[ntp:questions] Windows time question.

Terje Mathisen "terje.mathisen at tmsw.no" at ntp.org
Sat Apr 23 16:52:18 UTC 2011

Richard B. Gilbert wrote:
> On 4/23/2011 2:52 AM, unruh wrote:
>> On 2011-04-23, David J Taylor<david-taylor at blueyonder.co.uk.invalid>
>> wrote:
>>> "Richard B. Gilbert"<rgilbert88 at comcast.net> wrote in message
>>> news:ot-dnVdAGogjOizQnZ2dnUVZ_o-dnZ2d at giganews.com...
>>> []
>>>> Anyone who doubts the accuracy, legitimacy, etc. of pool services
>>>> should
>>>> be able to purchase a GPS timing receiver and get the time from "the
>>>> horse's mouth". The investment required is about $100 US at a minimum.
>>> Even less - US $34 now....
>> He did say a "timing receiver". That could mean a GPS receiver
>> specifically set up to supply time, including an on board temp
>> insensitive crystal to supply the time if the sattelite availability
>> goes to low.
> There is no such thing as a "temp insensitive crystal"! A crystal can
> be placed in a tiny temperature controlled oven to achieve the *effect*
> of temperature insensitivity.

Not true:

Depending upon the cut, quartz crystals can be more or less 
temperature-sensitive, so it is in fact possible to get one that will 
wander less when the surrounding temperature changes.

A TCXO avoids the problem, as you noted, by having a 
thermostat-controlled oven which keeps the temperature more or less 

The most energy-efficient approach is to have a tiny temperature sensor 
on or near the crystal, then use a sw lookup table to compensate for 
temperature swings.

Garmin have used this approach for many years in order to get 
sufficiently stable 10 MHz time bases in their GPSs.


- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

More information about the questions mailing list