[time-nuts] Simulating LORAN-A GRI Timing Generator

Magnus Danielson magnus at rubidium.dyndns.org
Sat Jan 8 06:27:28 UTC 2011

On 08/01/11 06:54, Magnus Danielson wrote:
> John,
> On 08/01/11 05:04, J. Forster wrote:
>> Magnus,
>> See my post of a few minutes ago.
>> The GRI and Delay counters need to be 5 decimal digits. My current
>> thinking is manual entry of the numbers via BCD coded thumbwheel
>> switches.
> If you only need a static location that will be fine. Naturally you tune
> it up to the ships current (final?) position. Good exercise for the
> visitors. :)
>> LORAN-A has 24 possible GRIs; in three blocks of 33 PPS; 25 PPS; and 20
>> PPS. It also has 4 RF frequencies.
> I did a fashinating read-up here:
> http://www.jproc.ca/hyperbolic/loran_a.html
> GRI is a LORAN-C term if I get it correctly, where as PRR is used
> together with frequency indication in LORAN-A.

Pondering some more on this, doing some calculations (spread-sheets) the 
actual scheme emerge... the range of 0 to 7 is not selected by accident.

The base pulse repetition rate is 30 ms, 40 ms and 50 ms. The pulse 
repetition rate modifiers modifies the repetition rate by -100 us per 
step. It would only take two (or possibly three) wheels to select PRR.
This explains the high rate starts with 33 3/9... it also makes the 
receiver design easier to understand.

So a 10 kHz source would suffice for PRR settings. A phantastron divider 
from a 100 kHz standard should do it in those days technology, if not a 
10 kHz standard was used directly.

Only the delayed pulse would need the full set of thumb-wheels.


More information about the time-nuts mailing list