General info on Inductors?

rwgast_logicdesign

Over the past few months I have found myself needing inductors more and more, I usually have clearly labled parts on hand making most projects fairly easy to get set up with but this is not so with inductors, they are always a thorn in my side!

My first question is how one would go about measuring an inductor using a scope. All the tutorials I've seen on Google are fairly complicated at least for me. I was wondering if there is a simple way to get a rough estimate of an Inductors size using a prop scope, If I remember right you have to pulse it to measure it, so I'm hoping the function generator on the prop scope could do this part. Most tutorials I have seen require you to set up a separate circuit on top of a whole bunch of stuff. Basically I have a few Inductors laying around and I forgot there value an it would be nice if there were a way to get a quick easy rough estimate of there size. It does not have to be dead on +/-5uH should give me a decent idea of what I am looking at. I have a meter that measures caps decently but the ones that can do Inductors are a bit out of my price range.

Secondly picking the right Inductor style is a bit beyond me. When I built my switching regulator I got some inductors from vishay and they are HUGE, much bigger than inductors on most switching regulator circuits, they are only rated for one amp....

This is a picture of a 500mA switching regulator using a 330uH and 20uH, I guess I ordered open core inductors? The Inductors I see on things like the prop boe are much smaller, and handle more amperage! Also I buy from DigiKey and there choice of through hole inductors is kind of crumby, is there any thing wrong with putting two 10uH inductors in series to make a 20uH inductor, will this raise ESR (if that is even applicable). Lastly when making simple LC filter is there any issue with the axial Inductors that mount like a diode/resistor?

And finally the last question, exactly how hard is it to just make your own inductors? It would seem to me that you would need to be able to measure them on a scope or something to do this well? I guess what im getting at is the core has to be sized correctly etc, etc, so is it possible to just buy some materials that you can keep on hand to wrap any size inductor you may need, maybe certain diameter ferrite cores in sticks you can cut, and a few gauges of magnetic wire to cover all your bases?

Beau Schwabe

This works fairly well:
http://wiki.ohmspace.org/Determining_an_unknown_inductor_value_using_a_%27known%27_inductor_as_a_reference_value

Here is how to make your own air core inductors:
http://wiki.ohmspace.org/images/e/e5/CoilCalc_08-18-2008_.xls

Don't know the wire gauge? Try this out:
http://wiki.ohmspace.org/Simple_method_for_determining_wire_gauge

Duane C. Johnson

Of course it would be nice to have a commercial inductance bridge.
However, you can make your own and get reasonable accuracies.
Bridges

Your PropScope has, presumably, a AC output which can be applied across this bridge.
You can then measure the 2 nodes, signified by the meters in my diagram.

Adjust the pot until the scope shows a balance.
The ratios of the 2 inductors will be the same as the ratios of the 2 halves of the pot.

Note! select the frequency so the reference inductor impedance is in the range similar to 500 ohms, in the example.

This general circuit will work just as well with capacitors.

Duane J

Dr_Acula

I've found an LCR meter very handy. They have come down in price a lot recently eg http://www.ebay.com.au/itm/LCR-RCL-INDUCTANCE-CAPACITANCE-RESISTANCE-METER-Leads-/130756957647?pt=AU_B_I_Electrical_Test_Equipment&hash=item1e71b8d5cf $20.

frank freedman

My Amprobe multimeter does inductance. Only one I found in the sub $200.00 range.

rwgast_logicdesign

O wow thank you for the answers, the link provided by beau seems to be the most simple solution, as long as you have something that outputs a known frequency. I also had no idea you could get a $20 LCR meter!

The one thing I still need to know is, is there any issues with putting inductors in series? For instance 2 10uH's to make a 20uH's, Like I said I mainely stick with digi key and I am having a hard time finding some values such as 20uH.

Dr_Acula

Inductors in series add together, like resistors do. So two 10uH in series is 20uH, like you say.

User Name

@rwgast: Just as an alternative to the excellent answers already given, I mention two methods I use a lot... The first involves tacking the inductor to a known capacitance and looking for resonance by sweeping with a signal generator. Simple math reveals the inductance. The second is a little kit that uses an LM311 and a PIC processor.

rwgast_logicdesign

Wow that is a nice looking kit, it is much more than the eBay meter, do you own one can you recommend it over the ebay meter for some reason?

As far as putting inductors in series, I understand you just add there values and the sum equals the total inductance. What I meant is this technically a good idea, especially with power supply circuits. For example putting caps in parallel will lower ESR while putting them in series will raise ESR. Are there any concerns like this with inductors?

CuriousOne

I have that kit from electronics-diy.

I bought the "deluxe" version, which still has two problems:

1. Measurement runout - you need to constantly re-calibrate in each 10 seconds or so, if you want results with precision to 1%.

2. They don't include the enclosure pictured on website! they just give you plain serpac enclosure, in which you have to file off the lcd screen opening.

3. Veird behavour with some capacitors (which are OK when measured on another meter or used in circuit) - the capacitance rises along the time. Say I've connected capacitor and it shows 22nf. Over time of say 5 seconds, the result may rise to 80nf !


Regarding the inductors, there are few things you have to consider:

1. Saturation current/max current - check your device datasheet for formulas to calculate.
2. Resonance frequency - should be well above the switchig frequency of your device. For some inductors, frequency is not given, but you can look in datasheet for test frequency. If it is say 10khz, that means, this inductor will not be usable with say, 2mhz converter.
3. Shielded vs. Unshielded. Generally speaking, shielded ones are always better, being only slightly expensive.
4. DC resistance. The lower the DC resistance is, the higher is efficiency of the circuit overal, but also, inductor size and price raises considerably.

Regarding your question about putting inductors in series/parallel, yes it is possible, but you should physically arrange them in that way, that no mutual inductance transfer occurs. Using shielded ones will help with this much. Sure, there are designs where coupled inductors are used, for example, SEPIC converters, but that is usually declared in description.

Dr_Acula

For example putting caps in parallel will lower ESR while putting them in series will raise ESR. Are there any concerns like this with inductors?

Well if you put two inductors of the same value in series the DC resistance will be double. So if you kept doing that over and over, eventually the DC resistance would be too high. It may not be a problem for just a few inductors though.

I have had a lot of fun with the LCR meter winding homebrew inductors. Toroids. Different lengths of wire. Potcores. Very interesting to measure the values as they change, particularly as you add more windings. Gotta get back to work - the formula for turns vs inductance is on the interwebs.

Mark_T

Dr_Acula wrote: »

Inductors in series add together, like resistors do. So two 10uH in series is 20uH, like you say.

But that costs twice as much - and its easy to adjust the number of turns if making your own of specifying a bulk order.

Few circuits depend on exact inductance values (ferrite cored inductors change value with temperature frequency and DC-bias anyway),
if you can adjust other R or C component values it is usually easier. Also adding inductors in series is only additive if there is no flux-overlap
between them (so toroids rather than open-bobbin style).

Circuits where precise inductance values are needed usually use trimmed inductors anyway (RF IF-transformers for instance).

Dr_Acula

Good point Mark_T - that costs twice as much. And to answer my own question about turns vs inductance - the inductance is N^2. So you need less than twice the wire to get double the inductance. 1.41 times I believe. (root of 2).

Completing the magnetic loop improves efficiency too. eg a toroid or a potcore is a lot better than a rod of ferrite. Keep those magnetic field lines inside the material rather than letting them travel through air.

Agree that the formulas only work if there is no flux overlap.

My LCR has been very handy testing all these things with real world experiments.

Duane C. Johnson

Hi Dr.;

Dr_Acula wrote: »

Good point Mark_T - that costs twice as much. And to answer my own question about turns vs inductance - the inductance is N^2. So you need less than twice the wire to get double the inductance. 1.41 times I believe. (root of 2).

Yes, more turns increase inductance by the square rule.
But, keep in mind, the AmpTurns rating for the core is still the same so as the turns increase the saturation current will decrease.

Duane J

Mark_T

Duane C. Johnson wrote: »

Hi Dr.;
Yes, more turns increase inductance by the square rule.
But, keep in mind, the AmpTurns rating for the core is still the same so as the turns increase the saturation current will decrease.

Duane J

Indeed, but I'm more thinking about small adjustments to get non-standard values rather than doubling. Optimizing
inductors is complex given all the variables (gap size, turns, ferrite material, core size)