time-nuts@lists.febo.com

Discussion of precise time and frequency measurement

View all threads

Re: [time-nuts] R&S XSRM Rubidium Standard

K
KA2WEU@aol.com
Sun, Sep 17, 2017 6:45 PM

correct !

In a message dated 9/17/2017 1:51:30 P.M. Eastern Daylight Time,
cjaysharp@gmail.com writes:

Yes.

An instrument with a calibration certificate is not  necessarily accurate
but it's inaccuracies are known and can be compensated  for (but only to the
accuracy of the calibration reference of  course.)

On 17 Sep 2017 17:39, "Magnus Danielson"  magnus@rubidium.dyndns.org
wrote:

Hi,

The  word "calibration" is overloaded with multiple meanings, and
incompatible too.

"calibration" is often used to describe  adjustments to make a device
operate correctly, such as passing the  performance checks.

"calibration" in legal traceability is  about measure the performance
against references to form a traceable  record of deviations from the

norml.

This may include adjustment to  ease compensation, but this is not
necessary. Regardless of wither  adjustments where done or not, the
calibration record will indicate  the errors that then needs to be applied
to the measurement for the  measurement to be traceable, and this in

itself

requires documented  knowledge about how to do the measurement.
Otherwise it's just a fancy  indication.

Adjustment to a reference thus do not imply legal  traceability, or even
full functionality.

For full  functionality, you have to go through the performance check and
see  that all values is within limits.

"calibration" can thus imply  different things.

I regularly see people use these terms  inconsistently. That people get
disappointed when they get the wrong  thing is to be expected.

Cheers,
Magnus

On 09/17/2017 05:23 PM, Scott McGrath  wrote:

As to the point most modern instruments have self  calibration,  Most of
the time 'calibration' is simply  the performance check adjustments are

not

performed unless  necessary

The difference being the instruments used in  performance test are
traceable to a national standards  body.

So whats referred to as calibration is in  reality performance

validation.

How do I know this by  becoming friendly with the local lab and years ago
when i worked  for govt i used to moonlight at one of the local cal

labs.

On Sep 17, 2017, at 8:57 AM, KA2WEU--- via  time-nuts

wrote:

Modern test and radio equipment have  self calibration capabilities,

older

analog do not.  Calibration is not always need for  just simple test,

but

for specification conformation it is useful. A bit  of luck also

helps.

In a message dated 9/17/2017 8:08:00 A.M. Eastern Daylight  Time,
drkirkby@kirkbymicrowave.co.uk  writes:

On 15  Sep 2017 10:45, "Scott  McGrath" scmcgrath@gmail.com  wrote:

Precisely my point,    But when purchasing i  expect to pay for  a

calibration at a  minimum.

I have on occasions  requested  sellers to send an item to the
manufacturer
(Agilent or Keysight)  for calibration before shipping it to  me,
offering
to pay the calibration  cost, but stating that I expect a full refund

if

the
item fails the  calibration.

If a test equipment dealer is  confident that something is  working

well,

they should  not object to sending it to the manufacturer for
calibration,
as long as the buyer is willing to  pay.

Of course if a  seller knows little  about something,  they are not

going

to
do  this,  but the item should be appropriately  priced.

One UK seller  (grace1403)  declined to send an Agilent N9912A FieldFox

to

Agilent,  because  "Agilent were too fussy"., failing items for  trivual
issues.    But he did agree to send it to  one of the cal labs he uses.

I

thought it  was a waste of  time going to one of the less fussy outfits,
but
bought it anyway. It was then clear on receipt that it was  faulty.

(The

spectrum analyser functionality was ok, but  it didn't work as a

network

analyzer).  He took it  back,  but then advertised it on  eBay 6 months
later. When asked, he said nothing had been done to  it.

eBay rules about who pays the return  shipping charge for an item  that

is

"not as described'  keep changing, and may be different on different
sites.
But on a heavy item shipped internationally,  the  postage cost  can be
comparable or exceed the  calibration  cost.

Dave.


time-nuts  mailing list -- time-nuts@febo.com
To  unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and  follow the  instructions there.


time-nuts  mailing list -- time-nuts@febo.com
To unsubscribe, go to  https://www.febo.com/cgi-bin/m
ailman/listinfo/time-nuts
and follow the instructions  there.


time-nuts mailing  list -- time-nuts@febo.com
To unsubscribe, go to  https://www.febo.com/cgi-bin/m
ailman/listinfo/time-nuts
and follow the instructions  there.


time-nuts mailing list  -- time-nuts@febo.com
To unsubscribe, go to  https://www.febo.com/cgi-bin/m
ailman/listinfo/time-nuts
and  follow the instructions  there.


time-nuts  mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the  instructions there.

correct ! In a message dated 9/17/2017 1:51:30 P.M. Eastern Daylight Time, cjaysharp@gmail.com writes: Yes. An instrument with a calibration certificate is not necessarily accurate but it's inaccuracies are known and can be compensated for (but only to the accuracy of the calibration reference of course.) On 17 Sep 2017 17:39, "Magnus Danielson" <magnus@rubidium.dyndns.org> wrote: > Hi, > > The word "calibration" is overloaded with multiple meanings, and > incompatible too. > > "calibration" is often used to describe adjustments to make a device > operate correctly, such as passing the performance checks. > > "calibration" in legal traceability is about measure the performance > against references to form a traceable record of deviations from the norml. > This may include adjustment to ease compensation, but this is not > necessary. Regardless of wither adjustments where done or not, the > calibration record will indicate the errors that then needs to be applied > to the measurement for the measurement to be traceable, and this in itself > requires documented knowledge about how to do the measurement. > Otherwise it's just a fancy indication. > > Adjustment to a reference thus do not imply legal traceability, or even > full functionality. > > For full functionality, you have to go through the performance check and > see that all values is within limits. > > "calibration" can thus imply different things. > > I regularly see people use these terms inconsistently. That people get > disappointed when they get the wrong thing is to be expected. > > Cheers, > Magnus > > On 09/17/2017 05:23 PM, Scott McGrath wrote: > >> As to the point most modern instruments have self calibration, Most of >> the time 'calibration' is simply the performance check adjustments are not >> performed unless necessary >> >> The difference being the instruments used in performance test are >> traceable to a national standards body. >> >> So whats referred to as calibration is in reality performance validation. >> >> How do I know this by becoming friendly with the local lab and years ago >> when i worked for govt i used to moonlight at one of the local cal labs. >> >> On Sep 17, 2017, at 8:57 AM, KA2WEU--- via time-nuts <time-nuts@febo.com> >>> wrote: >>> >>> Modern test and radio equipment have self calibration capabilities, older >>> analog do not. Calibration is not always need for just simple test, but >>> for specification conformation it is useful. A bit of luck also helps. >>> >>> >>> >>> >>> In a message dated 9/17/2017 8:08:00 A.M. Eastern Daylight Time, >>> drkirkby@kirkbymicrowave.co.uk writes: >>> >>> On 15 Sep 2017 10:45, "Scott McGrath" <scmcgrath@gmail.com> wrote: >>>> >>>> Precisely my point, But when purchasing i expect to pay for a >>>> >>> calibration at a minimum. >>> >>> I have on occasions requested sellers to send an item to the >>> manufacturer >>> (Agilent or Keysight) for calibration *before* shipping it to me, >>> offering >>> to pay the calibration cost, but stating that I expect a full refund if >>> the >>> item fails the calibration. >>> >>> If a test equipment dealer is confident that something is working well, >>> they should not object to sending it to the manufacturer for >>> calibration, >>> as long as the buyer is willing to pay. >>> >>> Of course if a seller knows little about something, they are not going >>> to >>> do this, but the item should be appropriately priced. >>> >>> One UK seller (grace1403) declined to send an Agilent N9912A FieldFox to >>> Agilent, because "Agilent were too fussy"., failing items for trivual >>> issues. But he did agree to send it to one of the cal labs he uses. I >>> thought it was a waste of time going to one of the less fussy outfits, >>> but >>> bought it anyway. It was then clear on receipt that it was faulty. (The >>> spectrum analyser functionality was ok, but it didn't work as a network >>> analyzer). He took it back, but then advertised it on eBay 6 months >>> later. When asked, he said nothing had been done to it. >>> >>> eBay rules about who pays the return shipping charge for an item that is >>> "not as described' keep changing, and may be different on different >>> sites. >>> But on a heavy item shipped internationally, the postage cost can be >>> comparable or exceed the calibration cost. >>> >>> Dave. >>> _______________________________________________ >>> time-nuts mailing list -- time-nuts@febo.com >>> To unsubscribe, go to >>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>> and follow the instructions there. >>> >>> _______________________________________________ >>> time-nuts mailing list -- time-nuts@febo.com >>> To unsubscribe, go to https://www.febo.com/cgi-bin/m >>> ailman/listinfo/time-nuts >>> and follow the instructions there. >>> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/m >> ailman/listinfo/time-nuts >> and follow the instructions there. >> >> _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/m > ailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.