Measuring the self resonant frequency of an inductor?

[LooseJunkHater]

I'm trying to learn more about inductors and their different electrical parameters. Looking at this [this https://www.vishay.com/docs/34104/ihlp-2525cz-01.pdf] datasheet, it appears that *generally* the inductor is most efficient (Q factor) when it's not used near its self resonant frequency (SRF). For this specific question, I'm trying learn how to easily measure the SRF of an inductor.

I came across [this video] (https://youtu.be/tjbK4LsOQRk?si=sCQUETSEut7MaGTo)(schematic shown at 16:50) which clearly shows how to easily measure the exact SRF, but I'd love to be able to see the SRF of an inductor on a graph, like [this https://images.nagwa.com/figures/explainers/108147821682/2.svg ] datasheet, it appears that *generally* the inductor is most efficient (Q factor) when it's not used near its self resonant frequency (SRF). For this specific question, I'm trying learn how to easily measure the SRF of an inductor.

Edit:

It looks like from [this](https://youtu.be/C3wYMZ_cjdQ?si=dR9qZnzfRtovJ27m) video, I can use a similar methodology to measure the SRF of a capacitor. 

[LooseJunkHater]

For future reference, it looks like [this] (https://youtu.be/FjLfrrwpjBs?si=w9Z6sLUT7vDWc1D8) video is a good reference on how to measure the saturation current of the inductor.

[LooseJunkHater]

Is this the solution, maybe? To use a frequency sweep using a AWG and the FFT mode of my oscilloscope?

https://youtu.be/DqjhTfOy9NY?si=BH0YSjCC2Id0hN7S&t=551

[LooseJunkHater]

Again, another possible solution?

https://youtu.be/uMH2hGvqhlE?si=g7B-Ipr9sbc3_idT

I'm not sure whether either of these two solutions will be appropriate for my use-case. Maybe someone smarter can provide me guidance?

Thanks!

[LinuxHata]

I usually connect inductor to VNA and self-resonance can be clearly seen on frequency sweep.

[CaptDon]

Take the output of your signal generator through a resistor of perhaps 470 ohms to drive the inductor with the 'lower' end of the inductor at ground return. Connect your oscilloscope across the inductor (low capacitance probe is best) and sweep the inductor. At the parallel self resonant frequency you will see a huge rise in voltage, usually even bigger than the applied voltage. You probably won't see a series self resonant frequency but if you did it would appear as a drop in voltage. The resonance can be very sharp so sweep slowly.

[LooseJunkHater]

I usually connect inductor to VNA and self-resonance can be clearly seen on frequency sweep.

Can you explain this some more? What is VNA?

[LooseJunkHater]

Again, another possible solution?

https://youtu.be/uMH2hGvqhlE?si=g7B-Ipr9sbc3_idT

I'm not sure whether either of these two solutions will be appropriate for my use-case. Maybe someone smarter can provide me guidance?

Thanks!

I tried following Dave's video and I have absolutely no idea how to replicate his DIY bode plot.

I set my function gen to output a 1v sinewave, 1s sweep from 10KHz to 1MHz. Do I have to use all of these exact settings to get an accurate bode plot on my scope? I have no idea.

On my Siglent scope, I could not figure out what settings to use either.

I tried to follow [this](https://youtu.be/DqjhTfOy9NY?si=KjPaDzponilCnw4b&t=551) video as well (using the FFT function), but I have no idea how the inductor is wired in the circuit.

I think I need a "schematic" essentially showing whether I need any resistors, how to connect the inductor (to ground or only on the positive of the AWG), and where to probe.

[LinuxHata]

VNA is Vector Network Analyzer.
Simply saying, you connect your component of interest to it, and run the frequency sweep.
On the screen you will see how S-parameters change in Realtime, so you can easily find the self-resonance frequency. Some VNAs can do that for you.

[rteodor]

I usually connect inductor to VNA and self-resonance can be clearly seen on frequency sweep.

Can you explain this some more? What is VNA?

Just a very quick example

Later edit: On NanoVNA Saver I got a peak of S11 Z at ~2.5MHz but directly on the display I get the peak at ~7.3MHz. I don't know what's going on.

[TimFox]

Here's a very simple circuit that I used with a spectrum analyzer:

The two capacitors C1 and C2 are chosen to have reactance much less than 50 ohms over the frequency range of interest.
Ltest and C_Parasitic form the self-resonant circuit.
I have used this with a known capacitor between C1 and the inductor to measure the inductance and resonant Q of coils.
J1 and J2 connect to 50 ohm source and load, respectively.

If you sweep (manually or automatically) the frequency, you will see a minimum output at the SRF, where it parallel resonates.

[T3sl4co1l]

Most efficient -- in what sense?

For example, the impedance of an inductor is maximum at its (lowest frequency, parallel) self resonance.  It never gets "more efficient" at choking an AC signal!

An alternate definition of Q can be used to express its "efficiency" near resonance.  Note we must distinguish between the Q factor of the terminal impedance itself (that is, the resistance and reactance of the component at a given frequency), and some kind of inference about what's inside it (e.g. an RLC parallel equivalent that describes the impedance peak).  In the latter case we can define a Q factor at resonance, and it can be measured for example by peak bandwidth, or some reasonable assumptions about the values of L and C approaching the SRF.

You may find playing with / thinking about this model, to be of interest:
https://www.seventransistorlabs.com/Calc/Coilcraft1.html
Granted, the skin effect / core loss elements aren't so straightforward, but they can be set to zero and the basic RLC circuit still used.

Tim

[LooseJunkHater]

I usually connect inductor to VNA and self-resonance can be clearly seen on frequency sweep.

Can you explain this some more? What is VNA?

Just a very quick example

How that is such a cool tool... But likely something I won't purchase for a very long time lol

[LooseJunkHater]

Here's a very simple circuit that I used with a spectrum analyzer:
(Attachment Link)
The two capacitors C1 and C2 are chosen to have reactance much less than 50 ohms over the frequency range of interest.
Ltest and C_Parasitic form the self-resonant circuit.
I have used this with a known capacitor between C1 and the inductor to measure the inductance and resonant Q of coils.
J1 and J2 connect to 50 ohm source and load, respectively.

If you sweep (manually or automatically) the frequency, you will see a minimum output at the SRF, where it parallel resonates.

Can you explain the purpose of C1/C2/R1/R2 and choosing the right components? Further, what would the expected waveform look like? From the videos I've watched on YouTube, I haven't seen anyone use the same type of circuit that you are using.

[TimFox]

R1 = R2 = 50 ohms are just to give reasonable terminations for the input and output.
In parallel with each of C1 and C2 are 100 ohms (series combination of 50 ohm source and R1, and 50 ohm load and R2).
Large values of C1 and C2 in parallel with 100 ohms are approximately equal to similar values of C in series with much smaller resistances.
(See series-parallel transformation for impedance/admittance in elementary textbooks.)
Therefore, you end up with a large capacitance and small resistance in series with the inductor under test.
You drive the left side (R1-C1) from a variable frequency sine wave:  this could be an AWG, the tracking generator of a spectrum analyzer, the source of a VNA, or other AC generator.
You can see the output at the right side (R2-C2) on any measuring device you want, with a 50 ohm termination:  voltmeter, oscilloscope, spectrum analyzer input, VNA, etc.
When the inductor self-resonates (parallel resonance), the total impedance across it is much larger than the impedance at R2-C2 and the signal will be a minimum.
I have never posted anything on YouTube, but I have used this circuit to measure inductors.
I just mounted two BNC connectors on a single-sided circuit board, with the two capacitors mounted on appropriate stand-off insulators to fit the inductor.

[LooseJunkHater]

Based on my understanding of that circuit, it doesn't seem like it would be possible to see a graphing of the SRF of an inductor with that circuit. I'm hoping to be able to see a graph[/b], like [this https://images.nagwa.com/figures/explainers/108147821682/2.svg ] and be able to create it using my oscilloscope + AWG.

[TimFox]

No, the resulting graph will be inverted from that example, since the self-resonant frequency will have a high series impedance and a low output.
Worked for me.
You need to make sure that there is minimal capacitance in parallel with the inductor to do SRF.

[LooseJunkHater]

No, the resulting graph will be inverted from that example, since the self-resonant frequency will have a high series impedance and a low output.
Worked for me.
You need to make sure that there is minimal capacitance in parallel with the inductor to do SRF.

Okay this makes sense!

My AWG (Uni-T UTG932E) actually has a 50-ohm termination mode; does that mean I can skip on R1, or should I include R1? Furhter, for C1/C2, is a 5uF ceramic capacitor large enough or should I use something larger?

I assume that R1/R2, and C1/C2 don't need to be "perfectly" matched; it's fine if there is like a 5% variance between the values? Is that correct? Will it effect the SRF measurement?

[TimFox]

Matching between R1/C1 and R2/C2 is not critical:  the important thing is that the reactance of each of the two capacitors should be much less than the resistors, over the frequency range of interest.
R1 and R2 are in series with the signal path:  both are needed.
The trick is to get very low resistance in series with the inductor (parallel-series conversion of the impedances at each end).
5 uF?  What frequency range are you using?
At, say, 10 MHz, that capacitor is probably above its series SRF and looks like an inductor.
At 10 MHz, I used about 5 to 10 nF.

[LooseJunkHater]

Matching between R1/C1 and R2/C2 is not critical:  the important thing is that the reactance of each of the two capacitors should be much less than the resistors, over the frequency range of interest.
R1 and R2 are in series with the signal path:  both are needed.
The trick is to get very low resistance in series with the inductor (parallel-series conversion of the impedances at each end).

So I guess R1 is not needed with my AWG...?

5 uF?  What frequency range are you using?
At, say, 10 MHz, that capacitor is probably above its series SRF and looks like an inductor.
At 10 MHz, I used about 5 to 10 nF.

Most of my inductors were pulled from either switching regulator circuits or power supplies, so I wasn't planning on testing above 1MHz initially. However maybe it is a good idea to measure the SRF of the ceramic capacitor before I use it? Should I go with a capacitor in the nF range instead, as it'll reduce loading of the AWG?

[TimFox]

I have always used R1 with a 50 ohm source, to avoid loading the generator with a seriously reactive load.
If you are testing around 1 MHz, then you need to increase the capacitor value by a factor of 10 from that used at 10 MHz.
Ceramic capacitors above 100 nF tend to be bad dielectrics: X7R or worse, that may give bad results in a measurement jig.
At or above 100 nF, your best bet might be a polypropylene film capacitor, although C0G/NP0 are available at 100 nF.

[mawyatt]

Here's something that folks might find useful wrt capacitor/inductor resonance.

https://www.eevblog.com/forum/testgear/lcr-meter-plot-software/msg4774100/#msg4774100

Best,

[TimFox]

That's a nice series of posts.
Note that in his measurement of a 5 uF polypropylene capacitor, the self-resonant frequency (series-resonant mode) of the 5 uF polypropylene capacitor is about 300 kHz.
In my simple circuit, you want to make sure C1 and C2 are operating well below their SRF.

[BILLPOD]

Good Morning LooseJunkHater,  go here:  http://www.learningaboutelectronics.com/Articles/LC-resonant-circuit.php

[fourfathom]

Here's a very simple circuit that I used with a spectrum analyzer:

Tim, can you explain why you use capacitors instead of resistors in the shunt path?  My typical test jig for measuring inductors looks like this (attached image).  I've got a series capacitor in the circuit to so I can calculate the inductance, that cap value will change according to the target inductance and frequency, or will be shorted out if I am only looking for self-resonance.  If I want a sharper response I can always reduce the shunt resistor values (at the cost of more attenuation).