[time-nuts] Beginner question - unexpected possible jitter in 1 PPS output of Motorola ONCORE UT+ module

[David Feldman]

I'm trying (in lieu of setting up a GPSDO) to use the 1 PPS output of a Motorola ONCORE UT+ module to do a one-time reference oscillator adjustment of my frequency counter (setting its reference oscillator which has a small trimmer).

My counter has  a "totalize" mode where it counts pulses from zero up, until the pulses disappear or the counter is reset (with a front panel button). I built a small CMOS divider connected to the UT+ 1 PPS output, intended to generate 0.05 Hz at 50% duty cycle (this is triggered on leading edge of the 1 PPS output, so it should not be affected by the imprecise duration (pulse width) of the 1 PPS signal). This (0.05 Hz signal) gates the reference oscillator output from my frequency counter, 10 seconds on (counting up), followed by 10 seconds off (not counting) into the counter input in totalize mode. I'd expect the counter in the totalize mode to count from 0 up to 130000000 during one 0.05 Hz cycle (10 seconds of 13 MHz, the ref osc frequency for this counter.)

Here's the problem: my counts vary significantly (+/- 10000 or more) from cycle to cycle, which is much higher than expected based on 50 nS period uncertainty of the 1 PPS output noted in UT+ specification (I think the 50 nS uncertainty should apply to the overall 10-second count duration in my circuit.) I'm using the ONCORE ENGINEERING NOTE as my documentation.

I am using TAC32 to monitor operation of the UT+ and it indicates (after about 10 minutes acquiring satellites) 7 or 8 satellites being tracked (with Eb/No indicated for each one.) Often (a few times per minute) one satellite signal seems to disappear, another is acquired in its place. My grid square is correctly indicated once 4 satellites are being tracked. The unit appears to be in the position hold mode by default. The module was bought on ebay and I do not know its prior history. The UT+ is powered with 5V taken from a small USB power supply (same result with two different power sources.)

My question is this: Under what conditions can the 1 PPS signal be relied on to meet its period uncertainty spec? The UT+ emits 1 PPS regardless of whether satellites are visible or not (it does this as soon as power is supplied, long before any satellites are acquired); the documentation doesn't seem to elaborate under what conditions the 1 PPS signal can be considered to be within the period spec, which is why I'm posting here. 

Thanks!