Hi Andrew, for completeness' sake, I'd like to point out that you're correct, it's possible to derive one signal from another. Put simply, you can count 10 Million clock cycles, and you get a PPS from the 10 MHz. But when do you start counting? If you need the PPS to deliver the same time to multiple devices, that won't help. To generate a 10 MHz from a PPS, you can go a similar (inverse) path, but because you only have a pulse per *second*, your loop bandwidth of whatever PLL generates your 10 MHz would be pretty small, and it would take a while to lock. This is all extremely simplified! But I hope it gives you an idea why we have both. Cheers, M On 18.06.2015 18:07, Andrew Clegg wrote: > Thanks Martin and Ian. > > Ian, I understand that 1 pps corresponds to time and 10 MHz corresponds > to frequency, but since the two concepts are directly related, I was > assuming that a 1 pps signal can also be used by a USRP to generate a > stable frequency reference and vice-versa. Perhaps a bad assumption! > > So, I will use the 10 MHz reference for my application, since I'm > interested primarily in frequency stability. > > Thanks again. > > Andy > > >> Date: Thu, 11 Jun 2015 09:45:09 -0700 >> To: usrp-users@lists.ettus.com >> Subject: Re: [USRP-users] Use of an external 10 MHz or 1 PPS reference? >> From: usrp-users@lists.ettus.com >> >> On 10.06.2015 18:20, Andrew Clegg via USRP-users wrote: >> > I ordered an inexpensive BG7TBL GPS disciplined clock >> > (www.ebay.com/itm/121530825744). The unit includes both a 10 MHz and a >> > 1 PPS output. Is there any reason to favor using one or the other >> > reference signal for the B200, or would the results in terms of >> > frequency stability and accuracy be exactly the same with either >> > reference? Is there any configuration that needs to be done for the B200 >> > to use an external clock reference? >> >> The B200 allows you to use both. You will need to set the time and clock >> sources to 'external'. If you had to choose (which it doesn't seem you >> do), the 10 MHz reference would be better suited for frequency >> stability, but doesn't provide any kind of time reference, which you'd >> need, e.g., for synchronizing devices in time. >> >> Cheers, >> Martin >> >> _______________________________________________ >> USRP-users mailing list >> USRP-users@lists.ettus.com >> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com