Why does this simple op amp create an AM modulated envelope over its carrier?

The circuit below is an ideal op amp (requires no power, can produce any output voltage) in LTSpice:

LTSpice shows it amplifying the 159.155 kHZ input, as expected. But it also shows a 2500 kHz AM modulation of the envelope! Where does that come from?!

• Stick 1 Ohm as a series parasitic resistance for the inductor. Commented Jun 14 at 23:01
• @periblepsis Could you please explain more? Commented Jun 14 at 23:02
• You've got 10 Ohms to ground and pretty close to no resistance in the inductor. You are driving this thing very close to the high-Q tank frequency. It's likely a beat. You could try using a different frequency with your voltage source (say 110 instead of 160) and if I'm right, you should see a different beat frequency. Also, try increasing the 10 Ohm to 100 Ohm, or to 1 Ohm. See what happens. But also try the 1 Ohm series I mentioned earlier. I don't use the ideal opamps, though. Commented Jun 14 at 23:07
• I've rolled this question back to reverse the correction made by the OP. That correction was initiated following my answer pointing out the error. Of course, if the correction was made with full recognition of my answer pointing it out then, that's fine but, it's not necessary to make any question amendment. In other words don't make already given answers appear wrong by correcting an error in the original post without due care and respect. Commented Jun 16 at 16:44
• @Andyaka Thank you for catching it the typo where I added a "k" which doesn't belong. Fixing it makes the question clearer, but doesn't answer the question, which remains "Why does this simple op amp create an AM modulated envelope over its carrier?". I'd therefore request that you leave the "k" out of the question, as it makes it clearer and more helpful to everyone. "k" or no "k", the carrier needs to be explained, which is what this question asks for, and what you answer below: The carrier comes from being almost, but not exactly, tuned to the resonant frequency. Commented Jun 16 at 17:25

Let me make one thing clear before I move on: -

If you do silly things in simulators you may get silly results


That is what you are doing. Your are expecting an op-amp model to respect real-life but, op-amp models don't expect to be abused. In other words they expect to be respected and not driven in ways that can easily confuse them.

Having said that, what you say is a 2500 kHz modulation is in fact a 2.5 kHz ripple in the high frequency (159 kHz) resonant frequency of the tuned circuit. And, this is expected when trying to stimulate a very resonant tuned circuit with an input that isn't quite tuned to its resonant frequency such as this response when driving a parallel 100 nF and 100 μH inductor via 100 kΩ at around 157 kHz: -

• Andy - You requested elsewhere "If you are still confused about something then leave a comment to request further clarification." I am still confused about: What is "silly" about this circuit? What aspect of it is "abuse"? Given its goal of amplifying a particular frequency, but not other frequencies, what aspect of it is "silly" and "abusive" and how should it be changed? Commented Jun 16 at 2:53
• Spice models are not designed and validated for incorrect use. Leaving the power inputs unconnected is incorrect use. Newcomers tend to treat simulators of any kind as 100% representative of the real world. They are anything but. Commented Jun 16 at 6:14
• @electroGeek are you trying to correct something I have written in my answer? Or maybe you are addressing the other comment? Commented Jun 16 at 10:10
• @Andyaka It looks to me like the asker made a mistake when writing their question originally (2500 kHz instead of 2500 Hz), then you pointed it out, then the asker edited their question to fix the mistake, and then you edited the question to reinstate the mistake. Am I understanding correctly? Can you explain why you did that? That seems like harmful behavior, but I may be misunderstanding something. Commented Jun 16 at 16:32
• @TannerSwett why not report it to moderators if you think I did something inappropriate? Or, you can take my word that correcting a question following an answer that pointed out an error (without proper recognition of the answer that pointed the error out) is regarded as bad form. There are plenty of examples each week where this happens. Think about how this site works. Commented Jun 16 at 16:41