I wonder why these Lucent units are so recommended on Time-Nuts. There seem to be a lot of compromises to them: 15 Mhz sine instead of 10, "dirty" 10 MHz output only available on redundant unit, low sensitivity 8-channel GPS, funky connectors (power and others), ... And so far the performance does not seem to match the alternatives out there. Am I missing something here?
BTW - my HUD has now risen to 6.5us. It was as low as 1.9us at one point. Not sure what's causing this? Maybe this is a normal pattern as the crystal breaks in?
If you want, you can go out and buy a brand new, shiny GPSDO with a 50 channel receiver that can hear a whisper from another planet. You probably don't want to know the price!
In the past, we were spoiled by units like the Trimble Thunderbolt and the other members of the Z38 series. Their performance is amazing. The Z3801 has a 6 channel receiver that needs a bit higher level. So what? It works great! The Thunderbolt only has an eight channel receiver. It doesn't need any more.
I don't think it's fair to say that the Lucent units were 'recommended' on the Time-Nuts mailing list. It wasn't discussed, but I suspect that people were hoping that these units would have similar performance to the Thunderbolt and other Z38xx units which are now either unavailable or uneconomical. They have good names behind them, a good pedigree and, of course, they were reasonably priced. Most of the chatter on Time-Nuts involved the reverse-engineering process and modifications to make the unit more time-nut-friendly. That's part of the fun! Until the units ran for an extended period, there was no way to know what its ultimate performance would be. I suspect that there's some disappointment within the list, but not enough to cause outright bitching and moaning. After all, these are surplus units rather than new and they were priced accordingly. They were just designed with different parameters than the Thunderbolt and the other Z38xx units.
The HUP will wander around a lot - particularly while the crystal is breaking in. On my well-broken-in Z3801, there's an interesting oscillation in the HUP. It goes from a high of 3 to 4 us down to a low of < 500 ns and back up. The period of this oscillation is about 7 days! This oscillation isn't visible on any other parameter in the box. I haven't attempted to get to the bottom of it. It seems like HUP values in the 5 us range are not surprising and there's significant unit-to-unit variability - probably determined by the characteristics of the particular OCXO in the unit.
These units do exactly what they claim. They provide an absolute frequency and time reference that's traceable to international standards and are within the reach of any hobbyist. For the majority of users, any imperfections will be irrelevant. If you're not satisfied with their performance, well, I guess that means you're a Time-Nut, too!
Ed