[time-nuts] NTGS50AA 1 PPS mod

EB4APL eb4apl at gmail.com
Tue Jun 16 21:20:18 EDT 2015


I has been making some tests with the 1 PPS output and here are the results:

Lady Heather cable delay commands works with both polarities, i.e. it 
can advance or retard both 1/2 and 1 PPS signals, I used as reference 
and external 1 PPS signal from a Rb oscillator.
The first impression is that the 1 PPS  edge leads the 1/2 PPS about 550 
ns and this difference is consistent after some cable delay commands and 
antenna disconnections, it is maintained during holdover.
Then I tried several power cycling and warm resets and this make an 
annoying thing to appear: during the acquisition and phase locking the 
difference jumped between 540 and 140 ns in 100 ns steps.  This is due 
to the 1/2 PPS synchronization with the SYS_CLOCK signal, so when the 
internal 1/2 PPS moved back and forth until the system phase lock is 
obtained, the output jumps between successive cycles of the SYS_CLOCK.  
The annoying thing is that its final state is not always the same, the 
final difference can be any of the mentioned steps, from 120ns to 550 ns 
and there is not guarantee which one is obtained while in my board the 
540 ns difference is the most common.  I don't know yet  it the 1 PPS is 
closer to the "epoch second" or it is the 1/2 PPS, I have to hook up a 
GPS timing module and an antenna splitter and see what happens. Anyway 
since Nortel specifies a tolerance of +/- 1 us of the 1/2 PPS with 
respect to GPS even second, any of the seen values are within specs but 
it is not very convenient for Time Nuts.


El 14/06/2015 a las 20:35, EB4APL wrote:
> Even if I get a cell site I would not use it for a "private network", 
> here all cell phones are GSM not CDMA.
> The only use for the 9.8304 MHz is as a master for deriving serial 
> comm clocks (i.e. 9600 is  9.8304 / 1024) but I don't plan to became a 
> "Serial Comm Time Nut" yet. ;-)
> Ignacio
> El 14/06/2015 a las 1:48, Bob Camp escribió:
>> Hi
>> Of course tomorrow you will stumble into a “great deal” on a complete 
>> cell site that needs a 9.8304 MHz clock :)
>> ====
>> One thing to watch:
>> The pps you now have may or may not be deterministic in its relation 
>> to the every other second output. It also may or
>> may not be in a fixed relation to GPS. I would bet money that it *is* 
>> in a fixed relation and that it’s actually better than
>> the other signal. Just because I believe it to be true does not make 
>> it true. It needs to be checked against something else.
>> Bob
>>> On Jun 13, 2015, at 1:56 PM, EB4APL <eb4apl at gmail.com> wrote:
>>> Hi,
>>> I just finish the mod.  It was easy, I cut the trace between TP14 
>>> and U405-6 and soldered a wire between TP14 and TP33. Now I have a 
>>> pretty 1 PPS on J5, the old 9.8304 MHz output. The signal has 0-5 V 
>>> levels, normally high with a 10 us pulse going down.  In my unit 
>>> this pulse leads the even second pulse by 539 ns.  I will check if 
>>> the Lady Heather command for compensating the cable length can be 
>>> used to move this if somebody needs a more accurate "epoch second". 
>>> I have to use the 1PPS from my FE5680A as a reference but now it is 
>>> disconnected.
>>> I have made a picture of the mod and I'll include it with my partial 
>>> schematic (I made some advances there) and the list of the TP 
>>> signals that I'm preparing for upload.
>>> I have checked that now I have also 4 additional 1 PPS outputs in 
>>> the 110 pin connector J2.  They are in the pins previously used by 
>>> the SYS_CLK signal.  They are differential LVDS as most of the 
>>> signals on this interface.
>>> Regards,
>>> Ignacio
>>> El 13/06/2015 a las 1:14, Ed Armstrong escribió:
>>>> Ignacio, I would very much appreciate a copy of whatever schematics 
>>>> you have, even if it is not guaranteed to be 100% accurate
>>>> I agree with you that the 9.9804 Mhz is basically useless, while 
>>>> the even second pulse is merely almost useless. However, as you 
>>>> have apparently looked the board over more carefully than me, you 
>>>> probably already understand why I did it the way I did. The 
>>>> location of the two output circuits were very easy to find, the 
>>>> path from the connector to them is quite distinctive. I just needed 
>>>> to find out where the signal got into the output circuit from, and 
>>>> when I flipped the board over, the trace bringing in the even 
>>>> second pulse was extremely obvious. There was no obvious trace for 
>>>> the 9.9804, and I didn't feel like probing all over the place and 
>>>> looking up a lot of chip numbers to try to figure out where it came 
>>>> from, as I have a very unsteady hand which makes poking around in 
>>>> these closely spaced components an invitation to disaster. So I 
>>>> just went with the obvious.
>>>> I found it interesting that the output circuit inverts the signal a 
>>>> few times. I actually would have preferred to invert it, so that 
>>>> the polarity was correct for a raspberry pie or a serial port under 
>>>> Windows, but it appeared some of the traces to do so were hidden in 
>>>> the layers of the board, and again the more I fool around the 
>>>> better my chance of shorting something out and becoming very unhappy.
>>>> I will be anxious to hear how your version of the modification 
>>>> works out, please do keep us posted.
>>>> I believe the antenna cable feed delay is going to work in the 
>>>> wrong direction here, I also seem to recall reading somewhere that 
>>>> the adjustment range may be limited. I did pretty much correct the 
>>>> offset by manually setting my position about 75M higher than what 
>>>> the device figured it to be, but I am concerned that would only be 
>>>> accurate for a satellite directly overhead, and may cause other 
>>>> inaccuracies by throwing off the geometry, especially for 
>>>> satellites close to the horizon. Based on what I am currently 
>>>> seeing from the Pi, I think the smart solution is to just ignore 
>>>> the offset altogether.
>>>> Ed
>>>> On 6/10/2015 11:30 AM, EB4APL wrote:
>>>>> Hi Ed,
>>>>> I am the one who discovered the 1PPS pulse while troubleshooting a 
>>>>> NTG550AA.  Instead of reuse the 1/2 PPS output and missing this 
>>>>> signal, my plan is to recycle the 9.8304 MHz output circuitry and 
>>>>> connector, the circuits are almost identical.  So I will cut the 
>>>>> trace that goes from TP14 to U405 pin 6 and also use a wire 
>>>>> wrapping wire to joint TP14 to TP33 so the 1PPS will be at J5.  I 
>>>>> think that I will do the modification this weekend.
>>>>> I don't imagine any future use of the X8 Chip signal but having 
>>>>> the even second output could be useful, at least to see the 
>>>>> difference with the 1 PPS.
>>>>> I had not measured the time difference yet, but I made a partial 
>>>>> schematic of the board for my troubleshooting and there I see that 
>>>>> the 1/2 PPS signal is synchronized with the 19.6608 signal that is 
>>>>> the source for the 8X Chip ( 9.8304 MHz), this is done in U405B . 
>>>>> The period of this signal is about 50 ns and this is the origin of 
>>>>> the 1/2 PPS width.  The 19.6608 MHz oscillator is phase locked 
>>>>> somewhere to the 10 MHz oscillator thus it is as stable as this one.
>>>>> I think that using the other half of U405, which actually is used 
>>>>> to divide by 2 the 19.6608 MHz signal, could render the 1 PPS 
>>>>> synchronized with the 1/2 PPS and also with the same width. 
>>>>> Probably the easier way to correct this is to use the command 
>>>>> which sets the antenna cable delay and compensate for the difference.
>>>>> I don't have a full schematic, even I am not sure that the partial 
>>>>> one is 100% correct but I can send it to anyone who wants it.
>>>>> Regards,
>>>>> Ignacio
>>>>> El 10/06/2015 a las 6:30, Ed Armstrong wrote:
>>>>>> Hi, this is my first post ever to a mailing list, so if I'm doing 
>>>>>> anything wrong please be gentle with your corrections :-)
>>>>>> A short time ago I purchased a Nortel/Trimble NTGS50AA GPSTM, I'm 
>>>>>> sure many on this list are familiar with it. At the time of 
>>>>>> purchase, my only interest was the 10 MHz output, for use with my 
>>>>>> HP5328b frequency counter and perhaps in the future also my 
>>>>>> signal generator. No question here, it just works great as is. 
>>>>>> However, it certainly seems best to leave these devices powered 
>>>>>> up all the time.
>>>>>> OK, now were getting close to my question. The unit pulls about 
>>>>>> 10-11 watts, which is really not very much. But it kinda bugs me 
>>>>>> to have it sit there using electric and basically doing nothing 
>>>>>> when I'm not using it. So, I bought a Raspberry Pi 2 with the 
>>>>>> intent of using it as an NTP server. I can't really say I'm 
>>>>>> enjoying my intro to Linux a whole lot, but I'll get there. It 
>>>>>> still needs some work, but it does function with the PPS output 
>>>>>> from an Adafruit ultimate GPS, which I bought for testing this 
>>>>>> and possibly building my own GPSDO in the future.
>>>>>> The NTGS50AA is a very capable device, but unfortunately it does 
>>>>>> not have a PPS output. Instead it has an even second output, 
>>>>>> which goes low for approximately 50 ns. The falling edge of this 
>>>>>> pulse marks the beginning of the second. During my search for a 
>>>>>> solution to this, I came across a post from this mailing list 
>>>>>> which I believe was discussing repair of one of these units. 
>>>>>> Someone in that post mentioned that there was a PPS signal at 
>>>>>> test point 33 which went low for about 10 µs. Thank you, that 
>>>>>> saves me a lot of probing.
>>>>>> The first thing I did was verify that this pulse did exist, then 
>>>>>> I decided to examine it a little closer. I kind of suspected that 
>>>>>> it may have been a rather raw pulse as received from the 
>>>>>> satellites. I found out that is not correct, once the unit 
>>>>>> successfully phase locks, this PPS signal is very accurately tied 
>>>>>> to the 10 MHz output, even when the unit goes into holdover mode. 
>>>>>> I was very happy about this :-) Next step was to see how 
>>>>>> accurately it was synced to the even second pulse. The bad news 
>>>>>> is that it does not occur at exactly the same time as the even 
>>>>>> second. The good news is that the offset is very consistent, 253 
>>>>>> ns before the even second pulse, +/- 1 ns.
>>>>>> My next step was to find out where the even second pulse entered 
>>>>>> the output circuitry. I then broke the trace taking the even 
>>>>>> second into the output circuitry, and ran a piece of 30gauge wire 
>>>>>> wrapping wire from the via at test point 33 to the via at the 
>>>>>> input to the output circuitry. The wire fit so perfectly it felt 
>>>>>> like the vias were made for just this purpose :-) Now I've got a 
>>>>>> very nice PPS signal available both at the front jack and at the 
>>>>>> backplane connector in the rear of the unit.
>>>>>> OK, here is the actual question. Do you think it is OK to 
>>>>>> consider a pulse which arise 250 ns early to be close enough? And 
>>>>>> no, I am not forgetting about that 3 ns, there is about 3 ns of 
>>>>>> delay added by the output circuitry.
>>>>>> Hope you didn't mind the long-winded post, and I thank you in 
>>>>>> advance for any advice you offer.
>>>>>> Ed
>>> _

More information about the time-nuts mailing list