CD74HC4046A: FM Modulation

Hi Patrick, 

Posting a link to our previous discussion for reference to anyone else reading this in the future: https://e2e.ti.com/support/logic-group/logic/f/logic-forum/1417401/cd4046b-pll-fm-modulation-and-demodulation-help/5466191#5466191 

It seems like your modulator/demodulator setup is functional, although maybe the demodulated output can be cleaned up with some additional filtering or a different probe setup. Can you clarify what you're looking for? 

Regards, 

Connor 

Hi again Connor, Thanks for monitoring and helping people in the forum. I really appreciate it. 

I would try to take look on the additional filter as the scope that is available on my disposal is for school equipment only. 

I have run some further testing that is needed for the application of modulation and demodulation. 

it does not show on the spectrum analyzer. I think the reason why cuz the center frequency is too high around 10^7 Hz which is why I need it to be around 100kHz ~ 200kHz range to the generate an analog fm signal .

The thing that I need as well is to demodulate a sine wave, and not square wave. When I tried using sine wave as an input signal, it demodulates that resembles like a positive half-wave rectifier. I am expecting it to be a full sine wave. The reason might be because the document is an FSK which makes this a digital modulation, and I need an analog FM modulation which is why the previous testing made was for square wave. The SLAA618 "Implementation of FSK for FM Mod and Demod". I didn't know that there is analog and digital fm modulation and demodulation which is my fault for not knowing them. 

I think you should be able to re-use the schematic from the FSK implementation application note even for FM modulation. You could try changing the passive component values to tweak the VCO frequency into the range you're looking for. For example, to get a center VCO frequency around the 100kHz range you might try something like R1 = 500kOhm and C1 = 0.1uF based on Figures 17 and 18 in the datasheet. 

Depending on what you want the modulation index to be (i.e. how much the VCO frequency can change relative to its average value), you should be able to adjust the amplitude of the input signal to the modulator. 

I would expect that inputting a sine wave should work correctly, can you send a picture of the scope measurement with that setup? Also, are you setting the DC offset of the input to VCC/2? 

Regards, 

Connor 

That's probably what I am missing on my application the VCC/2. This fix the fluctuation happening in steady state fluctuations before. 

Though the sinewave seems to clipped, and it is probably clamped as well?

Based on the your reply "FSK implementation application note even for FM modulation.", you were saying that FSK diagram can be used for analog fm signal generator?

Based on the figure 17 and 18, when I look at it, it only goes up by 800 Hz at R150K which is kind of far from 100kHz. am I missing something?

I am using this figure to grab my values. is it different from your figure?

That's good to see that changing the offset gives better results. I'm not entirely sure why the sine wave is clipping since it looks like the peak to peak is only around 600 or 700 mV, which should be within the normal VCOin operating range. How much do you have to reduce the input amplitude until the clipping goes away?

Yes you're right, I accidentally misread the scale of Figure 17 and 18 as kHz when actually it's in Hz. 

In my previous comment where I mentioned that you can re-use the same basic schematic, I just meant that FSK is just a special use case of FM. FSK usually implies that the input signal you modulate is digital, and the VCO will have discrete frequencies depending on if the input is a '0' or '1'. FM uses the same principal except the input signal is analog, so the VCO frequency can be anything within a continuous range. 

Regards, 

Connor 

I think the sensitivity is too high when adjust the Vpp and offset, it clips right away with just 100mV difference. Input signal: 2Vpp + 1.4Vdc

Based on what you said, i know digital is 0 and 1, and if I feed this in the input signal on the fm mod chip. this will become digital. 

If it is a sinewave, and it is continuous, the output will become analog right away without other modification? Sorry if this question should have been intuitive. 

Sorry, I should have clarified the offset of the signal should be half of VCC of the CD74HC4046A devices, not half of the signal's peak to peak voltage. I think I saw that you're using a VCC of 5V for the CD74HC4046A devices so I would try setting the offset of the input signal to 2.5Vdc. The reason for this is that the VCO frequency will start to clip if VCOin is less than around 1V or so. With your input of 2Vpp + 1.4Vdc, the low voltage of the signal is Vdc - Vpp/2 = 1.4V - 2V/2 = 0.4V. From the curves showing VCO frequency vs. VCOin voltage, it looks like the VCO frequency is not well defined when VCOin is less than 1V or greater than VCC. 

That's correct, essentially the circuitry for FSK vs. FM modulation/demodulation is the same, but the way you interpret the input/output signals from the mod/demod pair is different (analog vs. digital). 

Regards, 

Connor