Why there is a distortion in FM

Tom Albrecht wrote:

The zero point should be independent of the signal level, so something's not right. It's possible there is a problem with one of the tuned circuits, and that can be hard to see without a sweep generator. I usually find that if there are no problems, a fixed frequency alignment works fine, but if there are problems, it's awfully hard to see which tuned circuit is acting up.



Good info. I will go back and test again. It is possible I may have had the AM modulation turned on on the sig gen. Not sure if this would make a difference.

Tom Albrecht wrote:

Did you get a clear peak within the adjustment range for all the adjustments other than the secondary of the last stage?



Yes. I found the coils very easy to adjust with clear peaks in each one. The only exception were the two weird ones connected to the variable capacitor where turning the capacitor also pulls the cores out of the coils. I found it hard to adjust these without also moving the capacitor rotor. The one closest to the front (L4) that aligns dial to RF was pretty predictable, but the other one (L6) seemed to have little effect. Not sure what it was supposed to do.

Tom Albrecht wrote:

When you're adjusting the secondary for zero volts, it should be strongly positive at one end of the adjustment range, and strongly negative at the other end. In between, it should move in a fairly smooth linear fashion through zero. It saturates near the positive and negative extremes, but has a nice slope going through zero. Is that what you're seeing?



Yes. seemed nicely linear transition from pos to neg. Pretty cool actually how that works. I am still trying to understand how this circuit manages to perform FM demodulation though. How does it map frequency to voltage as necessary for FM? If the input IF signal goes slightly up in frequency how does this combination of rectifiers turn this into a voltage swing?

EDIT: How strongly positive and negative should the signal be? Is it +-1V? or more like 2? I think it is the limiter's job to keep this at full scale if I understand correctly, so how big is full scale?