4

I'm wondering why I can't find an LCR meter with frequency band that starts for example from 1 Hz. Is there some specific reason?

EDIT: My question comes from the need to measure the impedance of a generic "component". I create a meter with an wave generator and an oscilloscope based on this configuration:

enter image description here

The circuit inside the yellow rectangle is the impedance to measure. The source is my wave generator and the Vout is where I read the output voltage with oscilloscope. Then I compared the SPICE simulation with my measure:

enter image description here

The continue lines are the simulated impedance and the circles are my measures on a similar circuit made by discrete component with same values (resistors and capacitors through-hole). The values of the components are reported in legend.

With this example I want to focus your attention on the fact that this kind of impedance varies at very low frequencies, 0.1-10 Hz. This is the reason why I'm wondering there aren't LCR meter at that band.

thoraz
  • 389
  • 2
  • 13
  • 1
    reactance at 1 Hz would be less than the resistance. That's usually why, although for EHV insulation testing , they go to 0.001 Hz but then they use ~50k$ impedance Instruments – Tony Stewart EE75 Jul 03 '18 at 15:19
  • From what I gather, your chart shows an oscilloscope and signal generator yielded results with good correspondence to theory. So that answers your question! Why spend money for a dedicated instrument when basic equipment works fine at such low frequencies? – user71659 Jul 03 '18 at 19:21

3 Answers3

7

They are custom made. For example, this one which works down to 10 µHz (100,000 seconds period, 1.2 days).

But they're not commonly needed, not in electronics. Sometimes in material science measurements. When we had to make measurements of permeability of a special ultra-pure metal we just threw together an experimental rig with lock-in amplifier, power driver and some other instruments we had laying around rather than spending tens of thousands of dollars on a specialized instrument, probably similar to what you are doing. The very low frequency was necessary because of eddy currents. We made vector measurements and extrapolated down to extreme low frequencies.

Spehro Pefhany
  • 376,485
  • 21
  • 320
  • 842
  • Ok, I got it. With "custom" do you mean that the company designs the meter following specific request of a customer? – thoraz Jul 04 '18 at 07:37
6

At DC, capacitors are open, and inductors exhibit only their DC resistance. In other words, capacitors and inductors don't do what makes them special at DC. It takes changing voltage or current to see the effect of capacitance and inductance.

The closer you get to DC the harder it is to tease out the capacitive and inductive properties. Theoretically it's possible at any frequency above 0, but the lower the frequency, the less signal there is to detect out of the noise.

1 Hz is so ridiculously low that it is useless in a pratical sense for measuring anything but unusually large capacitances and inductances.

Olin Lathrop
  • 310,974
  • 36
  • 428
  • 915
  • 1
    Not so rediculous to those instrument designers in ELF field ( magnetic grid component insulation tests, earth sciences, etc. -1.5 – Tony Stewart EE75 Jul 03 '18 at 15:40
  • @Olin I edited the post to be more clear. Your answer is a good point. – thoraz Jul 03 '18 at 15:43
  • The classic HP 4192A gets down to 5Hz, a lower limit I assume to be down to the phase noise of its IF chain. – Dan Mills Jul 03 '18 at 15:50
  • +1, I'd add that 1 Hz is "human scale": ridiculously large caps and inductors can be measured by hand (e.g. discharge cap through a resistor). The plain old multimeter comes to the rescue when the more sophisticated LCR meter stops being useful! – anrieff Jul 03 '18 at 21:04
  • As Ali Chen's answer implies, the capacitor isn't likely to be a traditional parallel plate one, but rather a model for electrochemical energy storage, like batteries or supercapacitors. (The Cole-Cole plot on that instrument shows something that's a parallel RC in series with a C, e.g. higher internal resistance chemical storage coupled with a faster plate charge) – user71659 Jul 03 '18 at 21:07
  • I agree with @anrieff. In my example the impedance varies in the 0.1-10 Hz band and the component value are not so large! – thoraz Jul 04 '18 at 07:44
3

Commercial LCR meters are designed to measure equivalent parameters of fairly simple "component" models (inductors with some DC and maybe parasitic C, capacitors with ESR and leakage, etc). Your object is more complex and doesn't fit any standard model of commercial electronic components. So you won't find many commercial meters that anticipate your double-layer model. At best you can use some sweeping-frequency LCR network analyzer, and then compare the obtained transfer function with output of SPICE model, just as you already did. There are "chemical impedance analyzers" as Hioki IM3590,

enter image description here

which might meet your needs.

Ale..chenski
  • 38,845
  • 3
  • 38
  • 103