Hi

A real simple question that has puzzled me too long.

In terms of a phase accumulator.

a constant frequency out Fout , is a constant phase increment,

New phase = old phase + phase constant


For a linear up sweep of Fout

new Phase = old phase + ( phase constant * N)

where N is an integer, increases by 1 on each iteration


So for a log sweep ?

new phase = old phase + ( ??? )



so what do I put in for ???

 

I can't speak authoritatively within the context you've given, but "log sweep" generally means that the logarithm of the parameter changes linearly.

 

So instead of increasing by 1 in every iteration, N' will be
N'=e*N.
Sounds reasonable.

 

Thanks

can i turn the question around ?

If for instance I have a plot of a 'log sweep',
how could I 'prove' that it is a log sweep ?
how could I prove the parameters of the 'plot' ?
like ( start freq , stop freq, sweep 'speed' )

for a sweep down, would it be like constant number of cycles per octive ? or decade or something ?

My current best bet is to generate a plot algorithmically, then compare the points !

 

In the case of plotting the function, you prove it is 'logarithmic', if the plot is a straight line when plotted on a log graph.

http://www.printablepaper.net/category/log

 

would you consider a plot as sufficient,