[ntp:questions] Very large jitter and offsets on GPS ref clock after upgrade to "p5"

David Lord snews at lordynet.org
Wed Feb 8 17:45:05 UTC 2012


David J Taylor wrote:
>> Nearly that, but I'm splitting the power and planning on
>> having +9V and -Ve on two of the lines and a 5V regulator
>> since some of the GPS modules with interfaces take  over
>> 200mA so common OV with the -Ve at the far end should
>> remove any effect from voltage drop over that length of
>> cable.
> 
> OK - usually for low-frequency stuff the more ground lines the better!  
> If you're having +9V then I would be inclined to common-up the four 
> ground lines, and use a single line as the +9V.

Thats what I'm intending. The 9V +Ve and -Ve are essentially
floating at the supply with -Ve linked to 0V at the GPS.

Nothing bad has happened so far, rootdisp is 1.397 and offset
0.002ms.

> 
>> I might try rs422 for the PPS (and TxD lines) but they
>> are 16-pin chips and have unused sections. The 8-pin
>> versions with single Tx + Rx are much too expensive.
>>
>> David
> 
> It would be an interesting experiment in view of the recent discussions, 
> and allow you to terminate the lines correctly at both sending and 
> receiving ends.  Having said that, 18m will be about 54 ns divided by 
> velocity factor, say 70 ns, so hardly critical to a receiver chip that 
> likely has a couple of microseconds rise time input filter in any case. 
> Please measure waveforms if possible, both on the line and at the chip 
> input pin.  See what filtering may be included internal to the chip.

There is no filtering indicated from datasheets but + and -
inputs will be either biased by added 390R to +5V and 180R
to 0V or by 100pF in series with 100R to 0V. For differential
mode, the edges are supposed to be ramps caused by the
reflections with the outputs being triggered at the crossover
of the rising and falling edges.

I'm intending to use 6 x twisted pair so that there will be
spare 0V lines, and D15 connectors to prevent mix ups.


David

> 
> BTW: my own extensions over a ~4 m length have been with unscreened, 
> three-way or four-way cable (TXD PPS ground), but that has been within 
> one room.  I've used that to parallel the GPS 18x LVC output between two 
> PCs.
> 
> Cheers,
> David



More information about the questions mailing list