Variable inductance problem?

Thread Starter


Joined Oct 10, 2008
Hi everyone,

I recently made an adjustable inductor using a E core and coil. I adjust the inductance by gapping the core using screws and nuts which I mounted to it. When the core is tight I get an inductance of 5H. The problem is when I turn the 10-32 screw just 1/2 turn the inductance changes by an entire Henry (at that point the gap is only about 2 thousandth of an inch!) The core I am using has an Ae of 225mm^2, a Iu of 2300 and an AL of 3100.

I need the inductance change to be much more precise. I was hoping to get something that I could change the inductance by just a few mH per turn. How can I get more precision out of the inductor?
Last edited:


Joined Sep 9, 2010
The screw is adjusting the air gap? The air gap has a very much lower permeability than your core and the gap's size will dominate the calculation of combined permeability, and thus inductance.

Thread Starter


Joined Oct 10, 2008
Yeah the screw is adjusting the air gap.
I have looked into fine thread screws (200TPI) but they get a little pricey. Not only that but they are very delicate and I think the threads would strip easily if the screw and nut were not perfectly aligned.

What I am wondering is if a different size or shape of core would work better? Or any other options I might have to make the inductance more precise when I adjust it.


Joined Dec 13, 2013
try moving a magnet near the core while reading the inductance. look up magnetic amplifier, a changing magnetic field in the core can change the inductance. an exgtra winding can supply the magnetic bias to adjust the coil.


Joined Jun 17, 2014

What it sounds like you are doing is trying to adjust the gap. That's a standard procedure for adjusting inductance by small factors, but you should realize that this is a very precise parameter that usually has to be measured down to the thousandth of an inch or better. The gap reluctance could be thousands of times higher than the core, and the overall permeability is affected by the ratio of reluctances and distance, so changing the gap by even 0.001 inch could change the inductance twofold or even tenfold, depending on other dimensions.

So your attempt to change the gap is an effective means to change the inductance, but setting that gap to get a specific inductance requires a precise movement on the order of what you would measure using a micrometer.

What helps a little is if you have a core that already has a gap to begin with. That means increasing the gap causes less change per thousandth of an inch than if the core had no gap to begin with.

Removing turns is less problematic but of course this requires physically removing turns, unless you can make a slider with some turns bare on one side so the slider makes contact. This isnt very easy to do either.
If you could make a tapped inductor you could at least have several values to choose from, but of course that's not as variable as changing the gap.

When the whole gap gets changed it changes the magnetic gap by two times, because the flux passes through the gap twice in an EI core. This means lifting only one end could improve the adjustment resolution somewhat, although it is still a precise setting.

Pieces of "I" that are brought close to the gap also will increase the inductance, probably with better resolution. This would mean at least one side would need an additional set of "I' laminations mounted on a movable platform maybe using 4-40 screws or even 2-56 screws for finer adjustment.


Joined Mar 6, 2009
I guess one shouldn't be surprised at the outcome. Inductance will be essentially proportional to the inverse of the air gap length for very small airgaps. The anticipated change in inductance ΔL for a small change in gap Δg would be

\( \Delta L=-L_o \times \frac {\Delta g }{g }\)

where Lo is the inductance with air gap length g.

If Δg is commensurate with g then the effect can be significant.

Thread Starter


Joined Oct 10, 2008
Thanks for the suggestions everyone.
They have helped me a great deal with my variable inductor project.

MrAl, your suggestion on using a core that already has a gap is great. After looking at core specs I can see that the AL value does not decrease linearly with the gap. So if the gap already exists I can gap it further and the change in the AL value will be smaller than what I would get using an ungapped core. Just the insight I was hoping for.

Thank you very much for the help everyone!