# Avoid inductive coupling on two near coils

I have two coils that are placed parallel from each other and separated by a mere 1.5cm, lets say A and B. The core of them is built using soft iron. Right now they are wired and built in the same way.

Is there any way that I can use to prevent the induction from coil A to coil B when coil A is being pulsed by a DC current? If I place a scope with a channel on each coil, coil B shows the same pulsing frequency than coil A but with a reduced voltage.

Is there any way that I can use to avoid this behavior? Or this is how it should be and by physics it can't be avoided?

• If centerlines are perpendicular, induction is reduced. If they must be mounted with centerlines parallel, place the center of each coil at 57° to each other. That way the magnetic field crosses the other coil windings, it becomes self canceling. – Optionparty Feb 28 '14 at 1:36
• Thanks for your comment. What do you mean by "place the center of each coil at 57° to each other"? Doesn't this make them not parallel? – kR105 Feb 28 '14 at 2:07
• Another way to view any number of coils avoiding inductive transfer, is one ahead of the other -_ This was done long ago with TRF receivers, where several amplifiers were in series with coils between them. The amplified output was feeding back to the input, causing oscillation. Placing each coil at 57° prevented feedback unless the magnetic field was disturbed, so they then enclosed each coil in its own metal can. – Optionparty Feb 28 '14 at 16:01

## 4 Answers

Place the coils perpendicular to eachother or use a shielded inductors. If none of these options are possible you could make a shielding out of mu-metal. It provides a good shielding for low frequency magnetic fields.

If the two "independent" coils are each wound on gap-less high permeability cores then coupling is greatly reduced. The problem arises when the core material is low permeability or, (as in the case of a lot of inductors), there is a significant air-gap. This is because the magnetic field "fringes" due to air carrying the flux. Air is a poor concentrator of flux and coupling can happen because the lines of flux "spread-out".

The coupling also depends on operating frequency. Induced voltage is N$\dfrac{d\phi}{dt}$ and the rate of change of flux is proportional to frequency. However, flux is also dependent on the ampere-turns in the "transmitting" coil and as frequency rises (for a fixed inductance value), current falls proportionally.

There comes a frequency (and this is beyond my memory at the moment) where the choice of "protection" changes from using high permeability material to using a solid conductor between the coils. At low frequencies, the high permeability material allows flux to be taken away from a sensitive "area" and return it back to the original source; in effect the fringing that splays out from the "rogue" field is kind of short circuited by a low reluctance path that bridges north and south.

At higher frequencies solid copper (or even silver) conductor sheets become more effective and you see this type of thing in radios - a square shaped can sits over an inductor. Why does mu metal get worse at higher frequencies - mu metal's "effective" permeability reduces with frequency (due to eddy currents increasing) and it becomes less effective as frequency increases. The two effects tend to cancel but, because mu metal is a relatively poor conductor (compared to Cu) it doesn't do the job that a good Cu conductor does at high frequencies - the eddy currents induced in copper are many times that produced in (say for instance) iron or mu metal.

Where do you pitch the protection - both can be the best solution but one may be just as good as both if the frequency is low or high.

Toriod inductors will exhibit less coupling than other types, and as others have pointed out mounting them orthogonally to each other will also limit coupling.

Once you've done those two things you should have very little coupling, but if you need to go further you can add magnetic shielding, separate them further apart, and simply design your circuit differently so coupling isn't a problem.

As you can observe the effect, you can minimise it by rotating each coil until it has the least effect on the other coil. Then fix the coils in that position.