Losing time

[Alain Dekker]

Is there any good reason why some of the industrial machines we use run spot
on time (08:00 is correct every morning) and other lose as much as 2hrs in a
day? Machines are a combination of XP SP2 and SP3 and a mixture of Pentium 4
2.8Ghz single core and Intel i3 dual core 2.4Ghz, but are basically
configured almost indentically. Some systems with exactly the same hardware
and OS behave differently, losing differtent amounts of time over a day.

Thanks,
Alain

 

[MyNews]

They all need to be update to XP sp3

Support is ending for some versions of Windows
Support for Windows Vista without any service packs ended on April 13, 2010.

Support for Windows XP with Service Pack 2 (SP2) ended on July 13, 2010.*
< http://support.microsoft.com/kb/322389 > chick link for more info!

If you're running one of these versions after support ends, you won't get
security updates for Windows

The one losing time my needs

New Lithium Battery 3v
they run For about $2 to $3 a piece!

for they are getting 3 to 4 your old
will lose charge and you lose time!

 

[Paul]

Is there any good reason why some of the industrial machines we use run spot
on time (08:00 is correct every morning) and other lose as much as 2hrs in a
day? Machines are a combination of XP SP2 and SP3 and a mixture of Pentium 4
2.8Ghz single core and Intel i3 dual core 2.4Ghz, but are basically
configured almost indentically. Some systems with exactly the same hardware
and OS behave differently, losing differtent amounts of time over a day.

Thanks,
Alain

A good reason for a computer to be synced exactly, is it is accessing
an NTP server on the net, and correcting the time on a regular
basis. That computer, will then seem heroic. I have NTP enabled on
this machine, and the current time error is 3 seconds since last
sync (yesterday). NTP not only allows a single point correction
to the time, but it is also possible to compute a "drift rate", and
make corrections more continuously (I don't know whether the
Windows version does that or not).

When the computer is off, the time keeping is handled by the RTC
and the cylindrical 32768 Hz quartz crystal on the motherboard.
That is the same crystal, used in a digital watch.

When the computer is running, the OS reads the RTC, and copies the
time into software. From that point onward, time keeping is maintained
by counting clock tick interrupts. The drift of clock tick interrupts
is traceable to a different quartz crystal on the motherboard in
that case, than the RTC one.

If something interferes with the ability to service clock tick interrupts,
a one-sided error in time keeping will result (slowdown). The clock tick
interrupt has a relatively high priority, but there are other hardware
mechanisms available on a PC, that can "steal" time. Intel invented
the SMM feature, which is virtually undetectable when the computer
is running. That's just one example of something that affects the
computer, but you can't do anything about it.

http://en.wikipedia.org/wiki/System_Management_Mode

"This can destroy real-time behavior and cause clock ticks to get lost."

The Nforce2 based motherboards had some problems with time keeping,
but in that case, it seemed to be a bug involving the APIC.
It was some kind of interrupt issue. The bug showed up, when the
chipset was set to a non-canonical clock rate (typically, something
an overclocker might do). The drift was so large, and was also
+ and - the correct time value, that even with NTP enabled and
the update rate cranked to the wall, it still couldn't make
a good time piece from the computer. Huge errors could result.
So it is possible for a hardware bug, to trash the time keeping
of a computer.

Modern computers have a large number of timers and time keeping
options. You're likely to find a good overview of that topic,
on one of the virtual machine web sites, as virtual machine
design has to emulate some of those features, and deal with the
errors that result from less than perfect emulations. I don't think
I could do a good enough job of describing them all.

Paul

 

[Yousuf Khan]

Is there any good reason why some of the industrial machines we use run spot
on time (08:00 is correct every morning) and other lose as much as 2hrs in a
day? Machines are a combination of XP SP2 and SP3 and a mixture of Pentium 4
2.8Ghz single core and Intel i3 dual core 2.4Ghz, but are basically
configured almost indentically. Some systems with exactly the same hardware
and OS behave differently, losing differtent amounts of time over a day.

Thanks,
Alain

It just happens with some machines. It doesn't have anything to do with
old CMOS batteries or anything like that, like some here are telling
you. The internal clocks are not quartz-precise, that's all. You can
make the machines better than quartz-precise by installing an NTP
(Network Time Protocol) software on the machines. That way, you can sync
them to any NTP server on the Internet. A good freeware utility is the
AboutTime. You can set it to update itself on any amount of elapsed time
you like, a good amount would be once every one or two hours. This
AboutTime software also acts as an NTP server, so you can have one
machine syncing itself to the Internet NTP servers, while the rest of
your machines sync themselves off of this machine.

Yousuf Khan

 

[glee]

Is there any good reason why some of the industrial machines we use
run spot on time (08:00 is correct every morning) and other lose as
much as 2hrs in a day? Machines are a combination of XP SP2 and SP3
and a mixture of Pentium 4 2.8Ghz single core and Intel i3 dual core
2.4Ghz, but are basically configured almost indentically. Some systems
with exactly the same hardware and OS behave differently, losing
differtent amounts of time over a day.

Are they all kept on 24/7, or are they shut down at night?

 

[BillW50]

In MyNews typed on Mon, 10 Jan 2011 07:04:27 -0600:
[...]

New Lithium Battery 3v
they run For about $2 to $3 a piece!

for they are getting 3 to 4 your old
will lose charge and you lose time!

Actually just the opposite. The lower the battery voltage, the faster
the oscillator runs.

 

[BillW50]

In Bill in Co typed on Tue, 11 Jan 2011 14:10:44 -0700:

In MyNews typed on Mon, 10 Jan 2011 07:04:27 -0600:
[...]

New Lithium Battery 3v
they run For about $2 to $3 a piece!

for they are getting 3 to 4 your old
will lose charge and you lose time!

Actually just the opposite. The lower the battery voltage, the faster
the oscillator runs.

Interesting! Why is that? I would have thought it was a crystal
oscillator, whose frequency is fixed by the crystal.

True, but you can tweak the frequency a tad by varying the voltage you
use to make it oscillate. Most quartz watches has an adjustment to vary
the frequency a bit. Most computers lack this adjustment and depend
solely on the battery voltage. The OP mentioned some machines are off by
two hours a day. They didn't say which way it was off though. If it is
running slow 2 hours a day, well that would be very unusual. If running
fast by two hours a day, a very low real time clock battery would do it.

 

[BillW50]

In Bill in Co typed on Tue, 11 Jan 2011 22:13:57 -0700:

Wow, I'm surprised it could vary THAT much. I would just expect it
to stop oscillating (once the battery voltage falls low enough to do
anything).
I remember from my amateur radio days the difficulty in pulling a
crystal oscillator by more than a few hundredths of a percent, at
least as I recall.

Remember too, the clocks used in computers are about as cheap as you can
get. And all of the parts are all low quality including the crystal. Yes
two hours off a day is pushing it to the point that most oscillates
would stop running if the voltage dropped that low.

 

[MyNews]

BillW50

So were is a Links of Info to back you up?

 

[MyNews]

glee mmm

Are they all kept on 24/7, or are they shut down at night?
--
Glen Ventura
MS MVP Oct. 2002 - Sept. 2009
CompTIA A+
http://dts-l.net/

 

[choro]

It just happens with some machines. It doesn't have anything to do with
old CMOS batteries or anything like that, like some here are telling
you. The internal clocks are not quartz-precise, that's all. You can
make the machines better than quartz-precise by installing an NTP
(Network Time Protocol) software on the machines. That way, you can sync
them to any NTP server on the Internet. A good freeware utility is the
AboutTime. You can set it to update itself on any amount of elapsed time
you like, a good amount would be once every one or two hours. This
AboutTime software also acts as an NTP server, so you can have one
machine syncing itself to the Internet NTP servers, while the rest of
your machines sync themselves off of this machine.

Yousuf Khan

Got a lovely vintage German Kienzle battery driven off-white and subtle
yellow ceramic frame wall clock which makes sounds like a cuckoo clock
winding itself up and down but can never get that to keep accurate time
either! It is pre the era of Japanese quartz watches and clocks. I even
tried putting in a fresh battery, to no avail. In the end I gave up
trying to adjust it.

At the moment it keeps losing time, so I set it forward by 10 mins until
it falls back to being 10 mins slow. Consequently I just keep it for the
gentle tick tack and the sounds of winding itself up and down like a
cukoo clock.

But I still love it.

 

[choro]

Does it look a bit like this one?

http://www.tias.com/13486/PictPage/3923288832.html

Yes, but the ceramic surround is more art-deco if you like. Actually I
was wrong about the colors. Ivory scalloped left and right edges with a
subtle moss green as the main color.

The main color is actually more a shade somewhat between the skin and
the flesh of an avocado pear. Tried to find a image on the Internet but
without success. Unfortunately Can't send any binary stuff to NGs
otherwise I could have taken a picture of it and posted it here.
--
choro
*****

PS. But tell me in all "seriosity"...

What is it that goes in dry and comes out wet, and gives pleasure to two
people?

;-)

 

[choro]

Go here http://www.tinypic.com/ and select/upload an image

When it's 'done its stuff' select 'Direct link for layouts' (.jpg) -
hilight then copy/paste into your message .....

Like this: http://i54.tinypic.com/25ksor8.jpg

Folk may then view by clicking on the link!

I hope! ;-)

HTH

Dave

Haven't got my own website where I could post the pic and then give you
the link, I am afraid. What I meant was that I could have used my camera
to take a pic of my Kienzle wall clock and posted it to this NG.

And though I am still with the same ISP and in the old days I am sure I
could send images or attachments to NGs, I now find that these days
anything I post with an image or an attachment gets lost on the
ethernet. So much for progress!

 

[BillW50]

In MyNews typed on Thu, 13 Jan 2011 01:13:33 -0600:

BillW50

So were is a Links of Info to back you up?

Sure, just list what you want references for and I will provide them.
Are you talking Linux and viruses for example?

 

[choro]

Now you have completely lost me!

Do you see a picture when you click here?

http://i54.tinypic.com/25ksor8.jpg (it's the engine in my narrowboat)

Yes, of course I can see the picture of a very clean and obviously well
maintained engine in both Google Chrome AND in IE9.

But what I meant was that I did NOT come across an image of my Kienzle
clock on the Internet. So how can I post a link on this NG?

Now IF I had my own web page I could have used my camera to take a
picture of my Kienzle and put it up on my web page. Then I could have
posted a link to the photo on my web page on this thread.

With my ISP and can see messages with jpg and other attachments but my
ISP will NOT let me post any such stuff to NGs. Though of course I can
do this in my normal correspondence. But not in my postings to NGs.

Do I make myself clear now?
--
choro
*****

By Jove, you must have the cleanest engine anywhere, all polished up and
spanking shiny! Well done! Never seen a cleaner engine.

 

[BillW50]

In BillW50 typed on Thu, 13 Jan 2011 05:43:22 -0600:

In MyNews typed on Thu, 13 Jan 2011 01:13:33 -0600:

Sure, just list what you want references for and I will provide them.
Are you talking Linux and viruses for example?

Oh duh? I should read the subject first. Sorry, I just woke up. ;-)

You are talking about the RTC running fast if the battery is low, eh? No
I haven't seen anything on the Internet about this for years.

Where I first learned about it was from another fellow electronic
engineer on these newsgroups many years ago. And while it made sense
from the EE standpoint, I also did some experiments. And you can do the
very same as well. And when you lower the voltage, the clock does in
fact speed up. You can only go low so far and the clock then stops
altogether.

 

[choro]

Thanks ~BD~. Must remember this "tinypic" web site. May be I will take a
pic when I feel more up to it and put it up on this website. Feeling a
bit drowsy now. Been up for almost 24 hours.
--
choro
*****

On 13/01/2011 11:35, ~BD~ wrote: [....]

Now you have completely lost me!

Do you see a picture when you click here?

http://i54.tinypic.com/25ksor8.jpg (it's the engine in my narrowboat)

Yes, of course I can see the picture of a very clean and obviously well
maintained engine in both Google Chrome AND in IE9.
Thanks

But what I meant was that I did NOT come across an image of my Kienzle
clock on the Internet. So how can I post a link on this NG?

Well at least I provided a link where a similar clock *is* shown!

Now IF I had my own web page I could have used my camera to take a
picture of my Kienzle and put it up on my web page. Then I could have
posted a link to the photo on my web page on this thread.

Hmmm. If you have a digital camera, surely you can take a photograph of
your clock and then load that image into your computer - in a place
where you can find it again?

Assuming 'yes', you too can use the 'tinypic' facility to upload that
photograph to the *tinypic* web site and then copy the .jpg address into
a post just as I demonstrated earlier. It's free, byw!

With my ISP and can see messages with jpg and other attachments but my
ISP will NOT let me post any such stuff to NGs. Though of course I can
do this in my normal correspondence. But not in my postings to NGs.

I understand that you cannot *attach* a photo to a post, but you can
(I'm fairly certain) insert a link to a URL within a message.

Do I make myself clear now?

Thanks for explaining. Maybe *you* will better understand now and 'have
a go' as I've indicated? All it costs is a little time! (Hey, we're back
on topic of the thread 'Subject'!)

choro
*****

By Jove, you must have the cleanest engine anywhere, all polished up and
spanking shiny! Well done! Never seen a cleaner engine.

That's kind of you to say!

It doesn't always look /quite/ so clean - but not far off for our annual
rally with other boaters who have the same type of engine!

 

[Yousuf Khan]

Got a lovely vintage German Kienzle battery driven off-white and subtle
yellow ceramic frame wall clock which makes sounds like a cuckoo clock
winding itself up and down but can never get that to keep accurate time
either! It is pre the era of Japanese quartz watches and clocks. I even
tried putting in a fresh battery, to no avail. In the end I gave up
trying to adjust it.

At the moment it keeps losing time, so I set it forward by 10 mins until
it falls back to being 10 mins slow. Consequently I just keep it for the
gentle tick tack and the sounds of winding itself up and down like a
cukoo clock.

But I still love it.

Before they had quartz oscillators, clocks and watches were timed by a
capacitor filling and emptying. The repeating cycles of the capacitor
kept the time. But of course, capacitors are all individually unique, so
some may cycle a bit faster or slower than others.

Hell, I even had a cheap quartz digital watch that was off by huge
amounts every month. Even the quartz watches weren't that accurate at
one time. The quartz was just an advertising ploy.

Yousuf Khan

 

[Alain Dekker]

The title of the OP says it all "losing time". On one machine it was truely
terrible: We set the correct time and watched it lose about 2-3s per 10s. In
other words, if we set the clock to 10:00:00, then after one minute (one an
external, accurate, wristwatch) the wristwatch says 10:01:00, but the
computer was saying about 10:00:40.

Had a call from Service who say they've taken the same computer out of our
bulky machine and put it on the bench. The clock is now keeping up with the
wristwatch, suggesting there is something crazy happening with voltages in
the machine.

Alain

In Bill in Co typed on Tue, 11 Jan 2011 14:10:44 -0700:

In MyNews typed on Mon, 10 Jan 2011 07:04:27 -0600:
[...]
New Lithium Battery 3v
they run For about $2 to $3 a piece!

for they are getting 3 to 4 your old
will lose charge and you lose time!

Actually just the opposite. The lower the battery voltage, the faster
the oscillator runs.

Interesting! Why is that? I would have thought it was a crystal
oscillator, whose frequency is fixed by the crystal.

True, but you can tweak the frequency a tad by varying the voltage you use
to make it oscillate. Most quartz watches has an adjustment to vary the
frequency a bit. Most computers lack this adjustment and depend solely on
the battery voltage. The OP mentioned some machines are off by two hours a
day. They didn't say which way it was off though. If it is running slow 2
hours a day, well that would be very unusual. If running fast by two hours
a day, a very low real time clock battery would do it.

 

[Alain Dekker]

The computer is generally kept on at all times. The computer is also
generally not connected to the internet, allowing synching with an NTP time
server impossible.

When I saw this happening in front of my eyes, I quickly opened a command
prompt and typed in "time". As I understand it, this is the BIOS time so can
be different from that reported by Windows, right? Anyway, this showed the
same time as in the notification area, bottom right of Windows XP, so now I
don't know what to think.

Is there any hardware, even if it costs lots of money, that can be
implemented on these machines to get a more guaranteed clock performance?
These machine are used in high speed production lines and timing for
rejection is critically important!

Thanks,
Alain