[volt-nuts] Agilent calibration - certificate for a vector network analyzer (VNA)

[Dr. David Kirkby]

On 24 August 2013 18:55, Joseph Gray <jgray at zianet.com> wrote:
> David,

Hi Joseph

> That is quite a difference between the two certificates. The Techmaster one
> doesn't tell you anything about what they actualy measured, or the
> uncertainties. You have to trust that they know what they are doing and
> that they actually did it.

I think Techmaster certificates are useful to T+M dealers, as they can
sell things for a higher price. Their certaificates are also useful to
end users that just need a certificate to satisfy some institute or
body. But I don't believe it would satisfy a competent engineer who
really wanted to know if the instrument was working properly. A
competent enginner would not be too conviced.

When an Agilent VNA calibration expert (Ken Wong) wrote on the Agilent
VNA forum that "I will be very skeptical when a VNA calibration
service does not include a cal kit and verification kit." and that
Techmaster cal certificate did not use a verification kit, one has to
be sceptical.

The fact Techmaster also issued a cal certificate for a calibration
kit makes me suspicious too. Apparently you need access to primary
standards to verify they are in spec.

There is something else interesting on that Techmaster certificate.
Note it says "option 010 added". That is a software option (enables a
Inverse Fourier transform to convert frequency domain data to time
domain). It is no longer sold by Agilent. I do wonder if that was
legally added. Buying that option for a new VNA costs several thousand
dollars, but the 8720D was last made in around 1999, and has been
unsupported since 2004. I suspect these cal labs are useful for things
like getting software options added! Agilent will certainly not sell
the option.

> With the Agilent certificate, there is data to
> give you confidence that things were done properly.

Yes. Note the Agilent calibration is not acreditated by anyone. I
could have had an acredited calibration, but I did not want to pay the
extra 50% or so. From what I understand, the VNA would be calibrated
in exactly the same way. But it costs Agilent money to get acredited
for the calibration, so they pass that cost onto those that want it
acredited.

For me personally, I trust Agilent know how to calibrate VNAs. There
is going to be more VNA expertese in Agilent than arguably any other
institution.

> Thanks for posting those. It was educational.

I'm glad you found it useful.

> Joe Gray
> W5JG

Dave, G8WRB