On Aug 24, 2013, at 10:00 AM, Bill Turner wrote:
> And I disagreed, saying the theoretical bandwidth is theoretically infinite.
> So, do we agree?
No, we still don't agree, Bill. Reread again what I said about taking limits
of a sequence.
Here is another simple illustration (not needing to cite Greek mythology this
time :-):
Do you consider a Butterworth filter to have a finite bandwidth?
(I am using the Butterworth since it is pretty much the simplest filter
possible in the analog world -- an RC filter is a case of a first order
Butterworth, for example.).
The tail of the Butterworth frequency response goes to infinity, does it not?
In fact, since the Butterworth has no ripple, it never, ever falls to zero. A
Butterworth lowpass simply fades away slowly with increasing frequency, but
never actually reaching zero at all.
Yet, a Butterworth filter has a known bandwidth -- not infinity.
The mathematical key, by the way, is that the Butterworth transfer function
falls inversely as the power of 2n, where n is the order of the filter. Thus,
the area under that curve is a finite number -- and that is where you get an
equivalent bandwidth number from.)
The same is true for an FM signal. It has a finite bandwidth.
You might be mixing the two different concepts of (1) bandwidth and (2)
frequency response, Bill. One is finite, while the other can have a response
all the way to infinity.
73
Chen, W7AY
_______________________________________________
RTTY mailing list
RTTY@contesting.com
http://lists.contesting.com/mailman/listinfo/rtty
|