[time-nuts] Chinese NTP Time server

[Brian Lloyd]

OK, I tried the AD8055A op-amp in the buffer. Output now looks exactly
like the output of the OCXO (square wave). Unloaded output is 3.2Vpp.
Output into 50 ohms is +10dBm, nice and clean. So, if you buy one and
want to add the components to get 10MHz sq-wave output, you need to add
a TBA 1-1219 isolated DC/DC converter, and an AD8055A op-amp. Works
great. Mouser has them in stock. While I do have hot-air SMD-rework
capability, the SOIC-8 package was easy to hand-solder with a small tip
on my Hakko solder station so I didn't have to go get solder paste.

BTW, the price quoted for a unit WITH 10MHz output was $350,
substantially more than the $201 for the non-10MHz unit. For $10 in
parts I am happy. OK, it isn't sine-wave output. I don't have anything
that requires a sine wave reference.

So this meets my needs for NTP stratum-1 for the machines in my house,
and 10MHz for my radios.

Still, I wish to graduate to full time-nuttery status. I would like to
do the ADEV analysis. Other references I have kicking around:
Thunderbolt, Leo Bodenar GPSDO, Ephratom LPRO-101 Rb. Unfortunately I
don't have a good, free-running OCXO for short-term comparison. Still, I
suspect the Rb should work well enough for that. Also, the only device I
have to DO the comparison is a Fluke T7260a. Thinking about a TICC. Advice?

Brian, WB5BL
-- 


 <https://www.lloyd.aero>

Brian Lloyd
706 Flightline
Spring Branch, TX 78070
brian at lloyd.aero <mailto://brian@lloyd.aero>
+1.210.802.8359