[volt-nuts] Small capacitance

[Fred]

Thanks you all for the usefull info. Very interesting

I can not find the app note/book about the "shunt" methode I use. I
thought it was Agilent but I have a lot of books about network analysis
so I can be wrong.

Agilent app note: 5989-5935EN describes the way to use a shunt
measurement for very low impedances. Port 1 sends the incident to the
DUT and  and the voltage drop over it is very low, the current high and
the combined resistance of the ports convert that to a voltage. A sort
of Kelvin measurement. You use S21 and a custom trace or calculate the
result. 25(Rho/(1-Rho))

To measure high impedances you use a series measurement like Ivan
writes.

But then I have to calculate. What I did in my high impedance
measurement was placing the High Z capacitor in shunt with 50 ohm. The
parallel impedance of the XL and the 50 ohm are parallel allmost 50 ohms
and a VNA is very accurate around 50 ohm. But I do a 2 port sweep and
use the RX as 50 ohm shunt and measure S11 because my VNA needs a two
port SP2 measurement to do the 12 term error correction and calculate
the right Cp from the resulting corrected S11. On my old HP8407 I use
just a 50 ohm terminator paralleled with the capacitor and a directional
bridge so maybe shunt methode is the wrong word in this case and a
better word is maybe parallel measurement. But with the HP VNA the
problem is my homemade bridge, that has only a 30dB directivity but the
results are usefull. I will test the S21 series methode too. Did that in
the pre-tests while I was searing the4 different ways.

Today I made a coaxial capacitor but I made a mistake. I made two
plastic inner rings to hold the center conductor in the tube. After
calculation the ringparts together have a bigger capacitance as the
airpart and because I do not know if it is PE or nylon I have no clue
about the value. between 16-18pF total. So I am gonna change the
internal rings into endcaps so there is no dielectrium other then air in
the tube.Mount it on a board and make the inner conductor slidable so I
can change capacitance.

Fred


Ivan Cousins schreef op za 14-01-2012 om 10:38 [-0800]:
> Fred:
> You can measure the capacitance of a series connected capacitor with 
> high resolution.
> Have a look a manual for the Keithley 590 CV Analyzer.
> The concept is explained in the manual. See the diagram(s) and text.
> I regularly measure Schottky diode junction capacitance with sub fFarad 
> resolution.
> The analogy is to use a VNA to measure a capacitor in the series mode 
> (S21) rather than in shunt mode (S11). By measuring a DUT in a series 
> measurement mode the full dynamic range of the receiver can be used. For 
> measuring impedances that are less than 50 ohms, measure the DUT in 
> shunt mode. For measuring impedances that are greater than 50 ohms, 
> measure the DUT in series mode.
> This concept is also explained, (by looking at the equations and 
> diagram(s)) in the HP4275A LCR Meter.
> Some concepts can be used in many measurement instruments, i.e. a CV 
> Analyzer, LCR Meters, and VNAs.
> 
> Some of the early NBS documents showed how to build a ring guarded, 
> circular, parallel plate capacitor.
> I do not have a reference handy at the moment.
> A Google Scholar search with the right set of key words should turn up 
> something.
> 
> PS The Keithley 590 CV Analyzer is set to measure at 100KHz or 1MHz.
> I have used the Keithley 590 , HP4275A and VNAs for the last  decade (or 
> more).
> Some concepts work in all three instruments.
> 
> Ivan Cousins
> 
> _______________________________________________
> volt-nuts mailing list -- volt-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there.