Inductor Coil Winding

Thread Starter

gusmas

Joined Sep 27, 2008
239
Hello

I have a Dc-Dc converter and I need to wind my own inductor. The Inductor requires according to calculations 13.02A (rms value). Now the only wire I can find from a local supplier is rated for 11.5A. Obviously the current my Dc-Dc converter will use is bigger in amplitude than that of the wire. If I apply extra cooling (fan) onto the coil will the coil wire be fine? Obviously this is not ideal but its the only wire I can find at this point in time.
 

Thread Starter

gusmas

Joined Sep 27, 2008
239
You could wind two wires bifilar in parallel to give you the required current rating.
I am using the coil32 program to calculate the amount of turns the inductor should have, can you recommend any other software to calculate the amount of turns required, especially if I am going to use a bifilar configuration?
 

t_n_k

Joined Mar 6, 2009
5,455
There is more information required .......

Are you going to use a ferrite or other magnetic core?

What is the operating frequency?

Any other relevant details ..... such as inductance .
 

Thread Starter

gusmas

Joined Sep 27, 2008
239
There is more information required .......

Are you going to use a ferrite or other magnetic core?

What is the operating frequency?

Any other relevant details ..... such as inductance .
My apologies. Let me fill you in.

it is a sepic converter

Fswitch = 20kHz
Core = Ferrite Core
L1 Inductance: 389uH
L2 Inductance: 280uH
 

Brainbox

Joined Nov 15, 2010
25
I wont bother about the diameter too much.
A wire rated for 10.5A don´t melt at 13 Amps.
It only gives you a slightly more temperature rise.
The total temperature rise is the result of winding method (single layer winding or multilayer windings), inductor losses and ohmic resistance.
Inductor loses are most likely the major contributor of the temperature rise and are depending on the core material.
Stacked windings obstruct a proper head dissipation and results in a high temperature rise in the middle windings.
Ohmic resistance in DC/DC converters is most of the time negligible and only count when You want the maximum efficiency.
 

Thread Starter

gusmas

Joined Sep 27, 2008
239
I wont bother about the diameter too much.
A wire rated for 10.5A don´t melt at 13 Amps.
It only gives you a slightly more temperature rise.
The total temperature rise is the result of winding method (single layer winding or multilayer windings), inductor losses and ohmic resistance.
Inductor loses are most likely the major contributor of the temperature rise and are depending on the core material.
Stacked windings obstruct a proper head dissipation and results in a high temperature rise in the middle windings.
Ohmic resistance in DC/DC converters is most of the time negligible and only count when You want the maximum efficiency.
Awesome i am going to order the wire. is there a rule of thumb of how big the core must be with respect to the wire?
 

t_n_k

Joined Mar 6, 2009
5,455
At 20kHz the skin depth effect will reduce the current rating significantly.

Remember that resistance losses increase with the square of the current.

I would carefully review the losses issue before committing to the final design.
 

Brainbox

Joined Nov 15, 2010
25
There is a free tool at the TDK website named Magnetic Design Tool (MDT)
It calculates core sizes, wire counts, core material and let You "play" with those parameters (based on the TDK core ofcourse)
There is another tool from Ferroxcube on there site which is also very usefull.
 

crutschow

Joined Mar 14, 2008
34,284
Awesome i am going to order the wire. is there a rule of thumb of how big the core must be with respect to the wire?
Basically the core must be big enough to have room for the number of turns required with the chosen wire size. It's rather an iterative calculation, since the size of the core also affects the number of required turns, as determined by the core saturation flux.

So you pick a core and calculate the turns to see if it's large enough. If not, then you pick the next larger core and try again. And, of course, the core size and number of turns also affects the inductor winding resistance.
 

ian field

Joined Oct 27, 2012
6,536
You could wind two wires bifilar in parallel to give you the required current rating.
The more the merrier - more conductors means more surface area, which counteracts skin effect at higher switching frequencies.

A good resource is the Q&A column in Nuts and Volts magazine - they get frequent enquiries on how to wind switcher inductors - the columnist has got trotting out the formulas down to a fine art.
 
Top