[Lf] [Fwd: LF: toroid cores, from DK 8 ND]

[Andre Kesteloot]

Hans-Albrecht Haffa wrote:

> Dear Rik, look to the attachment.
>
> 55, Hans-Albecht
>
> Hans-Albrecht Haffa
> DK 8 ND
> cba
>
> Goeppingen, August 24th
>
> Dear Lowfers,
>
> ist a really pity that today`s experienced professionals in communication techniques have no knowledge about materials, components and magnetics. The calculus of a transformer on 136 kHz is really the same as on 50 Hz. Having several  years of experience in the development of switch-mode power-supplies I know that it is not possible to give a short summary, as many figures I just know by experience, not by design manuals.
> But I try.
>
> - You can choose every ferrite material with a relative permeability urel between 40 and 2500. If You have a material with a Permeability betwenn 3000  and 6000, reduce the magnetization. I never could measure any differences in materials like Epcos (Siemens) N27, N47, N67, N87 or Vogt FI324, FI323. Do not use Carbonyle iron powder cores neither for transformers nor for coils.
>
> - Choose a magnetization of about 100 mT = 100 Millitesla = 100e-3 Tesla and a current density of 2,5.....15 A/mm2.  15 A/mm2 for small cores, smaller than 20 mm at ist longest side, 2,5 A/mm2 for the big Toroids with 120 mm diameter. You are principally free to use more windings with a smaller core (more copper losses) or to use a larger core with more iron volume.  The chosen starting value of magnetization and current density is not a secret, it is the experience You get in a long time as a professional.
>
> - Choose adequate winding copper material, either round wire, copper-stripes or HF-Litz wire.
>
> - Look to Curves explaning the skin effect, that means that diameter or thickness of the copper is restricted, depending on the frequency used.
>
> - Measure the temperature rise with a thermocouple termometer or Your wet finger at room temperature. If it feels just hot, temperature will be about 60 ø C, so from 25ø C room temperature it will be 35 Kelvin  rise. A cheap add- on Thermometer for a digital multimeter You get from conrad.com in DL for about DM 50,--. If Your transformer becomes too hot, check the copper material or lower the magnetization. E.g. lowering magnetization is very effective. Check the  maximum temperature of your insulation materials, (130øC,  155øC or 180ø C). Self adhesive tape from the stationary shop is not an insulation material !!
>
> - a good book for HF-transformers is:
>  Jerry Sevick, W2FMI "Transmission Line Transformers", American Radio Relay League Inc. Newington
> - or also: Catalog from Amidon Associates
>
> - For Coils use gapped ferrites. Toroid Cores may be gapped with a Saw for tiles (bath-room plastering) in the DIY-market. It is no problem to cut a 10 mm gap into a  Toroid with 62 mm OD. I have already built up a filter here at home (LOC:  JN48UQ) in G”ppingen.
>
> If You mind,
> look to the home page of the following manufacturers of ferrites
>
> Kaschke   DL
> Keramische Werke Hermsdorf/ Tridelta/
> Vogt
> Siemens/ EPCOS
> Philips/ Valvo/ Ferroxcube   all above in DL
> (Ferroxcube also USA)
> ISKRA in Yugoslavia
> Amidon USA (It is not a manufacturer, this is Fair Rite)
> Neosid   DL and UK, UK for Power-Ferrites
> Stewart US
> Fair- Rite US
>
> A good choice may also be toroid cores of amorphous metals from VACUUMSCHMELZE in Hanau, DL or from Micrometals US, but usually they have never samples of theese wonderful metal glasses for development.
> So far in this night.
>
> 55,
> Hans-Albrecht Haffa
> DK 8 ND
> email: nospamleaveitoffbeformynameHaffaHA at t-online.de
>
> IDD:  ++(0)7161 - 87 847
>
> one side
>
> rsgb_lf_group at blacksheep.org schrieb am 23.08.02 16:48:31:
> > Hello LF group,
> >
> > I am trying to make a list of toroid core material that is suitable for
> > 136kHz power applications, for impedance transformers and low-pass filters.
> > I would appreciate any information about the material (brand-type), the
> > application (transformer or filter) and the power level it was used/tested.
> >
> > many thanks in advance,
> >
> > 73, Rik   ON7YD
> >
> >
>