Can someone tell me how is a cal kit fixed load verified bellow 3 GHz for it’s 52dB RL , that is other than at DC and without using just another cal kit/ network analyzer system?

Can someone tell me how is a cal kit fixed load verified bellow 3 GHz for it’s 52dB RL , that is other than at DC and without using just another cal kit/ network analyzer system?

- Thank you for a very quick if only partial response , I get around this problem by measuring them at DC and having a set of three loads that I have arbitrarily said must agree with each other to >55dB rl up to the frequency of interest this however has been flagged by an auditor for being questionable and as such I am looking for a more “absolute” process
- Looks like my intuition was right; from our in-house metrologists:

Drew: "We do use long lines to get down to 200 MHz and measure DC and then interpolate to the DC value.

We don't use just another load to c al with.

We use loads (ETs) specially selected for DC & low frequency performance.

This has been published."

and:

James: "Indeed we use TRL down to ~200MHz and the DC measured value. Between DC to 200MHz we interopolate, I left a reference to the paper use on your seat. Come see me if you have additional questions."

And the reference is (from our good friends at NPL):

Ridler, N.M.; Salter, M.J.; Young, P.R.; , "An interpolation scheme for precision reflection coefficient measurements at intermediate frequencies. II. Practical implementation ," Instrumentation and Measurement Technology Conference, 2001. IMTC 2001. Proceedings of the 18th IEEE , vol.3, no., pp.1731-1735 vol.3, 2001

doi: 10.1109/IMTC.2001.929497

Abstract: The practical implementation of an interpolation scheme for deriving the voltage reflection coefficients of short-circuit, open-circuit and matched load coaxial reflectometer calibration standards in the intermediate frequency (IF) range is described (where, broadly speaking, IF refers to the frequency range DC to 100 MHz). The scheme allows the use of commercially available devices (rather than specially constructed calculable devices) as reflectometer calibration standards with VRC values traceable to national measurement standards. Measurement results are presented which illustrate the capabilities of the scheme

URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=929497&isnumber=20096

And now you have a complete answer.

Obviously the DC value is easy to verify. The value down to some low frequecy (easily below 1 GHz) can be validated with longer line TRL kits.

I would guess from the DC value and the mid-frequency data from TRL, that a model for the load could be developed, presuming the load response is dominated by a RLC model that can be matched to the high frequency data. So, if you measure a load and you see that it has 10 pH of inductance at 1 GHz, and 50.2 ohms at DC, you can infer the response in-between, but the presumes some knowledge of the load contruction.

Long-lines is probably also quite reasonble to below 1 GHz, with a TRL line of about 16 mm giving enough phase shift. A standard 10cm airline should get you down to below 200 MHz.