Crystal Oscillator Drift

[uniserver]

from watching this video:

(orginally from technight in the Homebrew PRAM battery thread)

different types of Oscillators:

XO

TCXO

DTCXO

OCXO

RBXO

I'm just trying to figure out. Is his goal of synchronized data logging have to be solely based on super accurate oscillators?

So I think he said these loggers are about 50 meters apart, and they lay down 20 of them on the sea floor.

I am wondering if there is some kind of wireless technology or maybe even wired, where even if they drift they can all be synchronized,

as long as all the data logging across all units is synced up. Increasing battery life and decreasing cost.

Over all though, I would have to assume with technologies like GPS, Cellular and Internet. More expensive oscillators might not be used as much these days because they have more options now-a-days to time sync?

Or is it the opposite? High quality oscillators are much less expensive then what they used to be, there for, devices are naturally much more accurate?

And I guess my other question is regarding your mac and time drift, have any of you installed a known working pram battery in your mac, then let it sit in your closet for a couple years, booted it up and either noticed the time was right on or it had drifted massively + or - ?

Also I wonder how much oscillator drift effects modern screaming fast computers.

 

[James1095]

The cheapest easiest way to get a super accurate oscillator these days is to use a GPS receiver. Rubidium oscillators from decommissioned analog cellular sites are relatively easy to find and also extremely accurate. I'd think there are very few applications where this is really critical though. In my experience, the RTC chips in old Macs are quite stable, drift of a few minutes per year seems typical. Since I don't generally use old computers as a means of telling time, I never found this to be a problem.