1MHz Oscillator using op amp

[mathk]

Is it possible to design a 1Mhz sine wave oscillator using op amp?

According to some App Note from TI it is only possible design oscillator up to a couple of Khz with op amp. But if you take high speed op amp like some Analog Device "AD8055" which have a slew rate of 1400V/us and a 300Mhz -3dB bandwidth I am wondering if we can design oscillator that goes up to a couple of Mhz with it?

Also advice on designing such oscillator are more than welcome.

Thanks.

[c4757p]

Try it!

[fcb]

Of course you can design a 1MHz oscillator with an opamp.

One method might be a "phase-shift oscillator" (google it), the values are quite reasonable (220pF/300R) - i'm assuming that you want a simple oscillator with a sine out?

[JVR]

Just use a Xtal as a reference.

[mathk]

Yes @fcb simple sinusoid is what I need. I have try the "Wien bridge oscillator. I set the frequency at about 900KHz but the problem is that I have more a square wave than a sine wave and the frequency is jumping from 950 to 960 KHz (measured with a rigor 100Mhz bandwidth scope).

Then I tried a simple LC oscillator and the output is much cleaner, I still have a square wave but filtering the output give me what I need.

Any suggestion to increase stability?

Thanks

[fcb]

It's in the nature of the Wien for it's amplitude to grow until clipping.  You need some form of amplitude stabilisation - bulbs used to be used.

Once you've sorted the amplitude out, then tackle the f-instability.  Just be careful that it actually exists, and not some artifact of the Rigol.