Measuring dc resistance of inductor by measuring peak

Hi,
I need to measure the dc resistance of an inductor by passing a sinusoidal signal through it while in series with a known resistor. I will know the voltage without the inductor in the circuit and then measure the voltage with the inductor introduced. The frequency will be 100khz or so. (if it is variablle 70k-120khz better) I had in mind an spi dds generator and maybe a digital gain amplifier to adjust the gain according to the read voltage on the adc. I had in mind to measure the peak voltage of the signal to be able to find the new resistance. What is the best method to measure the peak of a 100khz sine signal. I am using PIC24F with 32Mhz crystal and it has a 500Ksps adc.

How do you recommend tackling this so I can start doing some research? I will only consider small value inductors 1uH-10mH.

My friend suggested using PWM and using active filtering on it and even to read with the adc.

To measure the DC resistance, you would want to use an ohmmeter...

My point is to include the skin effect and hysteresis in the scene. As I learned, this will effect the real part according to the frequency. Correct me if I am wrong. Maybe I shouldn't have mention 'dc'.

Hi dear jonny .
You can measure the DC current and voltage ( injected signal with the equipment that you designed it ) . and the simple divide !
Best Wishes
Goldsmith

---------- Post added at 00:36 ---------- Previous post was at 00:35 ----------

And about skin effect , it won't has effect on DC signal.

---------- Post added at 00:38 ---------- Previous post was at 00:36 ----------

And about hysteresis too.

As a school project, my lc meter, would be targeted for amateur switch mode power supplies where, modern designs are using these frequencies.

If you are addressing skin effect, you should ask specifically. DC resistance is a well defined term, that leaves no room for different interpretations. By nature, it can be measured with best accuracy as DC quantity.

What do you want to do exactly ? Do you want measure the AC component and DC component together ?
Best Wishes
Goldsmith

So DC resistance is just a misleading term. Reviewing the original question, I assume that you want to measure complex impedance (resistive and reactive part) for different frequencies. You can e.g. generate a current and measure voltage magnitude and phase. Peak voltage is only a magnitude parameter and doesn't tell about phase, so it doesn't contain the intended information.

My suggestion is to analyse the measurement circuit in terms of AC circuit theory to understand the meaning of involved quantities.

You should define an equivalent circuit for the device under test, e.g. a series circuit of an inductor and a resistor. The you can calculate, how both elements show in a particularly measurement circuit. A graphical representations of complex voltages and currents ("phasors") is usually helpfully for understanding.

While skin effect can be represented as frequency dependent resistor in the model, non-linear core parameters like "hysteresis" can't. All core related measurement results also depend on the signal magnitude. You'll hardly be able to access them exactly in your simple measuremet setup.

Combining a small signal AC measurement with superimposed DC current would be one of several options to access non-linear effects.

Regarding, the measurement of the complex impedance you mentioned - I will already have a value of the Inductance (therefore reactance).
The setup i am describing is only for measuring the resistance depending on the frequency and magnitudes(as you pointed out).


Your last sentence is of very interest to me. Can you specify further how I can create a real world scenario of the critical stages of a switch mode power supply? The design is preferably adjusted accordingly from a uC. thanks for your time

A real world power inductor will be loaded by large signal AC of non-sinoidal waveform. In so far, all standard small measurements don't meet the real operation conditions. They can be still useful to access particular parameters, e.g. skin effect losses. At the end of the day, you're asking for total inductor losses respectively it's share of SMPS efficiency and don't need to know how much is skin effect or core loss related. Even detailed manufacturer datasheets rarely tell about it.

To understand your actual intentions, it would be interesting, if you are designing power inductors yourself or applying catalog products.

In both cases, most parameters of practical interest can be at least estimated from datasheet information and don't need necessarily to be measured. A basic understanding of magnetic circuit theory is required however.

Oki, you gave me a more clear idea of the scenario. I had this idea in mind because when I asked about smps and how critical it is to know the resistance by core losses etc - some teachers told me that it is very important since today they use frequencies of about 100Khz. And yes, My intention is to wind my own inductors later on with ferrite cores and suitable winding wire. In your experience, is the resistance very critical(irrelevant if it is low or high) in the designing of (medium power) SMPS if one uses proper materials? What are the ranges of inductor values used? because that also makes a difference when considering the ratio of inductance to resistance.


Going further, are there formulas that can deduce the q-factor from physical properties and dimensions when winding your own inductors?

Answering the questions in detail would turn in lengthy post. As a general remark, copper losses matters (among others), particularly skin effect losses of thick wires at high frequencies. It can be described by a resistance factor. I suggest the free EPCOS ferrite tool for inductor calculations, it also includes a skin effect calculation for solid copper wire.

Inductance range is matter of circuit design.