[time-nuts] Thunderbolt stability and ambient temperature

[Poul-Henning Kamp]

In message <BLU125-W9E23FDF3E1618C93E603ECE440 at phx.gbl>, Mark Sims writes:

>I think that using a well insulated box [...]

The point for timenuts is not just the thermal resistance, but more
importantly the thermal *impedance*: you want to low-pass filter
the thermal changes so that they all happen in the area where the
PLL can cope with them.

Thermal resistance is about insulation, thermal impedance is
about (thermal) mass.

So you significant mass and volume (like a fridge) not light and
small (like a cardboard box).

When metrology people really want to keep things at the same
temperature, they mount them in oil-baths (for good thermal contact)
in the middle of a block of aluminium, typically 2'x2'x1' (for
thermal impedance) which is again insulated with 1" styrofoam, all
of this mounted in a plywoodbox, set on rubber-wheels to get it off
the floor (for thermal resistance).

If they are really into this, they cover the plywood with high-quality
(noble-) metal foil, to maximize reflectivity and minimize emissivity,
so that the black-body radiation from devices and humans in the lab
does not affect the temperature interface as much.

Then they leave it alone for "some weeks" in their temperature
controlled lab so the temperature can stabilize.

At this point they may start to wonder how they can verify the pt100
temperature sensor they put in the middle of it all actually works
when the temperature never changes...

Poul-Henning

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.