Using HF2IS Impedance Spectroscope as a DC Impedance Meter

Up until now, the blogs related to impedance measurement are mostly related to how to setup a measurement and optimize the ac frequency sweeps. As a matter of fact, the HF2IS Impedance Spectroscope can also be used as a dc impedance meter. The set up is not complicated but does require the users to take some precautions. Here is a description of how to properly obtain dc impedance with your HF2IS Impedance Spectroscope.

DC Impedance Measurement Setup

Hardware Setup

In this example, a dc impedance measurement with the 4-Term measurement setup is illustrated. The same technique can certainly be extended to the 2-Term measurement as well. Here, instead of driving the DUT with the high frequency Signal Output port, we use one of the four auxiliary outputs to generate desired dc voltages. One can sweep this dc voltage in the range of ±10 V, if required.

4T_DC_setup

Graphical User Interfac

To make the dc voltage and current measurement in ziControl, choose 4-Term Z setup, same as any other regular impedance measurement. Be sure to set the frequency to 0 Hz which reflects doing the demodulation at dc.

ziControl_GUI

  Then, to generate the dc voltage, one setup the Auxiliary I/O tool tab as below. The auxiliary output is set to manual mode with the offset dc voltage defined.

ziControl_GUI_aux

 

Source of Errors

The following sections described errors that should be calibrated out when measuring dc impedance with the HF2IS Impedance Spectroscope.

Intrinsic Measurement DC Offset

If we do a dc voltage sweep from 0 V to 1 V as shown below, we will see that at 0 V, the measured current (Signal Input 1) and voltage (Signal Input 2) are not zero (i.e. contain an offset). It must be noted that both the HF2IS Impedance Spectroscope and the HF2TA Transimpedance Amplifier input contain protection elements which can cause dc leakage. One of course does not notice this leakage-related offset when measuring at ac. The good news is that the dc offset is more or less constant which can be seen by the graphs below exhibiting a linear ramp behaviour (i.e. the measured current and voltage increase linearly with the dc bias.) This means that we can simply subract the offset at 0 V to get the true measured dc current and voltage value.

measured_current

 Measured current at Signal Input 1  vs dc sweep

measured_voltage

Measured differential voltage at Signal Input 2 vs dc sweep

Measurement Error by a Factor of 2

This error is less obvious to the users, especially in the 4-Term measurement since both the current and the voltage measurement values are both off by the same factor of two. So the calculated impedance will still be correct when the factor of 2 is cancelled out through the division of voltage by current. However, this error will show up when doing the 2-Term measurement when only the current is measured.

The reason for this factor of 2 can be explained with the example below. The screen capture shows a measurement of differential voltage (in FRA mode) of 1 V dc. Even though dc voltage does not have an rms value, ziControl still (unfortunately) shows the measured value as rms. One would then expect the value to be around 0.707 V when there is no input offset. However, one can see that the ziControl is displaying about twice the value, thus an error factor of 2. This factor of 2 arises from the fact that the demodulation always generates two sidebands: one at dc and the other at twice the modulation frequency. Obviously, if the frequency is dc, then both sidebands will also add up at dc which results in twice the dc amplitude measured. In short, when measuring at 0 Hz, one needs to also divide the measured value by 2 in order to get the right dc value. For more details, please refer to this Zurich Instruments blog.

voltage_error

Influence of HF2TA Transimpedance Gain

When measuring at dc, one might have a tendency to want to give a maximum gain since it also keeps the HF2TA Transimpedance Amplifier input bandwidth low. However, the input offset will also get amplified which can saturate the HF2TA very quickly even at a small dc voltage. Therefore, one should choose the transimpedance gain carefully to achieve good SNR without prematurely saturating the Signal Input port. In addition, due to the HF2TA Transimpedance Amplifier design, one should be aware of that both the different transimpedance gain settings as well as the measured device can have influences on the offset. Therefore, the calibration must be performed for each gain setting and each device separately. In general, keeping the transimpedance gain low is preferred.

DC Impedance Meausrement Steps

ziControl does not have features to calibrate dc impedance measurement. However, the existing ziControl tool sets can still be used to extract the dc impedance with some simple  manual post-processing. One can of course always automate these steps in Labview, Matlab or Python using the LabOne API, if needed. Here is a recommended procedure.

DC sweep of auxiliary output

The calibrated dc current can be calculated by:

IDUT_peak = 0.7071 * [p*I (@ bias voltage) – p * I (@ 0 V) ]

VDUT_peak = 0.7071 * [ V (@ bias voltage) – V (@ 0 V) ]

p  = +1 for positive wrapped phase, -1 for negative wrapped phase

The checking for the polarity is just to ensure that the calculation is correction when there is a negative current/voltage measured.

This step will allow one to verify two things:

  • whether the dc sweep is linear (and it should be for a reliable measurement independent of the dc bias applied)
  • the polarity of the dc offset

The example below shows that the 0 v offset is 11.5 μA. In addition, there is an inflection of dc current at about 800 mV dc bias. This means that the initial HF2TA Transimpedance Amplifier has a negative offset current at 0 V. As the bias is increased, the negative offset current slowly gets cancelled out until the actual device current starts to dominate at and after 800 mV. The formula will take care of this kind of inflection. Using the formula, the measured dc current at 1 V bias is 10.1 μArms in this example. The measured differential voltage (not shown here) is 0.995 V.

measured_current_example_R

Note: the poarlity of the parameter p should be considered with wrapped phase

One can of course measure the dc impedance without the sweep. The sweep simply helps to observe the dc offset behaviour.

Calculate DC Impedance

The impedance is then simply VDUT_peak/IDUT_peak. In this example, the measured impedance is 98.5 kΩ. This matches very well with the 98.7 kΩ measured by a multi-meter. Here are some sample comparisons between the HF2IS and the multi-meter. It can be seen that the larger the impedance, the small the measured difference is.

comparison_table

 

Conclusion

This blog describes how to use HF2IS Impedance Spectroscope as a dc impedance meter. The most important points to take care of is the normalization of measurement at dc and the dc offset. Otherwise, the dc impedance measurement is quite accurate compared to a typical multi-meter.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>