Re: [time-nuts] Question for the cesium nuts.
In a message dated 3/16/2007 18:32:58 Pacific Daylight Time,
martelno@yahoo.com writes:
Also, some high end rubidium (such as Perkin-Elmer)
manufacturers are able to develope 133Rb clocks having
450 000 hours MTBF! That'a a lot of nanoseconds!
Hi Jack,
we cannot expect the units to work that long. Hard Disks have >50K MTBF, and
fail all the time in much less time depending on how they are used etc.
MTBF is a mean, meaning in the real world the Perkin Elmer unit could fail
in the first 10 minutes, just that the statistics for that to happen have
pretty low probability.
When they calculate MTBF, they probably don't take into account external
effects such as AC voltage spikes due to janitors pluggin-in vacuum's or due to
lightning etc, earthquakes, user obuse, water damage, movement, effects of UV
light on plastik parts such as cables, Tin Whisker shorts, electrolytic
capacitor dryout, mouse damage, etc. I think it's pretty useless to say a piece
of equipment has a MTBF of over 50 years. Then again that's only an opinion...
bye,
Said
************************************** AOL now offers free email to everyone.
Find out more about what's free from AOL at http://www.aol.com.
Having worked on the HP 10816A rubidium clock 25 years ago,
it is hard for my to believe the lamp alone could last 56
years on the average. Even ordinary light bulbs don't last
that long.
Rick Karlquist N6RK
SAIDJACK@aol.com wrote:
In a message dated 3/16/2007 18:32:58 Pacific Daylight Time,
martelno@yahoo.com writes:
Also, some high end rubidium (such as Perkin-Elmer)
manufacturers are able to develope 133Rb clocks having
450 000 hours MTBF! That'a a lot of nanoseconds!
Hi Jack,
we cannot expect the units to work that long. Hard Disks have >50K MTBF,
and
fail all the time in much less time depending on how they are used etc.
MTBF is a mean, meaning in the real world the Perkin Elmer unit could
fail
in the first 10 minutes, just that the statistics for that to happen have
pretty low probability.
When they calculate MTBF, they probably don't take into account external
effects such as AC voltage spikes due to janitors pluggin-in vacuum's or
due to
lightning etc, earthquakes, user obuse, water damage, movement, effects of
UV
light on plastik parts such as cables, Tin Whisker shorts, electrolytic
capacitor dryout, mouse damage, etc. I think it's pretty useless to say a
piece
of equipment has a MTBF of over 50 years. Then again that's only an
opinion...
bye,
Said
************************************** AOL now offers free email to
everyone.
Find out more about what's free from AOL at http://www.aol.com.
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Having worked on the HP 10816A rubidium clock 25 years ago,
it is hard for my to believe the lamp alone could last 56
years on the average. Even ordinary light bulbs don't last
that long.
Rick Karlquist N6RK
Reminds me of the extra-ordinary light bulb:
http://www.centennialbulb.org/
/tvb
I believe the deal on that bulb is that it is being run at
considerably less than its normal voltage, and the life of
a bulb varies as voltage to the Nth power, where N is an
integer around -12 or something. Like those "lifetime"
130V bulbs.
Unfortunately, you can't throttle down a Rb plasma lamp.
Rick Karlquist N6RK
Tom Van Baak wrote:
Having worked on the HP 10816A rubidium clock 25 years ago,
it is hard for my to believe the lamp alone could last 56
years on the average. Even ordinary light bulbs don't last
that long.
Rick Karlquist N6RK
Reminds me of the extra-ordinary light bulb:
http://www.centennialbulb.org/
/tvb
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Am I the only one who have never heard of the 10816A before ?
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@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.
From: "Poul-Henning Kamp" phk@phk.freebsd.dk
Subject: Re: [time-nuts] Question for the cesium nuts.
Date: Sat, 17 Mar 2007 17:09:37 +0000
Message-ID: 11112.1174151377@critter.freebsd.dk
Am I the only one who have never heard of the 10816A before ?
A google for "HP 10816A" did not exactly turn up much. Except that it does
exist and works.
Cheers,
Magnus
That project made it as far as a pilot run of a half dozen
working prototypes and assignment of a model number, but HP never
sold any of them. I still have one of these "valuable"
collector's items :-)
You probably never heard of the HP 5063
cesium either, but that's another story...
Rick Karlquist N6RK
Poul-Henning Kamp wrote:
Am I the only one who have never heard of the 10816A before ?
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@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.
From: "Rick Karlquist" richard@karlquist.com
Subject: Re: [time-nuts] Question for the cesium nuts.
Date: Sat, 17 Mar 2007 13:48:11 -0700 (PDT)
Message-ID: 24661.204.213.79.145.1174164491.squirrel@webmail.sonic.net
Rick,
That project made it as far as a pilot run of a half dozen
working prototypes and assignment of a model number, but HP never
sold any of them. I still have one of these "valuable"
collector's items :-)
You probably never heard of the HP 5063
cesium either, but that's another story...
Isn't that the Navy cesium? To handle the physical shocks of the big 12 inch
canons if I recall correctly.
Cheers,
Magnus
You're thinking of the 5062 that fits into the hatch
of a submarine (a 5061 won't). The test is called
the "hammer blow" test.
Rick Karlquist N6RK
Magnus Danielson wrote:
From: "Rick Karlquist" richard@karlquist.com
Subject: Re: [time-nuts] Question for the cesium nuts.
Date: Sat, 17 Mar 2007 13:48:11 -0700 (PDT)
Message-ID: 24661.204.213.79.145.1174164491.squirrel@webmail.sonic.net
Rick,
That project made it as far as a pilot run of a half dozen
working prototypes and assignment of a model number, but HP never
sold any of them. I still have one of these "valuable"
collector's items :-)
You probably never heard of the HP 5063
cesium either, but that's another story...
Isn't that the Navy cesium? To handle the physical shocks of the big 12
inch
canons if I recall correctly.
Cheers,
Magnus
From: "Rick Karlquist" richard@karlquist.com
Subject: Re: [time-nuts] Question for the cesium nuts.
Date: Sat, 17 Mar 2007 15:29:38 -0700 (PDT)
Message-ID: 15616.204.213.79.145.1174170578.squirrel@webmail.sonic.net
You're thinking of the 5062 that fits into the hatch of a submarine (a 5061
won't).
Ah, thanks for recharging my neurons on that one.
The test is called the "hammer blow" test.
Why do I picture a large steel hammer on a swing setup with a DUT as targeted
endpoint?
OK, so what was the 5063 then? Now that you got us hooked!
Cheers,
Magnus
The test is called the "hammer blow" test.
Why do I picture a large steel hammer on a swing setup with a DUT as
targeted
endpoint?
Because that's pretty much how it's done :-)
OK, the DUT is sitting on a big steel table and the hammer hits the
table not the piece, but still...you would be amazed at how far the
pieces fly sometimes.
Tom Frank, KA2CDK
That reminds me of a tester I got to see while I was working on a project for
Motorola at one of their ALT (accelerated life test) labs. This was for testing
durability of cell phones.
The tester basically was a pendulum that was about three feet high. At the base,
you placed the PUT (phone under test), and you pulled the pendulum up to a
specified angle and let it go. It would swing down and whack the phone across
the room into a target. Amazing what they'll handle!
Another similar test was dropping a steel ball (maybe 3/4" diameter) onto the
display of the phone. The idea was to see how high you could go before it
shattered.
One of my more "amusing" projects.
Daun
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On Behalf
Of Thomas A. Frank
Sent: Saturday, March 17, 2007 9:41 PM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Question for the cesium nuts.
The test is called the "hammer blow" test.
Why do I picture a large steel hammer on a swing setup with a DUT as
targeted
endpoint?
Because that's pretty much how it's done :-)
OK, the DUT is sitting on a big steel table and the hammer hits the
table not the piece, but still...you would be amazed at how far the
pieces fly sometimes.
Tom Frank, KA2CDK
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Thomas A. Frank wrote:
The test is called the "hammer blow" test.
Why do I picture a large steel hammer on a swing setup with a DUT as
targeted
endpoint?
Because that's pretty much how it's done :-)
OK, the DUT is sitting on a big steel table and the hammer hits the
table not the piece, but still...you would be amazed at how far the
pieces fly sometimes.
Tom Frank, KA2CDK
It is described in MIL-STD-901. Depending on the mass of the equipment
you are testing, there is a light hammer and a heavy hammer.
I have had some of my equipment tested to that level. The peak
acceleration is in the 150-200G and the waveshape is very sharp, with
lots of harmonics, to make sure ALL resonances are excited.
Didier KO4BB
To pass the hammer blow test, the unit has to still function
electrically after the blow. It is OK if it suffers structural
damage as long as it still works.
Rick Karlquist N6RK
OK, the DUT is sitting on a big steel table and the hammer hits the
table not the piece, but still...you would be amazed at how far the
pieces fly sometimes.
Tom Frank, KA2CDK