[time-nuts] UBLOX LEA-5T Programming?

Bob Camp kb8tq at n1k.org
Fri Oct 9 18:02:19 EDT 2015


Hi

The PPS out of the GPS has a number of issues short term. Without sawtooth 
correction it likely is hopping and bopping 5 or 10 ns each second. Looked at as
frequency, the 1 Hz is +/- many ppb. 

The “frequency” output inherits this problem and adds the issues associated with pulse
drop frequency synthesis. More or less 10 MHz is a 16 MHz pulse train with 6 our of 16 
edges dropped out. That does not give you a clean spectrum or a predictable signal. 

Combine these two things and you have a signal that most PLL chips will be a bit bothered by
*and* a reference that is much less stable than the OCXO. The answer is to stretch out the 
comparison process to 100’s of seconds and to use sawtooth correction. Both of these things
are easily done with a  < $2 MCU. The gotcha is that you need code to go with it. 

A $25 Ref-0 with the same GPS and a (now) documented MCU will do the same thing as
any other GPSDO. That sort of sets an upper limit on how much you probably should spend
on this sort of thing. That’s not to say that similar logic *does* put an upper limit, as I’ve proven on a
large number of projects I’ve done …..

Bob

> On Oct 9, 2015, at 10:16 AM, Clint Jay <cjaysharp at gmail.com> wrote:
> 
> I am still learning and want to understand, if the PPS is good then why is
> the programmable output bad, as I understand it thus far, the PPS is
> derived from the same clock source or have I got that badly wrong?
> 
> On 9 October 2015 at 12:16, Bob Camp <kb8tq at n1k.org> wrote:
> 
>> Hi
>> 
>> Doing a GPSDO by locking to the awful 10 MHz output from any of these GPS
>> modules
>> is not going to work very well. Given the very long time constants
>> involved in a GPSDO
>> control loop, doing it without code is going to be pretty difficult.
>> 
>> A much easier approach:
>> 
>> Grab the GPS PPS and a scope. Use it to set the Lucent boxes on frequency.
>> Repeat the process
>> once a week.  The direct lock will flop around by (likely many) ppb. Your
>> manual set will probably keep it
>> under 1 ppb for a week.
>> 
>> Bob
>> 
>>> On Oct 9, 2015, at 1:39 AM, Perry Sandeen via time-nuts <
>> time-nuts at febo.com> wrote:
>>> 
>>> Hi,
>>> I have six of the older style (anodized solid aluminum case) Lucent
>> crystal oscillators that I’d like to turn into simple 10 MHZ output
>> GPSDO’s. I have some salvaged HP cases so I can have a linear PS and all
>> other stuff in one package.
>>> I’ve found two different choices of GPS relatively low cost units on
>> Ebay.
>>> One is a LUCENT M12+ GPS Timing Receiver W/100 Hz output  for  $19.99
>> Item 171886538138.
>>> The other is a UBLOX LEA-5T GPS module dev board 1PPS /USB/RS232/ntp ser
>> with two 1 PPS outputs for $49.90. Item 251785217093.
>>> IIRC, one output of a Ublox can be set to a 10 MHz.
>>> Can this model be set to have a 10 MHz output and is it easily done from
>> a PC? I have no coding ability.
>>> For more money I’d get a simpler phase detector circuit, which is all
>> I’m looking for. For less money I’d have to build more circuitry.
>>> A factor I don’t know is if the ublox is that much better/newer to
>> justify the price difference.
>>> I already have packed away two Oncore units with either a 1 KHz or 10
>> KHz output.
>>> Suggestions welcomed.
>>> Regards,
>>> Perrier
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at febo.com
>>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>>> and follow the instructions there.
>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>> 
> 
> 
> 
> -- 
> Clint.
> 
> *No trees were harmed in the sending of this mail. However, a large number
> of electrons were greatly inconvenienced.*
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.



More information about the time-nuts mailing list