[time-nuts] Simulating LORAN-A GRI Timing Generator
jfor at quik.com
Sat Jan 8 15:28:36 UTC 2011
> Loran c was complicated by the fact that there were phase reversals in the
> patterns that had to be emulated it doubled the size of the code. But to
> answer your question John its all done in simple readable basic language.
> Think of it as a endless loop pattern. The real trick is using a good
> reference. I think Magnus said 10 KC. What you do is count the 10 KC
> reference. Using a Parallax SXB chip that can run all the way up to 80
> The code runs so fast that 99.9% of the time the micro is just waiting
and counting edges.
OK. Thanks. You talk to the uP with a dumb terminal program?
> I like to create dumb simple stupid, reliable as heck programs like that.
> The Loran C sim allows the Austrons to compare to their limit of 1 e -13.
> There is a bit more cleverness in the C simulator external to the
> I thought there were different frequencies at about 1.8 Mhz.
There are 3 between 1.8 and 2.0 MHz, one higher that was mostly unused.
> I remember the gone-ometer (I still think thats a disease sailors can
> get), green crt black box about 2 ft tall.
That's for a DF loop, no? Was that used on the LORAN receiver? If so,
that's news to me. The crossed roughly 4' loops were common on ships, but
I didn't think they were LORAN antennas.
> If you really wanted to do a location you might need 2 or 3 chips at
> different GRIs and phases. But at $3.56 not a really big deal. Lastly
> must be some overall sync relationship between the chains that has to be
> maintained I might guess.
I'm not sure about inter-chain sync, but IMO it does not matter much.
> Oddly enough I emailed the group several times at the ship and never
> received a response.
> Figured they were all set.
If you're interested, I can make sure you talk to the right person.
More information about the time-nuts