Inductance Tester

Philldapill

For a project that is on the back burner, I need to build a fairly precise inductor. This inductor uses some big 10 gauge wire around a 4" diameter ferrite toroid. I know that if I build an RLC circuit using the inductor and a precise value capacitor, I can find the resonant frequency of the circuit. If I know the resonant frequency and the capacitor value, I can figure the inductance to a fair degree.

Has this been done with the propeller? If not, I'm thinking of a simple way to do it, involving the propeller adjusting an output frequency, and measuring the peak to peak. When the peak to peak is the highest, there is the resonant freq., and in turn you can calculate the inductance.

Leon

I've measured inductors with a signal generator and scope, or a diode probe attached to a meter, by measuring the resonant frequency of the coil in parallel with a capacitor. You could use the Propeller as a signal generator.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

Beau Schwabe

Philldapill,

There are several applications in the object exchange, that deal with reading coils (inductors)... some use a peak resonate method, others make use·of a current mirror.

There are a couple of other methods in the Propeller Forum as well:

http://forums.parallax.com/showthread.php?p=598919
http://forums.parallax.com/showthread.php?p=594827


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

heater

I haven't had time to check those links yet but one inductance/capacitance measuring technique is to create an oscillator out of a resonant RC tank and an inverter. Then just measure the frequency it oscillates at. Many people have done this with for example an op amp and a PIC or AVR uC. For example
www.elecfree.com/electronic/lc-meter-based-on-the-avr-microcontroller/


Now the Propeller timers have an inverter built in that will (in the correct mode) provide an output on a pin that is the inverse of the input on some other pin, so all it needs is the R and C and a couple of biasing resistors to get that to oscillate then use the timer to get the frequency.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

heater

Here is an example of an LC and inverter oscillator.
www.discovercircuits.com/DJ-Circuits/125kpowerosc4.htm

I'd just like to replace the inverter with Propeller timers built in inverter for a minimal parts LC meter. Havn't had the time to experiment with this idea yet.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

Beau Schwabe

heater,
·
Of course, somewhere there should be an inverter style LC oscillator where you can just count the pulses, but be cautious of this method.· An inverter style LC oscillator doesn't actually allow the LC combination to resonate at their resonant frequency because of propagation delays in the inverter.· These delays can·be significant.
·
The Current mirror approach is probably the fastest method... basically you "PING" a coil and use a current mirror to charge a cap.· The way that the cap is charged is proportional to how much energy is in the collapsing magnetic field after the ping.· You simply read the value of the·cap with an ADC.· It only takes about·a hundred micro seconds to determine a coil value using this method. (most of that 100us is reading the ADC)
·
The resonate approach by frequency sweeping is probably the most accurate, but takes·time to sweep the·frequency range of the coil in addition to·taking samples along the way.
·
·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Post Edited (Beau Schwabe (Parallax)) : 7/18/2008 6:58:40 AM GMT

heater

Beau, Interesting. Do you have any suggestions for a schematic to get a Props timer inverter oscillating with an LC tank. I like the idea of a minimal parts LC meter. There must be some useful accuracy over some LC value ranges despite the inverter delay.

No idea how the current mirror idea would look, any pointers ?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

Drone

Hi Phil,

Have a look at, "An LMS Impedance Bridge, By Dr George R. Steber WB9LVI" (link to free article below). This is an accurate LC meter made with a single-chip LM358 dual general-purpose op-amp. The software uses the least-mean-squares method in conjunction with the PC sound card (link to free software below). This technique could be adapted to the Propeller, or just use Dr. Steber's program with your PC!

www.arrl.org/qex/2005/Steber.pdf

www.arrl.org/qexfiles/LMSProg.zip

On page 64 of the June 2008 issue of Elektor Magazine there is an article by Martin Klaper and Heinz Mathis entitled, "2-Pound RLC Meter, impedance measurement using a sound card", the "2-Pound" part refers to a laptop computer.

www.elektor.com

The Elektor article uses a circuit identical to the one in the Steber article. So you might want to try out the Elektor software as well - you can download it for free without buying the article. In-fact there's no point in buying the Elektor article IMO as it is almost word-for-word and picture-for-picture identical to the much older Steber work! There is no reference to Dr. Steber in the Elektor article.

Let us know what you end-up with and how it works. BTW I need to measure inductances in tenths of microhenries. I haven't been able to find a PC-sound card technique that can go that low.

Best Regards, David

Philldapill

Wow, I never thought I'd get so many responses on something like this. Great!

I simulated your approach, heater, and using an inverter messes with the true resonant freq., as Beau mentioned. What I have in mind now, is fairly simple and I think it should work. I'm thinking running DC through a parallel RLC circuit, then abruptly cutting the power. The circuit should resonate for a few cycles before the resistance dampens the oscillation. During this time, I think I can use an op-amp and have it output a high signal when the voltage across the resistor is positive, and output a low signal when the voltage is negative. From here, it's a simple matter of feeding that into the propeller, and using a counter or something in assembly to count X pulses, and record the time it took for those pulses. From there you can get the period, then the frequency, and if you know the capacitance, voilla! There is your inductance.

Drone

Phil,

There's a toroid design program that may help you. It is called the mini-Ring Core Calculator. Find it at:

www.dl5swb.de

What is the expected value of inductance you are trying to measure? At what frequency will it operate? What is the expected length of the 10 gauge wire wound on the core? Do you have any information on the core?

Rgds, David

Leon

Another technique is to build an LC oscillator with another inductor, measure the frequency, and then put your inductor in parallel with the other one and measure the frequency again. Knowing the frequency difference you can calculate the unknown inductance. This technique is used in a couple of LC meters, like the AADE one popular with radio amateurs.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

Beau Schwabe

Leon,

Here is a method based on·what you describe to determine an unknown inductor based on results from a known inductor:


1) Determine the resonate frequency of your known LC values.

2) Substitute the known inductor with the unknown inductor, and measure the frequency.

3) The unknown inductance value can be determined by the following equation:


F1· = measured frequency from step 1
uH· = known inductor value
F2· = measured frequency from step 2
uH! = unknown inductor value


uH! = uH * F1^2 / F2^2


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Drone

Most posts in this thread require a "known" inductance to "measure against". My post with links to Dr. Steber's paper and associated software allow you to measure an "absolute" value for a reactive component, a "reference" if-you-will.