Dots on inductors

[Raveninghorde]

I read many years ago (National app note from memory) that one should
connect the inner layer of an inductor to the switch in a buck
regulator and the outer layer to the capacitor as the outer layer will
then act as a screen.

I don't know if it makes a difference in practice but it's free.

With many modern inductors it is impossible to see how they are wound
so the only indication of "polarity" is the dot. Is the dot arbitary
and down to the winder or does it actually relate in anyway to the way
the inductor is wound?

[legg]

It indicates phasing alone. Constructional details, where critical,
will only show on the winding sheet.

You won't get the winding you want without including winding
instructions in the transformer drawing and then batch inspecting each
purchased lot.

RL

[SoothSayer]

The dot is for phasing of a TRANSFORMER. In such a case, it indicates
the winding START point. In many cases, it is already part of the
bobbin. In some cases, it gets painted on as part of the winding process.
There is no need to mark the winding start of an inductor.

An inductor is a non phased device.

No, it matters not which 'way' it gets utilized. AT ALL.

[Tim Wescott]

I don't think that's entirely true.

Which way you use it doesn't change the inductive behavior of the part,
but it does change the parasitic capacitance of the part to neighboring
parts on the board, and/or how it radiates. If the outside windings are
on the "cold" end of the inductor they'll tend to provide some measure of
electrostatic shielding. If they're on the "hot" end they'll tend to
provide a bit of electrostatic coupling to the world at large, much the
same as if you had a similar-sized cylinder of metal attached to that
circuit node.

This may not matter for most practical purposes, but it certainly feels
like something that would have an effect on EMI, depending on how the
circuit is built.

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com

[John Larkin]

On Mon, 26 Nov 2012 15:44:04 -0800, SoothSayer

It can matter, just like the "outside foil" of a film cap (black band)
sometimes matters. There's less capacitive coupling to the outside
world, less radiated EMI, of the outside of the winding is at AC
ground.

Also, sometimes magnetic coupling between inductors matters, like in
filters. Polarity could matter there, too.

Someone used to, maybe still does, make choke inductors that have
internal biasing magnets, to increase saturation current in one
direction.

There are hum-bucking applications, too.


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

[Joerg]

Nope, no standard there. Fastest way to find out is crack one open and
look. Watch out for ferrite shrapnel. The gentle method is to measure
the RF leaking out with one end grounded and the other energized, then
reverse. Then talk to the mfg whether their production procedure (SOP)
calls out to always connect in this particular way. Typically they are
but it can be tough to obtain something in writing about it.

--
Regards, Joerg

http://www.analogconsultants.com/

[Bill Sloman]

On Nov 27, 10:44 am, SoothSayer <SaySo...@TheMonastery.org> wrote:

Wrong. If the outermost layer of a multilayer winding is connected to
ground at one end, the inductor as a whole will radiate appreciably
less than it would if one end of the outermost layer were connected to
a high AC voltage.

It only the stray capacitative coupling that changes, but that can
matter quite a lot.

--
Bill Sloman, Sydney

[Raveninghorde]

On Mon, 26 Nov 2012 17:27:46 -0800, Joerg <inv...@invalid.invalid>
wrote:

I've just checked with a German manufacturer and they confirm the dot
on their inductors is the start of the inner winding.

I appreciate not all manufacturers will necessarily be as consistent.

[Joerg]

Great, then at least you found one where they have a system to it. If
this is for a product design I'd still get that in writing though.

[legg]

On Mon, 26 Nov 2012 15:44:04 -0800, SoothSayer

It is highly unlikely that a dot will occur on a magnetic drawing that
has only one winding. Inductors with multiple, tapped and
phase-coherent windings do exist and are constructed with all of the
precautions required for transformers. The are called inductors simply
as a convention to indicate their intended use.

Every line-powered computer you've ever seen probably has one of the
more common examples; a common-mode choke to assist with EMC
compliance. Such a part becomes unpredictable (if not completely
useless) if phasing or winding instructions are ignored.

There are many intentional constructional details in magnetic
components that are not immediately obvious, even to the experienced
technician, that can determine their effectiveness in many ways.

RL

[Joerg]

> has only one winding. ...


It often does:

http://katalog.we-online.de/kataloge/eisos/media/pdf/744778920.pdf

[...]

[Raveninghorde]

On Tue, 27 Nov 2012 14:06:35 -0800, Joerg <inv...@invalid.invalid>
wrote:

Or

http://www.coilcraft.com/pdf_viewer/showpdf.cfm?f=pdf_store:lps5015.pdf

[Raveninghorde]

On Tue, 27 Nov 2012 10:43:12 -0800, Joerg <inv...@invalid.invalid>

What they said by email was:

The dot signifies the start of the winding on the inductor, thus this
would be the inner layer of the windings which would connect to the
hot side of the output of the DC/DC converter. The outer winding of
the inductor in effect self shields the emi on the inner windings.

[Joerg]

Perfect, then you got it in writing. I thought it was just a phone
conversation across the Channel.

[legg]

Interesting; both of them. At a quick glance, they are unique in the
range of both manufacturers.

For Coilcraft, the 'dot' isn't actually associared with either solder
junction, but appears equidistant between the two......so you'd have
to crack one open to figure out what they mean, and hope that the
meaning isn't subverted in the part's sales lifetime. It's not
guaranteed that the coil winding personnel have even a working
familiarity with the language of the spec......

The instruction 'winding direction' is unfortunately ambiguous in the
English language, as winding is both a noun and a verb. When
indicating prefered wire layout, I always indicate the direction of
the coil former's rotation (ie the verb sense of the winding machine's
function) and illustrate the movement of the physical part, to avoid
this (hopefully....).

RL

[Joerg]

It's not that unique, even in the distant lands of China they do it:

http://xfmrs.com/pdf/2xfs1m.pdf

But it's always good policy to obtain written confirmation about the dot
and winding orientation.

[legg]

On Tue, 27 Nov 2012 14:06:35 -0800, Joerg <inv...@invalid.invalid>
wrote:

and, of course, there was actually a noticeable effect on the EMI
plots, when checked........making the effort worthwhile?

There'a difference between anecdote and application. There's a
difference between a 200W toroid and a shielded bobbin core.

The costs of playing it safe, just in case, can whittle a budget past
the bare bone.

RL

[Joerg]

It already applies at much lower power levels. I finished a flyback
design a few weeks ago. If you get the winding orientation wrong on
those then all hell breaks loose at the EMC chambers.

The word "shielded" in ferrites is often more a marketing term. A closer
looks typically reveals a fairly large potting area. The potting
compound is colored almost the same as the core.

[John Larkin]

On Wed, 28 Nov 2012 15:29:24 -0800, Joerg <inv...@invalid.invalid>

Since some ferrites are conductive, it's not always obvious which
connection will radiate less.


--

John Larkin Highland Technology, Inc

jlarkin at highlandtechnology dot com

http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom laser drivers and controllers

Photonics and fiberoptic TTL data links

VME thermocouple, LVDT, synchro acquisition and simulation

[Joerg]

True, but fairly easy to test when you have a near field probe kit. That
kit is probably the most worn tool out here. Other than the bottle
opener :-)

[legg]

On Wed, 28 Nov 2012 15:29:24 -0800, Joerg <inv...@invalid.invalid>

The function of the rf-conductive core body has also to be taken into
account in the radiation process. Capacitance to the core from
different portions of a winding (or windings) has to be taken into
account, as well as possible core body and screen current paths.

>
>The word "shielded" in ferrites is often more a marketing term. A closer
>looks typically reveals a fairly large potting area. The potting
>compound is colored almost the same as the core.

A shielded bobbin core, on the other hand, specifically indicates the
employment of a sleeve structure with a controlled gap, whether SMD or
through-hole. This controlled gap is where the filled adhesive is
present.

RL