开云体育

On 13 January 2013 17:20, gianfrancocanale wrote:

Hi
since some time I have a HP8753B and I use it without problems.
I did not notice before that the numeric data (less significant figures ) of the markers
visualized in the Smith chart are unstable.
Is there anybody having a similar problem ?
Even better having also a solution ?
Thanks for help
Gianfranco

When you say the markers are unstable, if you set a marker to a
frequency (say 1 GHz), does it stay at 1 GHz? If not, then I'm
surprised.

But the fact the values of real and imaginary data are unstable in the
least significant figures may not be unusual. There is always going to
be noise on any measurement. The more averaging which is done, or the
lower the IF bandwidth, the less there will be.

What makes you think the issue is any worst than what is expected?

One would have to look at the specification of the analyzer and
determine if the amount of noise is outside limits or not.

Dave

On 14 January 2013 07:27, gianfrancocanale wrote:

Hi
I can give you more details:
full frequency scan 0.3-3GHz
1601 points
smith chart visualization
marker at 1GHz
50 ohm load on port 1
Marker indications in 2 minutes of observation:
- 49.604-49.586 ohm tot
- reactive part minus 1.2559-1.2383 ohm
- 126.14-128.53 pF
Thanks
Gianfranco

The critical parameters you did not say were the IF bandwidth, and
whether averating is turned on, and if so what averaging factor is
used.

1) Switching averaging on, with an averaging factor of >1 will reduce noise
2) Increasing the averaging factor will decrease noise
3) Reducing the IF bandwidth will reduce noise.
4) Reducing either the IF bandwidth, or increasing the averaging, will
both increase the mesurement time.

My gut feeling is there does not appear to be anything wrong there.
You can't expect perfectly stable readings. I think the only way you
will know for sure is to get Agilent to calibrate it, but a detailed
look at the specifications might be useful.

Dave