[time-nuts] Setting up PRS10 with M12+T GPS

[Magnus Danielson]

Hi Azelio,

On 05/02/2012 10:30 PM, Azelio Boriani wrote:
> Magnus, you are correct but can I avoid to list our beloved
> counters/analyzers? The time-nut main tool is the TIC.

Hehe, yes. But you can do great things even with a spare HP5335A for 
instance. It's quite respectable with about 1 ns resolution in this context.

I have not tried using the SIA-3000 for TIC purposes. Just "getting 
into" the machine GPIB-wise is a bit of a challenge.

> Anyway, to correct the PPS input for the PRS10 the hardware approach is a
> must: so a delay line and a microprocessor are needed. Unless the SRS can
> develop a suitable software for the PRS10 (an inverted TTL serial line is
> already present) to read the M12+.

Well, look at the Spectratime SRO-100, which already has that feature.

http://www.spectratime.com/products/isync/gps-disciplined/SRO-100

Architecture-wise it is very similar to the FEI 5680, a 60 MHz VCXO, DDS 
for FLL and then DDS for output. PPS input and output makes you think 
about PRS-10. The PPS input has about 1 ns resolution.

Now, there is a few things which differs from the PRS-10. It move it's 
internal (and later also its output) PPS to align up to the PPS 
reference. It hence does not need an output alignment. This makes the 
7,5 MHz clock (133 ns cycle) less critical. BTW I do the same in my 
products. The phase of the PPSOUT and the generated 10 MHz will probably 
not align up, something which is critical to some applications.

Also, it can take info from Jupiter-T, OnCore, M12+ and SuperStar II 
receivers for detailed timing info.

Cheers,
Magnus