Quick Question on Frequency detector!

[Megamox]

First of all excuse the drawings, MS PAINT seemed the handiest thing to use.

I just had a question on whether there are frequencies present in a signal which appears to be constant. The best way to describe the question is from the diagram. If I sent the waveform in red down the transmission line, I'd assume a frequency detector tuned to that frequency would 'BLIP' every time it detected it - Output (A). But I'm sure, if my memory's not playing tricks on me, such a carrier being keyed on and off produces the spectrum shown. This looks like the Sinewave is constantly present in the output signal and so the detector should look like Output (B) instead of Output(A). Which output would occur? Is there really a constant frequency present in the "blank" parts of the signal tool? What if the bursts were separated by half an hour! Would the detector still be reading constant detection?

Thanks!

Megamox

 

[JimB]

Yes it does do funny things in your head doesn't it!

The quick answer is that YES the frequency detector will blip every time the carrier is present.
But sitting here at the keyboard trying to find the words to explain it is hurting my head!


When Fm is small compared with Ft, the small pulse of signal will have a spectrum something like you have shown and the attachment.
Note that the sidebands will be a lot wider than the 3Fm component.

As the frequency Ft gets very low, the spectrum will only exist time when the carrier is turned on and the spread of the sidebands will be mostly determined by the risetime of the modulating waveform.

Does this description help?

JimB

 

[Megamox]

It does indeed!

I've been thinking about it a little more and I think I've gone off on a bit of a tangent, but it seems to make sense... so here goes.

Imagine watching a TV. The television flashes up pictures at around 25 frames per second. This is fast enough for your eye to see a continuous image. However the tv signal is not a continuously smooth image, it is a series of flashes or bursts of signal. So perhaps if the resolution of the eye were faster we could see the bursts.

Perhaps it's the same thing with the receiver. If the signal is bursting on and off so quickly, the receiver's "Persistence of Vision" cant tell the difference between this and a continuous signal. So then, there are no frequency components really in the gaps between the signals, but the detector will tell you that if its resolution isnt high enough to see them?

So for a signal that's keyed on and off say every half an hour, the modulating frequency is so small in comparison to the carrier signal that the sidebands shrink to being practically unoticeable. This means the detectors bandwidth "resolution" is now adequate to make out the bursts intermittently. However if we up the modulation frequency, the bursts come more often, the receiver gets to a point where it cannot distinguish one from another and believes everything is continuous and outputs a constant detection. However there are no frequency components in the blank parts of the signal, the problem lies with the resolution of the detector and if we broaden the resolution of the detector at this point to again detect the sidebands, the same signal now looks 'Blippy'.

Does this make sense, or am I going crazy

PS. Then interestingly, in answer to the original question, which output would be correct... they both are, depending on the bandwidth of the receiver at the detector.

Megamox

 

[JimB]

Does this make sense, or am I going crazy

More or less.

If the bandwidth of the detector was low compared with the modulation frequency, the detector output would not be a pulse but a spread-out hump which does not rise to the full amplitude of the incoming pulse.

If the incoming signal was present for a long time compared with the bandwidth of the detector, the output would rise to the full amplitude and then fall off slowly when the signal drops to zero.

JimB

 

[ronv]

Depending on what you are trying to do it may be easier to detect the time difference between bursts. You could for example charge a capacitor with a few cycles of your signal and discharge it at a known rate.

 

[circuitspecialists]

What are you trying to accomplish? Your ouput B will not occur with the waveform shown. You will never get a constant DC output from a receiver regardless of how often the Carrier is pulsed. This is called ASK or OOK (ON/OFF keying). You will however pickup noice (static) in the receiver during the portion of the signal between the bursts.