hi all,

i have made plenty of study of how to determine four C/L in the polynomial for the OPEN/SHORT cal kit however, i couldn't really find a method how to use the network analyzer to determine them, or verify the value.

First, i have browsed through a couple of threads here talking about this topic but i couldn't get the answer, as well as agilent application note and i even got the handbook of microwave component measurements which partially helping me.

Assuming i have the agilent cal kit 85032F and i would like to verify the C value, for my open cal kit, here would be my step

1) calibrate the PNA/ENA ( assuming they comes in N type connector )

2) Connect the open connector directly to the PNA port, and choose S11, or S22.

3) turn to smith chart, and from here how would i determine the 4 C value?

i am looking at the handbook page 560, which they shows a couple of frequency response open/short, in tme domain / freq domain and time gating function. May i know what's the correct procedure to determine it? do i need any mathematical software to help me?

thanks for your assistance.

i have made plenty of study of how to determine four C/L in the polynomial for the OPEN/SHORT cal kit however, i couldn't really find a method how to use the network analyzer to determine them, or verify the value.

First, i have browsed through a couple of threads here talking about this topic but i couldn't get the answer, as well as agilent application note and i even got the handbook of microwave component measurements which partially helping me.

Assuming i have the agilent cal kit 85032F and i would like to verify the C value, for my open cal kit, here would be my step

1) calibrate the PNA/ENA ( assuming they comes in N type connector )

2) Connect the open connector directly to the PNA port, and choose S11, or S22.

3) turn to smith chart, and from here how would i determine the 4 C value?

i am looking at the handbook page 560, which they shows a couple of frequency response open/short, in tme domain / freq domain and time gating function. May i know what's the correct procedure to determine it? do i need any mathematical software to help me?

thanks for your assistance.

In your case, you N-sma adapter may not have enough length to justify using time domain, and mismatches (likely due to the beads or teflon insert holding up the center conducor) might be distributed along the connector and so are more difficult to gate out.

The trouble is the open circuit capacitance is -nearly- flat with frequency, and for low frequency (below 6 GHz) you can do just about as well using a delay offset as capacitance. Normally it runs something like 70 fF+-5 fF. The real difficulty is in getting the delay to the open or short exactly correct. And mismatch in your adapter will yeild a ripple in the measurement that you might be tempted to infer is the variation of the capactitance, but is, in practice, error due to adapter mismatch that you want to remove from the open measurment.

In practice, the best thing might be to get an estimate of the C open, and short, and then do a calibration and measure a high quality airline with a short on the end. Ripple in that measurement is due to delay and C0 error between the open and short. Adjust the values and see if the ripple is reduced, if not, adjust the other way. When you have minimized the ripple due to delay, then start adjust C open.

BUT, for the connectorized case, you would be much better doing a TRL cal (which doesn't depend on the value of the reflection) and then measure the open directly. But for this, of course, you need a good TRL line set.