[time-nuts] Insulation - was: OCXO and Aging
shalimr9 at gmail.com
shalimr9 at gmail.com
Thu Sep 15 12:23:13 UTC 2011
This issue has been discussed at length several times. I remember in the last year discussions about using a large pool of water or concrete blocks to reduce temperature variations, and the discussions about overall thermal insulation (too much is not necessarily desirable since the oscillator power has to be dissipated) and reducing thermal gradients (most desirable).
Its all in the recent archives.
Does not mean that the subject should not be revisited with new information, but there is a lot of good stuff to read first.
Sent from my BlackBerry Wireless thingy while I do other things...
From: Bob Smither <smither at c-c-i.com>
Sender: time-nuts-bounces at febo.com
Date: Wed, 14 Sep 2011 23:36:09
To: Discussion of precise time and frequency measurement<time-nuts at febo.com>
Reply-To: Discussion of precise time and frequency measurement
<time-nuts at febo.com>
Subject: Re: [time-nuts] Insulation - was: OCXO and Aging
Will Matney wrote:
> The info they show is generally from the manual or paperwork that comes with
> the OCXO when new.
> If they are used OCXO's, then they will be more stable over time than a new
> one, off the shelf, as far as drift, or they should be. The OCXO acts
> similar to any heated device, and resistors function similar to this also.
> When first used, they will drift from spec more than they do after they have
> ran for so long in time. The factory generally does a burn in to stabilize
> them, but it still takes some time after that to completely settle down to
> where calibration isn't needed as often. There's generally a curve for each
> model that will show the drift over time.
> The main thing in using these, even though they're supposed to be insulated,
> is to mount them where temperature fluctuations are minimal. Also, some are
> better insulated than others, like the HP units, which are encased in foam
> inside a can.
I scanned the archives for this topic, so if I missed it and it has
already been beat to death, please point me to any previous discussions.
What is the list's opinion about putting OCXOs in an insulated
enclosure? That should reduce the effect of drafts in the lab. I have
some insulin shipping boxes and put an OCXO in one for a few days. The
OCXO is an C-MAC STP2145A. I measured the current draw of the OCXO and
the metal case temperature with the lid of the insulated box on and off:
Lid off: Is ~ 163 mA, Tc ~ 105F
Lid on : Is ~ 130 mA, Tc ~ 118F
I assume the oven operates at the same temperature in both cases.
Clearly there is less power needed as less heat is escaping. It looks
like the other circuit elements (outside the oven) are operating at
higher temperatures - likely not a good thing.
My measuring capability is limited. Within the precision (~10E-8) that
I have I could not see any difference in the oscillator frequency.
More information about the time-nuts