You should not blast an extremely high modulating signal (4V p-p) into the oscillator. Try 20mV peak (40mV p-p) because the input is from a microphone.

__________________
Uncle $crooge

 

You won't see much modulation at 20mV. My sim uses a 2V signal, but it doesn't show the over modulation. I think you're problem is a simulation setup one, and not a circuit error.

__________________
You don't need a quadraphonic Blaupunkt -- you need a curve
ball.

 

The peak loudest deviation for a 100MHz FM signal is only 75kHz, not 75MHz!
You will not see such a small deviation on a 'scope nor on a sim.

__________________
Uncle $crooge

 

Where do you see 75MHZ deviation? Not in any of my simulations you don't. The O/P asked about a demonstration of FM modulation, he didn't say anything about listening. In fact, that's irrelevant, since you can't hear a 1MHZ signal anyway. The levels and frequencies are chosen to demonstrate the process of frequency modulation, as per the O/P's request. I haven't seen any voice FM transmitter that connects a microphone directly to the modulator's input.

__________________
You don't need a quadraphonic Blaupunkt -- you need a curve
ball.

 

A transistor Colpitts oscillator used as an FM broadcast band transmitter uses a very low modulating signal, not anywhere near 4V p-p.

Most simple FM transmitters use a preamp transistor so that they can be used as a "bug" and detect voices anywhere in a room.

The first link in Google (there are probably many more) shows this FM transmitter that connects the low level output of an electret mic directly to the modulation input of the oscillator transistor:

Attached Thumbnails

 

__________________
Uncle $crooge

 

Those preamp transistor used in most transmitter have voltage gains of between 20 and 100. Mod4 has a dc voltage gain of about 21. Not 4v p-p, but much closer than 20mV. Either way as stated, the levels demonstrate the results of FM techniques. You can't hear FM, so listening won't do any good. Might as well turn up the gain and frequency so you can see what's going on. After all, what's teh worse that can happen? He might *gasp* learn something?

__________________
You don't need a quadraphonic Blaupunkt -- you need a curve
ball.

 

You can find out what kind of modulation you're getting and how much by doing an FFT on that point. LTSpice has many advanced FFT options.

__________________

 

I'm afraid there are way too many artifacts in the signal for the FFT to be effective in determining the deviation. I estimate a few hundred Khz deviation, nowhere near the 75Mhz previously mentioned. So what if that violates the 75Khz spec? The idea was to observe some frequency modulation, not to try to transmit the signal. I think we've accomplised that. The low-level modulation that isn't observable is done the same way.

I'm used to operating circuits outside their normal limits to learn what's going on with them. Indeed, I bring my high voltage projects up from voltages much, much lower than spec so I can keep a watch on the critical values while I crank up the power.

__________________
You don't need a quadraphonic Blaupunkt -- you need a curve
ball.