[ntp:questions] Re: CTS, IRIG-B, and a Sun v210
David L. Mills
mills at udel.edu
Fri Apr 15 15:06:17 UTC 2005
Martin,
The IRIG audio decoder autodetects IRIG-B (1000 Hz carrier) and IRIG-E
(100-Hz carrier. There are no provisions for DC shifts, as there are no
common I/O devices available for PCs and workstations that work with
those signals. It would in principle be possible to adapt the audio
driver to work with an A/D converter and DC shift signals, but that
would seriously degrade accuracy.
Once upon a time the IRIG audio decoder was the only driver that
seriously challenged the PPS performance with accuracy in the low tens
of microseconds. Unfortunately, at least with modern Sun hardware and
Solaris kernels, the accuracy has been seriously compromised to the
order of a few milliseconds. I would welcome reports on how well it does
with other hardware and other kernels.
Dave
Martin Burnicki wrote:
> Bjorn,
>
> Bjorn Gabrielsson wrote:
>
>>IRIG-Bs that I have used has been _analog_ signals - think sinus signal
>>tone. RS-422 is a differential _digital_ standard - either "0" or
>>"1". IRIG-B can be feed to a computer via an audio input.
>>
>>I'd suggest, some more reading in the manual and hooking up an
>>oscilloscope to the signal, to see what it really is.
>
>
> That's not quite the truth.
>
> "IRIG-B" is just what many people are talking about. In fact, the "B" just
> specifies the data transmission rate.
>
> The data signal can either be transmitted modulated on a sine carrier (what
> you're referring to), or unmodulated (DC level shifted, DCLS) which is a
> pure digital signal which can well be transmitted via RS-422 or other
> digital signalling modes.
>
> Normally you had to specify the IRIG code like "IRIG-Bxyz", where:
>
> "B" stands for 100 pulses per second data rate
> "x" is "0" for DC level shift or "1" for modulated carrier
> "y" determines the carrier frequency, where "0" is DCLS
> "z" indicates what type information is contained in the IRIG signal
>
> Example code specifiers look like B122 (modulated) or B002 (DCLS). For more
> detailed information, please see:
> http://www.meinberg.de/english/info/irig.htm
>
> I know it should be possible to decode modulated IRIG using the audio driver
> (haven't tried it, though), but AFAIK there's no driver in NTP which
> directly decodes the digital DCLS signal.
>
>
> Best regards,
>
> Martin
More information about the questions
mailing list