[time-nuts] Efratom LPRO Rubidium

[Charles S. Osborne]

I have one of the Lucent RFTGm-II-Rb Rubidium cell site standards. It's
misbehaving. Knowing the amount of experience on this list, I figured I'd
ask before I dive in much deeper where I have no schematics.

Initially I believed one of the DC-DC converters was not working after
warmup. But after remoting the voltages to test points, they seem to be OK.

The system did suffer a lightning strike some years ago, which took out the
UT+ GPS receiver in the RFTGm-II-XO. But that's working after the UT+ was
replaced. And I doubt the interconnected circuits would have been affected.
Power supply for instance was floated across 24VDC of gell cell batteries,
which usually can absorb a lightning spike pretty well.

The unit powers up about +200 Hz  above 10 MHz (mine is modified to no
longer put out 15 MHz). After 30 minutes it has dropped slowly to +150 Hz.
It briefly shows an online indication then. But a few minutes later that
goes off and the fault light comes on. The frequency drops as low as +129 Hz
as things continue to heat up. A couple of days later its still in the +129
~+149 Hz range. I can heat and cool it and move the frequency ( +157 Hz at
90°F inside the Lucent box, +129Hz at 110°F), so the loop definitely isn't
in control.

I also have its companion RFTGm-II-XO connected to it with the factory
wiring harness. The XO is GPS locked and working fine exceeding my ability
to read its error accurately but sub-milliHz. Counter is clocked by Z3801A
10 MHz as reference.

Lucent diagnostic monitor software says "unit failed", "oscillator
maintenance required". My question then, is there anything one can access
that might be adjusted to bring the OCXO back within capture range? +140 Hz
seems way beyond where it should be. I thought the OCXO tuning range on
something like this was normally +/- a few Hz.

The unit is circa 1999 vintage. Low mileage, probably has two years online
time to date (so likley not an Rb lamp problem). When I bought it, it was
new in the box and still in warranty, victim of a cell site buildup deal
gone bust and surplused out. At one time the two units would back each other
up, switching from one to the other once as they warmed up.

Anything user servicable in the LPRO? I assume a trip to the real service
center would cost as much as a good used car. So if its something I can't
correct, it'll become a dissection learning experience.

One final test I haven't run, is removing the LPRO and powering it up stand
alone, just in case the C field input is the problem. I assume it should be
capable of locking with those inputs floating if the LPRO is indeed OK?
Anyone know what: BITE (Built in Test?), Xtal Mon, Lamp volt, and C-Field
should be normally?

Thanks,
Charles Osborne, K4CSO
Duluth, GA