At 02:19 PM 4/19/2007, Hulen Smith wrote:
>Hello Flexers,
>
>Some time back I saw a post from someone that explained they were getting rf
>output (very low) when they were keyed up in SSB but with no mic gain or
>  It's almost
>like a balanced modulator in an unbalanced condition. Yea yea yea, I
>know.... old school. I'd appreciate any ideas. True that while in the
>carrier modes this will have little consiquence, however in SSB, this won't
>fly. Oh BTW it isn't oscillation. It's an on frequency signal.

It is precisely the same as the unbalanced modulator problem.

The SDR1K, in transmit, is basically a programmable oscillator (the 
DDS), fed through a vector modulator, and the vector modulator inputs 
are AC coupled (C20,C5, C27, C28), with no provision for adjusting 
the DC offset (which is what you'd need to suppress the carrier 
fully).  The signal is, as you've noted, quite low level, but it will 
be fixed.  If you put some modulation on the signal (i.e. feed 
something into I/Q) you'll have a much larger signal to compare 
against, so it's easy to get the 40dB or so carrier rejection 
(compared to the desired signal).

Most sound cards are also AC coupled (and, of course, if you're using 
transformers to eliminate ground loops, this is especially so).

The FET mux used as the QSE is pretty good in terms of balance 
(because it's mostly determined by the on resistance of the FETs, and 
they're all on the same die, so they're pretty well matched).  There 
might be some small imbalance in the clock signals (the DDS output 
filters are composed of discrete components, and they match well, but 
not perfectly), and there's small leakage currents in various places 
that all add up.

This is a fundamental problem with using a SDR1K as a signal 
generator, by the way.  It's got great dynamic range on the desired 
output (80-90 dB, easy), but it has this "leakage spur" that's always 
there, and impossible to cancel without a hardware change.  If you 
really need low power AND low DDS leakage, you could run it through 
some pads.  The challenge is if you need both low and high power, 
since there's no "adjustable attenuator" in the RF path.

Jim, W6RMK 



_______________________________________________
FlexRadio mailing list
FlexRadio@flex-radio.biz
http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz
Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/
FlexRadio Knowledge Base: http://kb.flex-radio.com/
FlexRadio Homepage: http://www.flex-radio.com/

Reply via email to