AnsweredAssumed Answered

Can one use a 16453A dielectric material fixture with a 4285A LCR meter?

Question asked by drkirkby on Mar 29, 2015
Latest reply on Mar 30, 2015 by yeechuang
Whilst the 16453A dielectric test fixture is not supported with the 4285A LCR meter, is there any way the 16453A dielectric material test fixture could be used with the 4285A LCR meter? I realize that there is no direct support in the 4285A's firmware, but can anyone give me any advice on how it might be possible to calibrate them, so I can use some external software to determine the permittivity of a dielectric? 

I have the following relevant parts - see picture of some 

* 16453A Dielectric Material Test Fixture (1 MHz to 1 GHz) 
* 4285A LCR meter. (75 kHz to 30 MHz) 
* 16085B 4TP to 7mm/APC7 adapter. 
* 16048A which are 1 m BNC cables that extend the 4285A's test ports to 4 male BNC connectors. 
* 4 x female-female BNC adapters - needed because the 16048A has male connectors, as does the 16085B terminal adapter. 
* 85050B APC7 calibration kit. 

(I also have an 8753ES VNA, if that would be any help in a calibration). 

With that combination of parts I am able to: connect the 4285A LCR meter to the 16453A dielectric material test fixture. 

I achieved *reasonable*, but *NOT* good results using the following procedure, although I can see fundamental flaws why this will never be very accurate.

1) Set the cable length correction to 1 m in the LCR meter.

2) Separated the electrodes of the 16453A dielectric material test fixture,  using the latch, then performed an open calibration of the 4285A LCR meter - see photo. 

3) Shorted the electrodes of the 16453A dielectric material test fixture and performed a short calibration of the LCR meter - see photo. 

4) Inserted the material under test (MUT) into the dielectric test fixture - the photo shows a piece of PTFE which is 1.05 mm thick. 

5) Measured the capacitance on the LCR meter (0.717 pF) 

6) Computed the dielectric constant from the capacitance by assuming an electrode diameter of 7 mm, which is what the diameter of the smaller electrode of 16453A. 

Using a piece of 1.05 mm thick PTFE I calculated Er to be 2.21, which is not a totally unbelievable result. 

A very quick calculation, assuming a 3.5 mm gap and a 7 mm diameter electrode, would suggest that approximately 0.1 pF of capacitance would be present between the two electrodes when the "open" calibration is performed. So really the open is not really a very good open. In a VNA, one defined the capacitance of an open, but I don't see any way to do this in the 4285A. 

I suspect the accuracy would improve for thinner samples and higher permittivity materials, as then the non-ideal characteristics of the fixture would be less significant, as the E-field would be more concentrated in the dielectric. 

I note the procedure one is *supposed* to use with the 16453A involves using about 0.7 mm of PTFE as a load calibration standard, and entering the thickness into one of the supported impedance analyzers, but that is not much use to me, as I don't have any of those analyzers. 

I have considered the possibility of trying to model the fixture with a finite element program to compute the capacitance, although the little "fingers" which  I think are used to make a sliding connection to the upper electrode probably make it practically impossible to do a good job. 

I can think of a way which may allow me to get an accurate measurement of the fringing capacitance of the "open", but I've still not no idea how I could use it to improve the calibration of the fixture with the LCR meter. 

* Connect a cable with an APC7 connector to the 8753ES VNA
* Calibrate the VNA with the 85050B APC7 calibration kit
* Make a low-inductance short between the bottom electrode of the 16453A and use a port extension on the VNA to get a phase of 180 degrees
* Remove the low inductance short, and the VNA should show me the fringing capacitance, but I don't think it would be too accurate, as VNAs don't work too well at very high impedances. 

Does any of this make sense? I know the simple answer is to buy an impedance analyzer supported by the 16453A, but these are all quite expensive, whereas I have both the 4285A and the 8753ES. 


Edited by: drkirkby on Mar 29, 2015 9:41 PM