-1
\$\begingroup\$

I have put up a circuit for Wien bridge oscillator using 741 op amp, with R= 390 ohm and C = 2.2nF, for operation around 200 khz ( gain is 4). However, it did not work so I switched the resistor in parallel branch with potentiometer. This is the closest I could get.

enter image description here

Here is the circuit, with Rf = 8.2K and Rs = 2.2K

enter image description here

Any change in potentiometer does not yield any output.

Can someone please tell me some way to improve this to get good sine wave ?

\$\endgroup\$
12
  • 1
    \$\begingroup\$ Schematic is needed. \$\endgroup\$ Commented Sep 20, 2014 at 16:53
  • \$\begingroup\$ Its a normal wien bridge oscillator; i thought people might be knowing the circuit of it. \$\endgroup\$ Commented Sep 20, 2014 at 17:05
  • 1
    \$\begingroup\$ What does your power supply look like? \$\endgroup\$
    – EM Fields
    Commented Sep 20, 2014 at 17:47
  • 2
    \$\begingroup\$ Your power supply sounds fishy. Please show your power supply schematic. \$\endgroup\$
    – markrages
    Commented Sep 20, 2014 at 18:26
  • 1
    \$\begingroup\$ It should work. You should measure 24 V from negative to positive supply terminals of the op-amp. \$\endgroup\$
    – markrages
    Commented Sep 21, 2014 at 2:14

2 Answers 2

1
\$\begingroup\$

The problem is the 741, which doesn't have enough gain at 200kHz to overcome the loss of the bandpass filter.

Here's a schematic of the identical circuit with a decent opamp, which works, and the LTspice circuit list is here if you want to play with the circuit.

enter image description here

\$\endgroup\$
9
  • \$\begingroup\$ The gain of the wien bridge oscillator should be greater than 3. That's what I have provided. So how does gain come into the picture here ? \$\endgroup\$ Commented Sep 21, 2014 at 1:37
  • \$\begingroup\$ It appears that your 741 has a gain-bandwidth product of less than 600kHz (a gain of 3 times a frequency of 200kHz) so no matter what resistance ratio you provide with Rf/Rs, the opamp itself is inherently incapable of providing that gain. \$\endgroup\$
    – EM Fields
    Commented Sep 21, 2014 at 3:10
  • \$\begingroup\$ Infact, following your advice, I decided to vary the gain as mentioned in comments above, giving me a good sinewave. Thanks \$\endgroup\$ Commented Sep 21, 2014 at 3:21
  • \$\begingroup\$ @plutoniumsmuggler: I doubt if you will succeed. A 741 type opamp never can be used for a 200kHz oscillator (due to its small-signal as well as large-signal limitations). If you select another amplifier watch the slew rate!. \$\endgroup\$
    – LvW
    Commented Sep 21, 2014 at 7:42
  • \$\begingroup\$ @LvW . I am getting a 150 khz now using 220 ohms and 2.2 nF. So around 200 khz should be easy by using lesser value resistor. \$\endgroup\$ Commented Sep 21, 2014 at 8:15
1
\$\begingroup\$

To Plutonium Smuggler: I think, it is more appropriate to respond to your last comment in this way.

1.) Undoubtly, the nominal frequency for the Wien oscillator is fo=1/(2*Pi*RC). The other "resources" resp. formula you have found will, certainly, belong to another type of oscillator circuit.

2.) The above formula for the Wien type assumes an amplifier having a gain of "3" (without any parasitic phase shift). However, this cannot be provided by a 741 type opamp at a frequency of 200kHz.

3.) Does this mean, that this opamp cannot be used for oscillating circuits around 200kHz? No - it can be used. However, one must be aware of the consequences: It is not possible to DESIGN such an oscillator; it will be - more or less - a trial-and-error process because the opamp introduces additional phase shifts (which are not known with sufficient accuracy). More than that, these non-ideal opamp properties are different for the various opam types.

4.) Example: Using your values (220ohms and 2.2 nF) I have tried to produce oscillations (Spice simulation) for two different 741 types:

a.) µA741: 400mV at 135kHz for a gain of 1+3=4.

b.) AD741: 180mV at 142 kHz for a gain of 4.

5.) Evaluation: Of course, it was possible to create oscillations - however, at much lower frequencies as desired (335kHz) because the additional opamp phase shift must be compensated by a corresponding inverse phase shift of the Wien network. The required gain was found by trial and error.

6.) Question: Is it possible to derive the necessary gain values from the BODE diagram (loop gain ac simulation)? The answer is NO!. The reason is as follows: The oscillating circuit provides - as can be observed (400mV, 180mV) - automatic amplitude limitation caused by slew rate limitation. This is a non-linear effect (additional phase shift in the time domain) and cannot be seen in the ac simulation.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks for this explanation. With reference to your point 4 and 5, it is not clear whether the simulations gave 135 Khz or 335 Khz. And you mean that I CAN increase the frequency, but only at the cost of amplitude ? That is in fact very good, because I can always feed it to a CE amplifier to get required amplitude. So now I am confused; what limits us or prevents us from using it for high frequency ? ( I thought it was Gain BW prod which limited this ) \$\endgroup\$ Commented Sep 22, 2014 at 12:31
  • 1
    \$\begingroup\$ Actual value: 135 kHz (335 kHz was the target value using 220 ohms and 2.2nF); Yes - by trial and error method you can produce higher frequencies than expected for "classical" designs - but the result cannot be predicted and varies from opamp to opamp, and hardware will give other results than simulation. It`s no DESIGN - it is just "try-and-error". \$\endgroup\$
    – LvW
    Commented Sep 22, 2014 at 13:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.