[time-nuts] A couple of Tbolt questions

[Arnold Tibus]

On Fri, 24 Jul 2009 12:18:21 -0700, Tom Van Baak wrote:

>> Unfortunately I can't help with the phase noise measurements either right  
>> now but it does seem like an area worthy of further investigation.
>> 
>> regards
>> 
>> Nigel
>> GM8PZR

>Since I have fancy phase noise systems and references here
>I ran both ADEV and phase noise measurements on *every*
>TBolt to make sure the ones that went to TAPR were all good.
>(Now you know why months go by between TAPR group buys).

>Typical TAPR Thunderbolt phase noise is about -100 dBc/Hz
>at 1 Hz, falling to about -150 at 100 Hz, with a floor just above
>-160 beyond 10 kHz. Variations on this, and spurs, seem to be
>a function mostly of power supply. For example, see:
>http://www.leapsecond.com/pages/tbolt/noise.htm

>I would be very surprised if *any* TAPR TBolts showed much
>worse or much better PN performance that what you see at:
>http://www.leapsecond.com/pages/tbolt/spec-hp.gif

>I can't make any claim about the untested TBolts that you see
>on the global surplus market. A number of them have problems
>and they go on my discard pile.

>If you have a pair of OCXO that you'd like me to measure here
>in my lab, contact me offline. I take equal delight in unusually
>good as well as unusually bad examples. Arnold, I'd be happy
>to independently test your two samples. Perhaps this would
>shed light on what that little "phase noise" label means.

>/tvb
-------------------------------------------------------------------------------------------------------

Tom, 
I just got your message and I appreciate your offer very much! 

I have both OCXOs in front on my table and
I thought to dig out the pinout of them (3 pins I don't know yet, 
I will have to measure at the live card) and then to make a setup
with a Dual Mixer Phase Measurement System. But before I am 
looking to get or build a really very low noise 5 Vdc power supply 
and as well a stable heater supply. I do mistrust the switching 
supplies...

So it would be great to do the professional measurement - 
the bad point is the distance between us and the shipment
 inconvenience, if you were my neighbor I would have already 
asked you. You have all my confidence. 
The Idea is very good indeed! 

The OCXO with the ominous or misunderstandable sticker is from 
that TB with the not (correct) working temperature measurement 
system. I did not want to think in this direction like Nigel, but I had 
already a bit the feeling that perhaps the stability is not so good
as I wished, but feeling or believing... not good to  

That is the box where I (therfore) decided to exchange with the 
10811-60158 from Fluke I. The point is that even this one may be 
not ok either. But all this is not definitive until verified. 
In contrary, I know well that you did take over a big task to do the 
job to verify all the items leaving your desk!

The second TB from your last shipment I started to run with LH 
just some hours ago. This one seem in fact to behave somewhat better. 
But you made already the valid statement about the real measurement 
with an external reference that only does count...

Let me think about a moment, you took me by surprise and I am 
too tired already this late hour to get something valid out of my brain...

Thank you very much for you proposal Tom,

regards
Arnold