Oscillator cap difference

[upand_at_them]

I have this circuit on a breadboard:


But it will only oscillate when I use 500V caps (ceramic). When I use the little 50V caps (ceramic, same uF value) it just won't oscillate. All the caps measure in spec on my multimeter.

The 9V battery isn't exceeding the capacity, so...What gives?

 

[alec_t]

Is the oscillation frequency what it should be (given the R and C values), or is it some spurious value ? (I'm wondering if parasitic inductance is playing a part).

 

[upand_at_them]

Yes, with the high voltage caps, it oscillates very close to the expected frequency. And the scope trace looks very clean.

 

[simonbramble]

Are you building the exact circuit shown (with the same component values)? The LM358 is a pretty crappy op amp. If you try to exceed the frequency of the circuit above, the chances are it is the characteristics of the op amp (and not the caps) that are coming into play. Do you have a better op amp to hand? I am thinking the input bias current might be upsetting the biassing of the Wien Bridge network and causing the thing to sulk...

 

[audioguru]

The 9V battery needs a 220uF to 1000uF capacitor across it because the battery might be old and have a high resistance.
I agree that the LM358 is a very crappy low power (severe crossover distortion) and low bandwidth opamp.

 

[upand_at_them]

Yes, exact component values as the schematic and a new battery. I dropped in a 100K pot and got the "bad" caps to work when Rx was below 44K. I guess it's just a temperamental circuit? It's odd, though, because the 500V caps operated over a wide range of Rx and all of the caps measured to within 10% of .01uF.

The output is pretty clean:


You're saying there is an even better op-amp I could stock up on? This was just a simple sine wave generator that I'm using to compare to my PIC-generated sine waves. But if there's a better device for the same price, I'll get some.

 

[audioguru]

An LM358 and LM324 are low power so their output transistors do not have enough bias current which causes crossover distortion.
The low current causes the high frequency output to be poor above only 2kHz.
A TL071 single, TL072 dual and TL074 quad audio opamp are much better and are inexpensive because they are used a lot. They are low noise and have very low distortion.

 

[upand_at_them]

One more thing...In the above circuit I'm going to replace the Rx resistors with pots for some variability. Would I be correct in assuming that the current consumed is the peak-to-peak output (2V) divided by the resistor value? I picked up some thumbwheel pots from RadioShack and I want to make sure I'm below the 0.05W rating.