Clock runs way too fast

G

Guest

My computer clock is running 1 minute fast every three hours. It syncronizes
at most once a day at 'nist1-ny.WiTime.net'. The two other sites were always
giving errors when trying to syncronize.

Is there some way to syncronize much more frequently? Do I need to use
another site?

Thanks
 
D

db

it use to be that if a
clock seemed to be runnnig
away with the time, then
the likely cause was a failing
battery.

- db
My computer clock is running 1 minute fast every three hours. It syncronizes
at most once a day at 'nist1-ny.WiTime.net'. The two other sites were always
giving errors when trying to syncronize.

Is there some way to syncronize much more frequently? Do I need to use
another site?

Thanks
 
D

Detlev Dreyer

JJ said:
My computer clock is running 1 minute fast every three hours. It
syncronizes at most once a day at 'nist1-ny.WiTime.net'. The two other
sites were always giving errors when trying to syncronize.

Is there some way to syncronize much more frequently?

Why? The synchronization takes place once a week by default. Once a day
means that you're either running unnecessary third-party software or
that you've already changed the default interval by a registry tweak
unless you synchronize *manually* once a day.
Do I need to use another site?

Nope. You need to know the culprit since this problem is caused by poor
software running in the background. Norton/Symantec crapware for
instance is notorious to cause that problem, just one example:

"System clock loses accuracy under Windows XP after installing Norton
AntiVirus 2002"
http://service1.symantec.com/SUPPOR...85256edd00478df3?OpenDocument&src=bar_sch_nam
 
B

BillW50

D

db

his response is a pretty good one but
beyond the scope of mortals.

if i recall the formula is something like
when Voltage = Resistance x Current.

so if v decreases it is likely due to r decreasing.

therefore, the unregulated frequency of C is affecting the crystal and the time to
speed up then eventually slow down to
a halt.

Not unlike a sun becoming a quasar then a blackhole.

Honestly, Bob... Stop showing off...!

Bob I said:
Sounds like a out of spec crystal on the motherboard. But here is the
frequency setting.

http://www.pctools.com/guides/registry/detail/985/

Nice one Bob! But 99% of the time, a fast running clock means the
CMOS/Setup/Clock battery is running low. As the lower the voltage, the
faster the clock runs. Although the fanciest syncronizer in my mind is
the free Dimension 4. It shows a history and everything.

http://www.thinkman.com/dimension4/
 
B

BillW50

Well it is a bit more complicated than that. The simplest oscillator is
a cap and a coil in parallel. Give it a quick zap and it will oscillate
at a frequency determined by the values of capacitamce vs. inductance.
The oscillations will stop from loses without being refreshed by more
timed zaps.

Piezoelectric is like this (found in digital clocks), but it is like a
cyrstal rock between two metal plates. Smack it with a hammer (like in
those electric cigarette lighters), they will kick back with a voltage
(spark). Although kick it with a quick voltage spike and they vibrate.
And this vibating (oscillations) causes voltage oscillations as well.
And the lower the voltage, the weaker but faster oscillations they
produce. Thus cystal control clocks runs faster as the battery gets
weaker. Well until they get so weak that they can't cause it to vibrate
anymore.
 
B

Bob I

The CMOS battery voltage should only come into play when the system is
disconnected from the mains.
 
D

db

interesting.

so the synopsis is: Just replace
the motherboard battery?



Well it is a bit more complicated than that. The simplest oscillator is
a cap and a coil in parallel. Give it a quick zap and it will oscillate
at a frequency determined by the values of capacitamce vs. inductance.
The oscillations will stop from loses without being refreshed by more
timed zaps.

Piezoelectric is like this (found in digital clocks), but it is like a
cyrstal rock between two metal plates. Smack it with a hammer (like in
those electric cigarette lighters), they will kick back with a voltage
(spark). Although kick it with a quick voltage spike and they vibrate.
And this vibating (oscillations) causes voltage oscillations as well.
And the lower the voltage, the weaker but faster oscillations they
produce. Thus cystal control clocks runs faster as the battery gets
weaker. Well until they get so weak that they can't cause it to vibrate
anymore.
 
D

Detlev Dreyer

Bob I said:
The CMOS battery voltage should only come into play when the system is
disconnected from the mains.

Correct. It takes approx. 3 minutes to find that out.
 
B

BillW50

Yes, like you stated the first time. <grin> Although it is also a good
point if it is on the mains 24/7, the battery won't play a part at all.
Like mine is on an UPS and it will run the clock without power for weeks
or months (computer powered down of course). Although in my case, I do
hit the power strip (from the output side of the UPS if I am not using
my computers. So mine uses the battery to run the clock for about 8
hours a day.
 
A

Axalon

Hi JJ,
Try this out and see if it fixes your issue:

1. Start->Run cmd.exe
2. net stop w32time
3. w32tm.exe /unregister
4. w32tm.exe /register
5. net start w32time

It's basically a reset for the time synch.

Hope it solves your issue
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads


Top