开云体育

Hi all,
It seems this audience might be interested in an impedance analyzer that I designed/built that works from DC-100kHz with 0.1% accuracy over much of the impedance range. The motivation for this was needing to get precise impedance measurements at arbitrary frequencies without spending a lot. I was also interested in learning more about impedance measurements and circuit design.?

Some key details:

• Link:
• Auto-balancing bridge front-end circuit
• Data acquisition/digitization I use a NI-DAQ 6363. You could use other hardware.
• Software written in Labview
• TDMS file saving
• It can measure impedances from 10mOhms to 10MOhms
• Test at frequencies from DC-100kHz (up to 250 kHz still needs validation)
• Perform frequency sweeps on log or linear spacing
• If a battery-powered computer and DAQ is used, the entire measurement apparatus can be both portable and floating.
• BOM cost ~$50-$100 (depends which INAs, BNCs, etc. you use). Not including NI-DAQ.?
• Four-wire (Kelvin) measurement
• Impedance range selectable by jumpers

Board photo?

Example screen shot from GUI:
Frequency sweep


Single frequency impedance statistics/time trace.


Accuracy data (1kHz, resistors) See ReadMe for more details.?


I'm avoiding putting all of the measurement data here to avoid clutter but happy to add more if there is interest. At higher frequencies (toward 100kHz) accuracy is closer to 1%, especially at loads much above or below the shunt resistor value. Also, very reactive loads (e.g. high Q capacitors) have closer to 1% accuracy.

One interesting use case is looking at the impedance of a solenoid (~1 mm OD wire) wrapped on a copper tube (~50 mm OD, ~1.6mm wall), and compare it to a Litz wire air core toroid.?
First the Litz wire:


As anticipated, Litz has constant resistance, and air core toroid has constant inductance.?

Now you can see how profound of an effect the tube has increasing the real(Z) and decreasing L, as well as the skin depth increasing AC resistance.



My interpretation of this:

• Below 50 Hz or so, it is essentially a resistor, the inductive impedance is very small.?
• At 50 Hz - 1 kHz, the solenoid's impedance has an increasing real component and decreasing inductance (decreasing slope of imag.). This occurs as the inductor couples more strongly into the copper tube. The inductance decreases because the magnetic field in the center of the tube is decreasing as the wall effectively shields the interior, and inductance is a measure of magnetic field energy stored.?
• Between 1 kHz and 10 kHz the resistance nearly plateaus and the inductance starts rising proportionally to frequency. Presumably at this point, the skin depth of the copper tube has been met, so the magnetic fields are not significantly penetrating to the center of the tube.?
• Above 10 kHz the skin depth of the wire is being reached and the AC-resistance is increasing as the effective copper area for current to flow in decreases.

There still a lot of room to improve the system, but it is quite effective for day-to-day use in a lab where I do a lot of filter design/building. Going forward I'd like to test it on calibrated loads (R,L, and C), but I don't have access to a wide range of 0.1% components, so it is a bit of a challenge to calibrate the system. Currently I use an Agilent 4263B LCR meter at 1kHz, 10kHz, and 100kHz.

-EM