[time-nuts] Z3801A Problem

Ed Palmer ed_palmer at sasktel.net
Tue Nov 6 03:11:39 UTC 2012


Hi Azelio,

On 11/5/2012 4:09 PM, Azelio Boriani wrote:
> Very interesting... is it using the binary protocol?

Yes, standard Motorola binary format.

> Maybe a serial link
> error, the binary protocol has a checksum (yes, NMEA too). Check the serial
> link levels with a 'scope, maybe that the Z3801 firmware waits to see some
> consecutive errors before actually reporting that something is wrong,
> meanwhile the satellite number is set to 0.

The link is TTL, not RS-232.  The power supply to the VP is 4V93 and the 
logic levels on the scope are maybe 0V2 and 4V8 so no problem there.

I've tried disconnecting the 1 PPS and/or the data line from the VP to 
the Z3801A.  The Z3801A reports the loss of 1 PPS quickly, but in either 
case, waits for quite a while before getting upset.  I haven't been able 
to corrupt a message to see what effect that has, but I think it would 
have to be a very specific form of corruption to be accepted.  In any 
case, although that might explain the drop to zero satellites tracked, 
it wouldn't explain the situation where the VP reports fewer satellites 
tracked than the Z3801A does.  This occurs frequently.  I don't know if 
the two situations (drop to zero and fewer satellites) are related or not.

By the way, I was initially using SynTac and Z38XX to gather the data.  
Since I couldn't believe the results, I replaced both with a program I 
quickly hacked together to gather and parse the raw data from both 
units.  The results were the same.

Ed

> On Mon, Nov 5, 2012 at 10:38 PM, Ed Palmer <ed_palmer at sasktel.net> wrote:
>
>> I've had a Z3801A for about a year.  It's always had an issue where the
>> number of satellites tracked will drop to zero for one or two readings and
>> then jump right back where it was ( often at 5 or 6 satellites ).  This
>> often (but not always) affects the EFC and/or PPS and/or HUP.  My Tbolt is
>> running on the same antenna via an HP58536A splitter with no problems.
>>   I've changed cables and splitter ports with no improvement.  I've also
>> checked the  power supply voltages and the frequencies of the two crystals
>> on the Z3801A board.  All are fine.
>>
>> I assumed that the VP was flaky so I recently pulled it out for testing
>> and found no problems.  I then tapped into the link between the VP and the
>> Z3801A and found that the number of tracked satellites reported by the
>> Z3801A is often less than the number reported by the VP.  Specifically,
>> I've seen the VP report 6 satellites tracked at the exact same second that
>> the Z3801A is claiming that there are no (zero) satellites tracked.
>>
>> Does anyone know if there are any hidden parameters that the Z3801A is
>> using to decide if a satellite is acceptable?  I looked through the manual
>> but I couldn't find anything.  Is there a command I'm missing?
>>
>> Thanks,
>>
>> Ed




More information about the Time-nuts_lists.febo.com mailing list