Hi Henk, yes I have play a little too.......presumably the phase error which would be got by inaccurate cutting of the cable length "could" be compensated by appropriate facilities in the software?? I have certainly seen the small shifts that are available to comensate for sample differences in some sound cards. I wonder what is possible/practicable??
An alternative to that would be a small hairpin section (just sufficient to allow the adjustment) of 50 ohm strip-line (forgive me if that is not exactly the right term for a pcb trace above a ground plane) could be etched in a way that would allow it to be trimmed (shortened) by a series of optional straps using 0R0 "resistors" Alan G3NYK > Posted by: "henkkeppel" [EMAIL PROTECTED] henkkeppel > Date: Tue May 13, 2008 3:59 pm ((PDT)) > > Well this is a good idea, use two simple quad diode mixers. > I wrote before in thse columns that I tried this in the sixties of > the former century with Germanium diode quads but these diodes were > not fast enough then, nor were the Si ones. I selected them on equal > voltage on conducting because that determined the carrier suppression. > The 90 degrees phase shift is on VHF much easier obtainable then on > HF. Just take a quarter wave coaxial cable. Take into account the > velocity factor, so about 1ft+. And with the now available thin > coaxial cable it will be much smaller then with RG58 or59 of those > days. We used a griddipper to determine the quarter wave. > For transmission we also needed a 90 degrees shift between the 2 > audio signals. That was very difficult and was realised with a > complex RC filter with many poles and zeroes of non standard values. > If I had enough time I would give it a try. > > --- In [email protected], "Alan Melia" <[EMAIL PROTECTED]> > wrote: > > > > I have watched the comments about 2m operation and wonder if there > are some > > other possible solutions? The multiplexer/ switching mixer chip is > not fast > > enough for 144Mhz mixing but diode ring mixers are readily > available at that > > frequency. So then to generating the quadrature LO at 144Mhz....its > a bit > > fast for easy logic....but if you used the standard quad circuit > > used at hf > > and mixed I and Q up in phase you would generate a 144MHz > quadrature LO > > signal. The problem is obviously the extra mixing stage and birdies > but > > these should be easy to avail and the second mixer could be used to > sweep > > the signal over a wider portion of the band (DDS??) the > alternativeis to mix > > down and then do the quadrature bit, which might be a lower parts > count. I > > have no idea which would work out best though.....any ideas if > a "direct > > conversion" 2m RX would be worth doing like that?? I suspect > the "IF" vesion > > with some IF gain would be perhaps better. > > > > Alan G3NYK
