Big offset voltage in opamp

Thread Starter

fragrance2008

Joined Nov 18, 2009
22
Hi folks,

I have some trouble with my current circuit.

I want to build a photodiode receiver. When i only put the supply voltages for the opamps without input voltage source, i got a huge output ac voltage of about 2V. In addition, the offset voltage of the first opamp is about 8mV. It should not be so big, isn't it? What's wrong with my circuit?

Attacted is my circuit in pspice. In my circuit, the first stage is to convert photo current to voltage. The input ac voltage source and the input resistor are just used to simulate the photo current from photodiode.
 

Attachments

beenthere

Joined Apr 20, 2004
15,819
In general, 741 op amps are not good to use. The design is so old and deficient in many ways compared to newer op amps. Three of them with DC connections multiplies their offsets, so 8 mv is not out of the expected. That huge 1 uF cap in the feedback of the first and last 741 is turning the circuit into an oscillator - think of the phase shift they cause. The very slow slew rate of the 741 output contributes to the oscillations.

A 741 has an input impedance on the order of 200 Kohms. For photodiode work, that figure has to be much higher. Think in terms of at least 1G.
 

Thread Starter

fragrance2008

Joined Nov 18, 2009
22
In general, 741 op amps are not good to use. The design is so old and deficient in many ways compared to newer op amps. Three of them with DC connections multiplies their offsets, so 8 mv is not out of the expected. That huge 1 uF cap in the feedback of the first and last 741 is turning the circuit into an oscillator - think of the phase shift they cause. The very slow slew rate of the 741 output contributes to the oscillations.

A 741 has an input impedance on the order of 200 Kohms. For photodiode work, that figure has to be much higher. Think in terms of at least 1G.
I don't know for sure how to calculate the phase shift. When I simulated the circuit with a pulse signal, it didn't have any oscillation signs. could you give me some guidance? Thank you in advance.

PS:i am sorry that i was not clear. I used the 741 op amps only for simulation and actually i have used op amp AD797 to build the circuit. I did in this way just because i used a student version of Pspice and i didn't have many choices for op amp.
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
That huge 1 uF cap in the feedback of the first and last 741 is turning the circuit into an oscillator.
No.
The first capacitor turns the first opamp into a standard integrator (single-pole lowpass filter) and the other two capacitors form a standard Sallen and Key second-order Bessel lowpass filter.

The first opamp has a loss of at least 10 times instead of some gain. It is inverting with a 10k input resistance which is far too low. The opamp should be non-inverting with a high input resistance and some gain.

The second opamp is also inverting with an input resistance of only 1k which many opamps cannot drive. The resistors could be 10 times higher.

I cropped the schematic and turned it so it is not sideways.
 

Attachments

Thread Starter

fragrance2008

Joined Nov 18, 2009
22

Thread Starter

fragrance2008

Joined Nov 18, 2009
22
I agree with you circuit analysis, but the oscillation has to come from something. I am always suspicious of relatively large value capacitors in circuit with shaky op amps.

To fragrance2008 - you might find this link interesting - http://www.national.com/onlineseminar/2004/photodiode/photodiode.html

There are several good sources that turn up Googling "photodiode amp circuit".
Thank you so much. It is very helpful. I have already replaced all the caps to 22pF, but it seems that the oscillation still exists. it is very confusing that the circuit oscillate at around 2.7 MHz without any input signal.
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
A 741 opamp has such a low slew rate that its output swing above 1MHz is zero and its gain at 1MHz is 1 like a piece of wire and is an attenuator above 1MHz. It can't oscillate above 300kHz. It is probably rectifying a nearby 2.7MHz radio transmitter.
 
Top