[time-nuts] Z3805A cooling requirements?

Volker Esper ailer2 at t-online.de
Sat Dec 22 03:55:19 UTC 2012


Although I am dog-tired it gives me no peace...

I come to the following conclusion:
- The long term Allan deviation gets worse, as long, as the effect of 
EFC compensating is in the range of tau
- But: It gets back to its normal value after that
- The short term deviation, however,  increases slowly, but it doesn't 
settle. It's increasing more and more.

So it is the short term stability, that is affected, rather than the 
long term stability.

Am I right or wrong?

Perhaps I'm to tired to decide.

Volker



Am 22.12.2012 03:54, schrieb Volker Esper:
> Here come the curves...
>
> You'll find two diagrams. The first (named "1_DF9PL...") shows five MDEV
> curves (Modified Allan Deviation), each of them measured at different
> times. Total time span is 30.5 hours.
>
> At small tau values (up to 1000 s) only a slight increase of sigma over
> time can be noticed. However, at a tau of 5000 s or greater you can
> watch sigma making a big bump. Ok, that's what we expected before.
>
> In diagram no. 1 it's somewhat fussy to recognize the change of a
> particular sigma(tau). Now, that we've got curious, we want to see, how
> the sigma(tau) changes over time. So I've been providing a second
> diagram ("2_..."), where sigma(tau) is a function of the time.
>
> You can see, for example, the curve of tau=20480s developing a big hump,
> and falling back to a proper value after about 1800 minutes. All curves
> at a tau greater or equal 2560 do so.
>
> At smaller values the curves are esentially less affected, but - they
> are not back at their starting value after 1800 minutes (30 hours)! You
> could guess, that the hump moves up to longer times with increasing
> sigma - but it doesn't. There is something significantly different below
> tau=2560s.
>
> What is it?
>
> Volker
>
>
>
> Am 21.12.2012 14:42, schrieb Volker Esper:
>> (oh, I've got to split the posting)
>>
>>> The picture enclosed can give you a first impression. What we see is
>>> the difference time between the GPS signal and the OCXO (blue)
>>> ("PPS-TI"), which is an HP 10811. In red we can see the EFC. The total
>>> span is 24 h.
>>>
>>> Before I applied the fan, the noise was at a maximum of about +/- 20 ns.
>>> Some hours after starting the fan the noise is much greater. That should
>>> have a significant impact on the ADEV.
>>>
>>> I don't put the ADEV curves here, I make up for it when the EFC
>>> compensation is completely out of the scope, that will be in about 12
>>> hours. I don't have the ADEV at 1 s, but the ADEV at 10 s has been
>>> almost constant. The ADEV at about 1000 s has a nasty bump now.
>>>
>>> IMHO that fits to the physical facts: the airflow will surely not affect
>>> the 10 s ADEV since the OCXO tries its best to isolate the oscillator
>>> from short time temperature influences. However, the turbulent air flow
>>> that I applied will influence the longer time ADEV.
>>>
>>> Have a nice solstice
>>>
>>> Volker
>>>
>>>
>>>
>>>
>>> Am 21.12.2012 12:44, schrieb Volker Esper:
>>>>
>>>> Yes, I made such a setup, it's now running 22 hours. I'll post the
>>>> results in two hours or so (if nothing evil happens to the earth,
>>>> meanwhile).
>>>>
>>>> Volker
>>>>
>>>>
>>>> Am 21.12.2012 03:35, schrieb SAIDJACK at aol.com:
>>>>> Wish I had more time to play with this setup.
>>>>>
>>>>> How about fellow time nuts spend some time and present similar test
>>>>> data on
>>>>> their OCXO's to compare?
>>>>>
>>>>> I was interested in the 1s to 100s ADEV, and my runs were from 8
>>>>> minutes to
>>>>> 20 minutes, certainly enough time to capture data for 1s to 100s ADEV
>>>>> measurements..
>>>>>
>>>>> bye,
>>>>> Said
>>>>>
>>>>>
>>>>> In a message dated 12/20/2012 14:17:59 Pacific Standard Time,
>>>>> magnus at rubidium.dyndns.org writes:
>>>>>
>>>>> On 12/20/2012 01:34 PM, Bob Camp wrote:
>>>>>> Hi
>>>>>>
>>>>>> Temperature transients are not a good thing for an OCXO. If you
>>>>> deliberately use the fan to create a transient, then yes the OCXO will
>>>>> not be
>>>>> happy. The question it - what happens after the transient has settled
>>>>> out? The
>>>>> plot you have still looks a lot like a step function.
>>>>>
>>>>> I agree. Temperature steps stresses the OCXO oven loop and easily
>>>>> creates a gradient over the crystal. As the oven loop tracks in, the
>>>>> frequency returns to around normal. The trouble with forced air over a
>>>>> crystal is that the metal shield couples very well and acts like a
>>>>> heat
>>>>> sink. A think plastic cover over it and forced convection doesn't have
>>>>> the same effect. There is even being used by at least one vendor.
>>>>> Works
>>>>> very well for the extra cents of manufacturing cost.
>>>>>
>>>>> The HP10811 is recommended to be put in a airflow-quiet corner of the
>>>>> world. Look at it's mounting in the HP5370A/B for instance.
>>>>>
>>>>> Cheers,
>>>>> Magnus
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> time-nuts mailing list -- time-nuts at febo.com
>>>>> To unsubscribe, go to
>>>>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>>>>> and follow the instructions there.
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> time-nuts mailing list -- time-nuts at febo.com
>>>> To unsubscribe, go to
>>>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>>>> and follow the instructions there.
>>>>
>>>>
>>>
>>
>>
>>
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.





More information about the time-nuts mailing list