not go all the way up to the supply voltage. So you might be better splitting the supply at 2.0 V rather than 2.5 V.

[xchip]

Hi IanB, firs toff thanks for taking the time to draw the schematic!

About the short circuit, I don't really see it I think I need to look deeper at that, I'm trying to find any pages or videos that explain that but i cannot find any!

As for the gain causing clipping and having to split the supply at 2.0, yes I saw that as soon as it started working but thanks for pointing it out!

I want to use the op amp to amplify a tuned LC tank, I didn't want to have any resistors that could damp the oscillations.. what could i do now? Do I really need to use a mosfet? Can it be done with a opamp then?

Thanks!

[IanB]

You need R6 and C1 because you are referencing your input signal to the negative supply rail instead of your virtual ground. You need R6 to bias the amplifier input to the virtual ground voltage and you need C1 to block the DC offset present in the input signal.

If you base your input signal directly on the virtual ground you can do away with C1 and R6, as you can see below.

[xchip]

Ok now I got it working in the simulator as you say and I understand it

I am now using a inverting amp config and I am using a buffer so the LC tank sees a high impedance and it can resonate freely.
Unfortunately I am getting the results of the picture (has the circuit on the bread board) regardless of how do I set the 100k pot (that sets the gain) I am getting a clipped sinewave, and changing the vgnd pos wont fix it

If I add a cap between the buffer and the inverting amp all works fine though. which is quite puzzling!

What on earth could be going on? Thanks!

[IanB]

Well, look at this! The simulator shows the same results!   

Maybe the output of the op amp is not able to go above about 3.8 V?

[xchip]

Eheh I see, I should spend 90% of the time with the simulator and 10% on the breadboard, thanks for taking the time to put my circuit in the simulator!

So why on the first circuit do I get such a dc offset? After the buffer I should be getting 2.5v, but it could well be that the resistor divider is giving me 2.501 and then that .001 gets amplified and ends up showing up as a huge dc offset... I tried lowering the gain of the amplifier with the pot and no matter what I think I was still getting that offset so it has to be something else.

As for the 2nd circuit, the one with the cap, I am not using R6(nor the vgnd), and it turns out I get a nice oscillation as in your sim... that is pretty scary because the circuit shouldn't work at all

Thanks IanB for keeping me curious

BTW I happen to have a coil and a varcap that I can tune to 77.5Khz and I am hoping to receive the signal of an atomic clock that transmits in that frequency, this is just an excuse to learn about LC and hopefully some radio stuff Now I'm still at the LC part

edit1: I tested the circuit with the clipped sinewave in my ltspice and I get a proper sinewave, how comes your simulator and mine dont agree?

edit2: I noticed your AC source has 0.1 volts of DC offset, and that is what is causing the 1volt dc offset at the output, did you put it on purpose?

I'm feeling close to understand something great!

[IanB]

edit1: I tested the circuit with the clipped sinewave in my ltspice and I get a proper sinewave, how comes your simulator and mine dont agree?

I don't know. The output clips because the maximum output voltage is Vcc - 1.5 (see datasheet). So if your supply voltage is 5 V, the maximum output voltage is 3.5 V. Any attempt to make the output go higher than this will cause clipping. (Edit: it looks like your simulated output peaks at 2.63 V, so it won't clip.)

edit2: I noticed your AC source has 0.1 volts of DC offset, and that is what is causing the 1volt dc offset at the output, did you put it on purpose?

In your earlier picture you showed the input signal as if it were biased above the mid-point voltage:



The picture of the output signal seems to show it sitting above the 2.5 V line, which is what would happen if the input signal had an offset:



I therefore added the 0.1 volts of DC offset to the simulation in order to replicate what I was seeing on your picture.

Notice above your output signal is clipping somewhere above 3.5 V, which is what the datasheet suggests will happen.

[xchip]

Thanks!

Now it's working, I have a buffer(x1) + invopamp(x100) + inovamp(x100) and I am just using an cap after the first invopamp in order to get rid of some offset voltage.

BTW I was expecting the lc tank to pick whatever is being transmitted on the 77.5Khz but all I get is a constant oscillation :-/ Why would that be?

[Zero999]

Congratulations, you've made an oscillator!

I don't think you'll have much success with the LM324 which is only suitable for DC and extremely low frequencies. Look at the graphs for the large signal bandwidth and gain bandwidth product.
http://www.ti.com/lit/ds/symlink/lm124-n.pdf

[xchip]

LOL thanks!!

I am now using more stages with a reduced gain to increase bandwidth and it stopped oscillating.

I used a AD9850 to performa a frequency sweep and I got a nice spike at 77.5Khz. (The AD9850 has just a wire connected to its output that works as an antenna, it is not connected to my LC tank) So up to a certain point we can say mission accomplished :-/

Some people told me 'a radio doesnt look like that! It needs FETs!' But nobody was able to tell me why my design is bad, anyone?