[time-nuts] Chinese NTP Server Part 2

[Brian Lloyd]

On Wed, Mar 4, 2020 at 10:26 PM xaos <xaos at darksmile.net> wrote:

> Hello everyone,
>
> So, the first Box I got was missing 2 parts. Namely the Voltage
> regulator and output driver.
> However, the OCXO was there. Weird...
>

Yes and no. Yes, there is an OCXO there but its output is square wave, not
sine wave. If you get the version that has the 10MHz output, they change
out the OCXO for one that has sine output ... and charge an extra $150.

In my case, I didn't care. Everything I have that uses a 10MHz reference
ends up squaring the signal anyway. So I just got the isolated switching
regulator and found a usable op-amp to use as a buffer.


> I got in touch with them and they seemed very nice. They sent me a new
> box and I've been running it for a few weeks.
>
> Brian Lloyd added the 2 parts to his unit and he was able to get a
> decent output wave. The weird thing is
> that his post might have been read by these guys, as the driver on my
> unit is actually sanded down (yes I said sanded!)
> so as to not show the part number !
>

Ha!

Ah, yours has a sine output!


> Here are some pics:
>
> https://www.maximaphysics.com/GPS.shtml
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>

I spent some time measuring the frequency accuracy and the accuracy of the
1pps compared to my T-bolt. The NTP server 1pps always leads the T-bolt by
anywhere from 20ns to over 80ns. I would not plan to use this for any sort
of accurate timekeeping, unless it was just something like a human clock
... or maybe an NTP server. ;-)

-- 



Brian Lloyd
706 Flightline
Spring Branch, TX 78070
brian at lloyd.aero
+1.210.802-8FLY (1.210.802-8359)