custom ferrite magnetic cores- sources?
Spehro Pefhany
Anyone bought small quantities of custom ferrite cores?
I'm looking for a couple of special shapes, but made from common
material like 3C90 or whatever (it's not very critical).
Size will be fairly large-- several cubic inches of material.
Any idea of sources and ballpark NRE costs?
Tim Wescott
AFAIK ferrite is hard to manufacture (which is why _my_ question was
about iron powder!)
Could they be cut from flat plates? Plates are obtainable, and I'll bet
that they're cutable by laser or water jet or some such.
Spehro Pefhany
wrote:
>On Mon, 23 Nov 2009 14:16:59 -0500, Spehro Pefhany wrote:
>
>> Anyone bought small quantities of custom ferrite cores?
>>
>> I'm looking for a couple of special shapes, but made from common
>> material like 3C90 or whatever (it's not very critical).
>>
>> Size will be fairly large-- several cubic inches of material.
>>
>> Any idea of sources and ballpark NRE costs?
>
>AFAIK ferrite is hard to manufacture (which is why _my_ question was
>about iron powder!)
Yes, now that I read it, your application has some similarities. Large
air gap too.
>Could they be cut from flat plates? Plates are obtainable, and I'll bet
>that they're cutable by laser or water jet or some such.
Water jet uses abrasive that is basically ferrite AFAIUI (garnet) so I
don't know how well it would cut. Maybe my shapes could be cut from
flat plates and stacked up/epoxied together. Maybe. Who sells these
ferrite plates of which you speak?
legg
<spef...@interlogDOTyou.knowwhat> wrote:
>On Mon, 23 Nov 2009 13:37:30 -0600, Tim Wescott <t...@seemywebsite.com>
>wrote:
>
>>On Mon, 23 Nov 2009 14:16:59 -0500, Spehro Pefhany wrote:
>>
>>> Anyone bought small quantities of custom ferrite cores?
>>>
>>> I'm looking for a couple of special shapes, but made from common
>>> material like 3C90 or whatever (it's not very critical).
>>>
>>> Size will be fairly large-- several cubic inches of material.
<snip>
>>Could they be cut from flat plates? Plates are obtainable, and I'll bet
>>that they're cutable by laser or water jet or some such.
>
>Water jet uses abrasive that is basically ferrite AFAIUI (garnet) so I
>don't know how well it would cut. Maybe my shapes could be cut from
>flat plates and stacked up/epoxied together. Maybe. Who sells these
>ferrite plates of which you speak?
>
Planar EI configurations offer plates with large surface areas that
would translate into significant cross-sections, when stacked. The
last ones I bought were polished on both surfaces ~ flat.
EPCOS, Hitachi, NEC/Tokin, TDK and FDK market ferrite polymer
composite materials. This was initially in sheet form, but it is
supposedly suited to injection molding.
The FDK and NEC materials are low temperature and lossy, the EPCOS is
high temperature (200C) and useful in resonant power circuits, though
permeability is low. The TDK markets composite material mainly for
magnetic apps.
http://www.epcos.de/inf/80/ap/e0001000.htm
Tooling for an injection mould can be pretty steep, unless amortized
over large production volumes.
RL
Paul Hovnanian P.E.
IIRC, ferrite cores are made by sintering powdered ferrites. I don't
think they machine them any further. So its probably a matter of having
a mold constructed and specifying the exact mix of powder to acheive the
required magnetic properties.
I wonder if one can obtain the powders, make a ceramic mold and borrow a
small pottery kiln to roll your own.
--
Paul Hovnanian mailto:Pa...@Hovnanian.com
------------------------------------------------------------------
Parity on, dudes!
Tim Wescott
Dangit! I _distinctly_ remember seeing this in the Fair-Rite catalog,
and now it isn't there!
Adrian Jansen
You need to find a supplier for the highly specialised ferrite powder
composition you want. Mostly the making of that is a propietary
process. I doubt any ferrite manufacturer would sell a small quantity.
Then you need a high pressure press capable of about 500 Kg/sq cm and
tool-steel dies to do the powder pressing. Life of a steel die is around
500-2000 cycles. Production dies are spark eroded from tungsten
carbide, expensive.
Firing needs a temp around 1400 celsius, and a controlled atmosphere,
air in first part of cycle, tapering on a strictly controlled curve to
pure nitrogen during cooling after peak temp. This is way beyond what
you can do in a typical pottery kiln.
Cutting shapes from block blanks would be possible using diamond
grinding wheels and drills. Ok for very small qty, but still expensive
for the tooling.
Water cutting was not around when I did this, but maybe with silicon
carbide or diamond grit, it might be possible.
--
Regards,
Adrian Jansen adrianjansen at internode dot on dot net
Note reply address is invalid, convert address above to machine form.
Tim Wescott
I want someone to make ferrite Lego.
Spehro Pefhany
About 7000 PSI.
>Life of a steel die is around
>500-2000 cycles. Production dies are spark eroded from tungsten
>carbide, expensive.
500 cycles is plenty.
>Firing needs a temp around 1400 celsius, and a controlled atmosphere,
>air in first part of cycle, tapering on a strictly controlled curve to
>pure nitrogen during cooling after peak temp. This is way beyond what
>you can do in a typical pottery kiln.
Yup.
>Cutting shapes from block blanks would be possible using diamond
>grinding wheels and drills. Ok for very small qty, but still expensive
>for the tooling.
>
>Water cutting was not around when I did this, but maybe with silicon
>carbide or diamond grit, it might be possible.
Thanks for the detailed info, Adrian. Very useful.
Do you remember about how much the ferrite changes in size after
firing?
Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
sp...@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Adrian Jansen
Typical shrinkage on firing is 20-22%. IE you make the die 20% bigger
than the final size you want. We used to test each batch of powder for
a number called the "Dry Press Density" which was around 2.7 gm/cc when
pressed at 500 Kg/sq cm. Final fired density has to be close to 4.8
gm/cc. You can easily calculate the linear shrinkage from the density
ratio.
The fun part starts when you press a long shape, like the leg of a U
core. Then friction between the die wall and the powder changes the
density along the leg. So the bottom and top get denser, the middle
less dense, leading to a leg tapered outwards at each end, and of course
the bottom of the U ends up denser than the legs, also leading to odd
distortions. You have to size a die to take account of all this.