[time-nuts] 5>10 doubler

Charles Steinmetz csteinmetz at yandex.com
Wed Feb 4 21:11:27 UTC 2015


Andrea wrote:

>Can you put in another graph the calculated difference to a pure sine wave?

I'm not sure what you mean by "the calculated difference to a pure 
sine wave."  I already reported the amplitudes of all of the visible 
spurs (that is, the ones above the simulation noise floor), which 
define the departure from a pure 10MHz sine wave.  I am attaching 
below the simulated spectrum analysis from which I took those 
reported amplitudes, if that helps.  (There is no new data here, it 
is just graphical rather than tabular.)

Again, this is from a simulation, and I purposely introduced 10mV of 
gate imbalance to model imperfectly-matched FETs.  It is the raw 
output from the doubler, with no traps installed.  The breadboard 
circuit performs similarly, although the 5MHz and 15MHz components 
are about 10dB lower from the breadboard than they are shown in this 
simulation (this depends on how well matched the FETs are -- I was 
able to get a better balance in real life than the imbalance I 
purposely introduced for this simulation).

Best regards,

Charles

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Frequency_doubler_spectrum.gif
Type: image/gif
Size: 45655 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20150204/870dae29/attachment.gif>


More information about the Time-nuts_lists.febo.com mailing list