Glenn, You're right. I would expect quite a bit of jitter. My comment about USB-Serial converters probably should not be on a "timing" news list. However, the 1PPS coming in on the DCD line of a standard serial port jitters too, but is still plenty accurate for "loose" timing systems like NTP servers who are only looking for millisecond type resolution.
I think one of the main reasons we still sell a lot of timing receivers to these customers is really because of the TRAIM function. They really don't need ns type accuracy, but they DO want the receiver to alert them to large anomalies due to sick satellites, and to automatically remove the satellites from the solution. Randy -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Glenn Sent: Friday, August 11, 2006 10:39 AM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] SiLabs Is the USB bus really a good way to get 1 PPS into a machine? I would think it would have quite a bit of jitter in it. -glenn Randy Warner wrote: >Didier, > >I too have looked at the SiLabs parts. I may just have to get one of >their eval boards. > >I use the SiLabs CP2102 USB-to-Serial converter in my new receiver Eval >Boards. Tiny little thing, no external components, and it also has the >ability to handle more than just RxD and TxD traffic. That was always a >problem with the FTDI USB converter in that you could not put the 1PPS >out on the USB bus. A lot of times I ended up using the USB just for >power and used the serial port for comms. Seems like an awful waste of >a USB connection. Anyway, with the 2102 I can put a the 1PPS on a >"virtual" DCD line and keep TAC32 happy. > >Randy > > _______________________________________________ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts _______________________________________________ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts