[ntp:questions] Re: radioclk DCF doubling

Serge Bets serge.bets at NOSPAM.laposte.invalid
Sun Jul 17 21:42:46 UTC 2005


Hello Jonathan,

 On Sunday, July 10, 2005 at 10:59:54 AM +0100, Jonathan Buzzard wrote:

> Depending on your serial port it will be going from ground to
> something close to 10V, but the switch point between a zero and one is
> around 1.5V. Hence the leading edge of the pulse is detected quicker
> than the trailing edge.

How much delay does this effect add? Is it stable? People with reverse
polarity boards should use another time1 offset?


> On Sat, 09 Jul 2005 12:04:23 +0200, Serge Bets wrote:
>> pulses that should be in theory 100 ms (or 200 ms), when seen by
>> "radioclk -t" are always shorter, around 92.9 ms (or 190 ms).
> I considered it basically unimportant, as it is pretty much constant.

And anyway in normal operations radioclk 1.0 doesn't care about precise
pulse lenght nor falling edge.


> For meaningful statistics in addition to the min/max/mean you need to
> know what the standard deviation of the pulse lengths is.

Redone stats on new data, always in µs. Longer 1 pulses lenght seem more
accurate and stable than shorter 0 pulses:

| 0 values=1157, min= 87930, max= 97234, mean= 93021, stddev=1794
| 1 values= 476, min=185793, max=194271, mean=189817, stddev=1389


> the idea is to make a cheap, *SIMPLE TO CONSTRUCT*, reliable device
> that provides performance that at least matches other radio clocks. In
> my opinion all design goals have been met, and performance of the
> device outstrips almost all other radioclocks.

I fully agree. I'm very happy with your project. It permited to a
complete electronics illiterate that is also a poor man, me, to build a
very good refclock with it's own fingers and be proud of it. And total
cost of pieces was 17.27€ only. Mostly 10.76€ for the Conrad receiver,
4.40€ for the cable, 1.10€ for the DB9 connector, plus other components.

And I have left some cable and components to build a second refclock,
the day I can find an MSF receiver. Unfortunately near impossible to get
those here, even online.


BTW a bug report: radioclk 1.0 will not deal nicely with the added leap
second next December 31.


Serge.
-- 
Serge point Bets arobase laposte point net




More information about the questions mailing list