This page was setup to share information about my work (and the work of others) on the surplus PCOM 9-13 GHz synthesizers from their Point-Point radios (many at 23 GHz). If you have anything to add, feel free and email me at the address at the bottom of the page......
Lance Lascari WS2B, 5/14/2000
Last updated 12/3/2001 by WS2B
-- just made the site available through a link, no action on this stuff in a year, sorry :(
-fixing broken links. I have come into work to play with the units some, my goal is to figure out why the sources "wiggle", I don't know whether this is a loop instability or if there's something else going on. I actually can see many mV of noise on the actual tuning line at the varactor -- very very bad.
Conclusion 1: The PLL is generally just not that stable, and the loop filter has several cascaded R-C poles at about 150 Hz. With that type of configuration, I'm not sure the loop ever was quite stable. Removing some of these poles may make the stability better (it's hard to say), but the reference spurs go way up as the 1502's phase detector really stinks. I think the loop was always very trashy, and these poles were put in to clean up the stuff that was happening in the loop.
Conclusion 2: Something absolutely has to be done about the microprocessor clock. On some units I see about 2.5-5 kHz of wiggle just because of the beat note between the 8 MHz/4 X5 and the 10 MHz TCXO. One solution is to just stick a different crystal on the micro, a colorburst crystal worked fine for me, but the best plan will be to find one that still will allow communication at some reasonable baud rate for setting the frequency. If the frequency is only going to be set once, the crystal could be changed after that's done.
Conclusion 3: The DC-DC converter may not be contributing that much to the noise. The loop is in bad shape and I think that's the biggest offender. Another offender is the RC filtered reference voltage at one of the pins of the op-amp loop filter. Adding a 1uF across the 0.1 or 0.01 uF cap that was there improved the noise quite a bit.
Anyone have any other ideas or findings? Lance 4:25 PM Saturday 10/14/2000
-Added contributions by Luis Cupido CT1DMK (see below)
He's done a great deal of work in figuring out how these synthesizers are programmed without any modifications. See the link below to the information he provided.
Over the past day or so I performed a couple experiments with my "modified & reprogrammed" PLL source.
Stuff to look at
10/9/2000 contributions by Luis Cupido CT1DMK
(Lance note: I suggest using this method of programming for now, it requires no modifications to the units. In the future, when I have some real software written, we may want to reconsider this recommendation... but that might be a while)
Here's a zip file with the following:
Want more info? Email me and we'll talk. I need some inspiration to get working on this again....