Measuring Inductance, best way?

Go To Last Post
24 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

One of these days, I plan on building a meter circuit that also measures inductors and capacitors (like the microchip "Engineers Assistant", but modified to do more).

 

I had planned on building this little LCRF meter project from EPE Electronics's (UK Magazine) John Becker.  I spent a little time with it, but it was kinda buggy and used a PIC processor (which as I understand it, is now virtually obsolete).  The method he used for both L/C was to use two different oscillators, one for L and one for C, and inset the DUT into that circuit and measure it's change (like putting one cap in parallel with another).  But in his description for the L measurement, frequency wasn't quite enough, and he did something with the phase.  Never really figured out what.

 

After searching on the internet, I found several other methods, but none I particularly l iike.  I don't want to have to generate a nice sine wave or anything.

 

One method I used many years ago to measure capacitors with a frequency counter was to use a constant current source, apply that to the cap, and measure the time it takes to charge (then trigger a flipflop and do the same thing, but on discharge).  It worked quite well and produced a very nice triangle wave.  But that was around 40 years ago and I can not find my notes on that circuit.  I know I have them though, just packed up.

 

But I'm thinking, I should be able to do (virtually) the same with with an inductor.   Has anybody tried this?

 

Or is there another preferred method for measuring inductors with a micro controller?

 

Just looking at ideas, and working on the code to make them a reality, just little "mini-projects" so to speak, no real project in the works yet.  Any suggestion would be appreciated.

Just gettin' started, again....

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This is a more modern take on an ages old circuit. Should be easily converted to AVR - it probably has been already.

 

The first time I saw the basic circuit was in an american electronics magazine in to 70's or 80's - no microcontroller involved. 

 

http://www.vk6fh.com/vk6fh/lc_me...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The method you use depends strongly on the frequency at which you want to make the measurement. Hertz? KHz? MHz? GHz?

 

Are you interested in inductors for power (e.g. switch mode converters) or RF (typically filters and tuned circuits)? For receivers or for transmitters?

 

What sort of accuracy? 10's of percent? 1%? 0.1%?

 

Do you want just the effective inductance or capacitance or do you want the reactance and resistance?

 

Lots of questions to answer....

 

Jim

 

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I added AVR to the last search:

 

http://www.kerrywong.com/2010/10...

 

The same old oscillator circuit coming up again......

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

OldMicroGuy wrote:
But I'm thinking, I should be able to do (virtually) the same with with an inductor.   Has anybody tried this?

 

Yes, you can do a LR oscillator, similar to a RC oscillator, but the resonant  circuit linked above seems to be popular, so would be a good place to start.

For wider dynamic range, some meters switch in varying resonant elements.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

in the dutch elector there has been a couple of designs that were on LRC meters. cannot imagine that when you go and look for them you would not find them.

I even recall one version were they made a design that in the first episode had a pig processor, but a couple of editions later they have also places the same analog stuff but now connected to an AVRmega as there had been a lot of comment on the pig being very hadr to get.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

meslomp wrote:
e had a pig processor

 

Those processors are definitely not kosher.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I picked up one of these cheap from china, it seems to work amazingly well, wish I had the case to go with it as shown here: https://www.amazon.com/Mega328-D...

 

Jim

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


 

This might give a few clues as to what you could do. I've got one of these...

 

 

...which looks like this inside...

 

 

The processor, U3, is a PIC16F866

 

U4, U8 and U9 are 4051 analogue muxes

 

U5 and U6 are LPV324 op-amps

 

U10 is a 4046 PLL

 

U7 is an MCP6284 Quad op-amp

#1 Hardware Problem? https://www.avrfreaks.net/forum/...

#2 Hardware Problem? Read AVR042.

#3 All grounds are not created equal

#4 Have you proved your chip is running at xxMHz?

#5 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand."

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

For a simple RL circuit, Tau = L/R.

 

In one of my kid's physics classes a few years ago they had to make their own inductor of whatever value it was.

 

We wound lots of wire on lots of metal rods but the value of L needed was quite high, (I thought).

 

We went to Lowe's and bought ~ 10, 12 inch long big spike nails, bundled them together, wound lots more wire, --> no go.

 

Finally we bought a couple of toroid's on line and finally got enough inductance to meet the requirements.

 

We started with a simple PB Switch to turn the battery voltage on and off, later I did it with a micro and a slow square wave.

 

We measured the RL curve on an O'scope to test our various coils, and to get the winding right when we finally got the toroids.

 

For your project, one could switch the Vin with a small relay, or a FET.

With a known Vsupply, and therefore a known voltage for a given Tau, one could use a voltage comparator to trip on the signal.

 

Then Use a micro to measure the pulse width, do the calc, and display the result.

 

The micro could switch different R's into the circuit to set a reasonably long RL time constant, for decent accuracy.

 

JC 

 

Edit:

Added a photo of the homemade inductors.

Can't seem to find a photo of the O'scope curves.

 

Last Edited: Thu. Jul 30, 2020 - 06:34 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Wrapping wire around nails is building a rail gun, not an inductor.  Hopefully few of those nails wound up halfway though the wall of the barn!  Your toroids were made of ferrite, which is significantly different (see:  eddy currents).  Anyhow, sounds like a fun experiment for a curious physics student.  Wish more schools did that.  S.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have one of those but I'm not that crazy about it.  Short/fussy test leads, turns off too quick.  But it works well enough.  I wonder if they're using just the VCO in that 4046 or maybe the whole thing to measure the 'phase thing" John Becker was speaking of.   I'll post a partial of the schematic see what you guys think, just a gated oscillator (aside from the power supply, micro and LCD).  And 3 4051;s, that's a lot of multiplexing, ranges?  Switching parts in/out for ranges?   I'll have to open it up later and look around.

 

DocJC, that sounds like what I was talking about.   Just like measuring the charge/discharge slope on a capacitor, I'd think you should be able to do the same thing with an inductor.  I just used a constant current source to control the slope I guess.  Did stretch the time out and make it more linear.  I was thinking of doing the same thing with an inductor, but just never saw anything on that out there.   Then I could use the same basic circuitry for both parts.   If I remember correctly, I had two comparators in there with precision voltage references, one for 1/6th and one for 5/6th of the supply (standard "time to charge" levels).   Worked well, direct readout on a frequency counter, T=C I think (so many years ago...  dang I wish I could find those notes.  Found a bunch of stuff I didn't know I still had,  but no notes).   

 

Yep, toroids are almost always better.  I used them often when I build little DC/DC converters.  Hate winding them, but I have a bunch of cores.  Most people don't realize how easy it to wind your own, just consider the ratio of number of turns in/out.    (that was where we saw such weird stuff on those circuit breakers, instead of toroids, we used a split bobbin, half the number of turns on one, half on the other.  Works Ok up to about 2KA, but above that, you get the two bobbins/coils, "fighting" each other.  One trying to drive the current high, the other trying to drive it low, VERY high voltage results)

Just gettin' started, again....

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Unless you are very careful about it, the curve is not a great way for measurement...it is very "sensitive" (in the wrong way), since it is exponential.  At least the freq response curve (resonant freq vc L) is a 1/(f^2) shape, which is "somewhat more linear" than the exponential..

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Thu. Jul 30, 2020 - 10:05 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Of the 'simple' options:

1. measure charge/discharge time

2. do that enough times you get an oscillator

3. or measure the complex impedance by applying a sine wave and measuring the phase and magnitude. Sweep the frequency. Aka VNA.

 

#3 was usually done with a component analyser which was an expensive piece of equipment. Nowadays with modern technology, it is much easier. Methinks Analog Devices have such things.

 

Also note that measuring anything but an air cored inductor is challenging - depending on the current through the inductor and the core specs, the inductance can vary significantly. Maybe that's an area that needs a better solution? Ie measure an inductor at various frequencies and current. I dare say those who make switchmode supplies would be looking for such a device. Measure at 50kHz increments at to 2MHz and 50mA up to 4A would probably satisfy many. I day say a switching technique would be used - generating a sine at 2MHz and 4A isn't something you'd do with a breadboard!

 

But, if you want something cheap 'n' cheerful, then the lm311 circuit seems to have stood the test of time. Simply measure the frequency, do a little calc and you've got a value.

 

The original circuit I referred to in the american electronics mag used an eprom to lookup the solution rather than a microcontroller and displayed the result on a 7 seg display. I might see if I can locate the magazine in my pile.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

https://www.mikrocontroller.net/articles/AVR_Transistortester

It is amazingly accurate for little more than an AVR and some resistors.
I have two of the Chinese produced units and a European based version in a nice hand held case.

Plus you get BJT, FET, diode, etc testing capability as well if you want.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This could turn into a "real" project before too long.  While I was looking for my old notes, I found another old notebook of mine that has tons of great information on inductors and transformers, just data I collected from various sources.  I like the idea of building on it, measuring various aspects of inductors etc...   We'll see.  Right now I just want something on my bench to tell me what I actually have (I have a lot of parts I salvage from stuff).

 

But I would like to see that project with the EPROM.   Sounds interesting, always up for something "new".

 

JamieScott......  That's a great website!  Lots of stuff on there.   Good thing google translates from German into English, I never learned to read German (though I did speak some when I lived there, but that was before I learned to read and write!)

 

Lots to sift through, found the project, think I downloaded everything, but hard to tell.  4 files, one of which is over 100Mb, so I think I got evvveeerrrrthing (at least pertaining to that project).

 

But looks like lots of other goodies out there too.  I like it!  Thanks for that.

Just gettin' started, again....

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This probably goes without saying, but I'll just mention it...

 

As you start playing with stuff, make sure you order some inductors, preferably with tight tolerances, and with attached data sheets.

 

Then label them so you know which is which!

 

You will then have some "reference standards" against which to compare your project's reported value, and against which to compare your unknown values.

 

Sounds like a fun project.

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Not sure how far it can go.  It'd be a good learning experience for sure.  But as I mentioned, my workspace is currently my lap.  My fiance's "kids" (40 years old!) need to move out (again) first.  Then the goal is to get a big ol' Rv and travel around and see everything we haven't seen.  But the stock market is putting a delay in that.  I can still do a good bit in an RV, but just have to either plan ahead and take what I need, or order stuff on the road.  But supplies will be minimal for sure.   I'm currently trying to compact all my "goodies" into a few small tubs that I can take, or take what I need and store the rest.  (parted out a whole bunch of computers, including two SG O2's and 2 SG Irix's, good parts!)  Hopefully I'll get something started soon enough.  (I'm telling ya, when we hit the road, changing the phone numbers, and no forwarding address, we'll send the "kids" a postcard a couple weeks after we leave a place.... "oh it was so nice, you should have been here, Ha!  Just kidding!!"

Just gettin' started, again....

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:

 

You will then have some "reference standards" against which to compare your project's reported value, and against which to compare your unknown values.

 

 

You're right, I have plenty of precision resistors (like .01% Thank you Vishay....) and a few precision capacitors, but no inductors.  Never needed them until now I guess.

 

Just another unfinished project from years ago, but now I can build on it and really learn something.   That's one reason I like this and the Engineers Assistant, covers the whole gamut of programming.  Lots to learn.

Just gettin' started, again....

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Digilent Analog Discovery's Impedance Analyzer has a reference resistor.

Wavegen 1 into Load and reference resistor as an impedance divider, Wavegen 1 into Scope 1, Load and Resistor node into Scope 2

Hardware Setup | Using the Impedance Analyzer [Reference.Digilentinc]

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yay! I found the magazine and a link!

 

Note - the download is 30+MB

https://ia801703.us.archive.org/...

 

See page 41.

 

There's also some line carrier modems for a bit of retro interest.

 

Happy reading!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yay! I found the magazine and a link!

 

Note - the download is 30+MB

https://ia801703.us.archive.org/...

I forgot what a good electronics magazine looked like...could barely put it down...saw a lot of old friends. 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I used to read that long ago.   That and Popular Electronics.   I should have thanked Forrest M. Mimms for the "LED Scope" idea.   Still have one of his books of projects.

 

EDIT:

That rings a few bells.  I forgot about Don Lancaster.  Used to read him all the time.   Back on page 105 I think, top/middle of the page is an ad for a voice remover.  Can't believe that guy was advertising that way back then.   Been seeing that ad for years.  I almost went to work for him in Stone Mountain (he moved).  Went and spoke with him, he made a small fortune off that thing and the other stuff he sold.  But he wanted to move on to other stuff.  So he wanted me to take over for him (doing what made him a fortune) for minimum wage.   Had to tell him 'No thanks'....   Sheesh.

Just gettin' started, again....

Last Edited: Sat. Aug 1, 2020 - 02:16 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How about something from Dick Cappels...

 

http://cappels.org/dproj/Home.htm

 

...several LC meters there?

#1 Hardware Problem? https://www.avrfreaks.net/forum/...

#2 Hardware Problem? Read AVR042.

#3 All grounds are not created equal

#4 Have you proved your chip is running at xxMHz?

#5 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand."