Ferrite material with square shapped BH curve

[MichaelPI]

Hello,

does anyone know a supplier of ferrite toroids using material that has a square shapped hysteresis loop? 

I would like to do some experiments with magnetic memories. I do not really want to buy an entire ring core memory from ebay for a lot of money for some experiments, that could be broken.

Unfortunately, Ferroxcube (before Philips) discontinued the 3R1 material last year. There seems to be the possibility to order 3R1 toroids from Farnell, but the leadtime is approx. 30 weeks (which is not a problem, if it is possible to get it at all), but Farnell does not ship to end customers in Germany. So I do not know, if this is a viable option at all.

Ring cores from Vacuumschmelze (not ferrite, hence pretty expensive) are also difficult to obtain with a Z characteristic.

Maybe someone knows other brands, that still manufacture material with this characteristic? Suggestions are welcome.

Thanks.

Regards
Michael

[uer166]

Big or small? Toshiba MS is what I experimented with way back https://www.digikey.com/en/products/detail/toshiba-semiconductor-and-storage/MS-21X14X4-5-W/4701157?s=N4IgTCBcDaICoHsDOALAlgIwIYAICyAyiALoC%2BQA

But it seems you need really small cores, not large sensor cores so idk if that fits.. If you have real volumes I can recommend some suppliers, but you won't find anything in distribution and it will be custom-ish.

[TimNJ]

Basically same answer as above, but generally speaking: Toshiba AMOBEAD amorphous cores or Hitachi METGLAS amorphous cores.

www.toshiba-tmat.co.jp/pdf/en/product/3-1_am_parts_absse.pdf

There are lots of other manufacturers in China and Taiwan these days, but you might have trouble sourcing them.

[MichaelPI]

I am looking for small toroids - small cross-section and magnetic path length to reduce inductance and required saturation current.
Thank you both for the hint. I will look for Toshiba / Hitachi cores at mouser and digikey.

[Terry Bites]

I'd be scouring the Ferroxcube website. Not.
But see this- they that the pn for the cores is in the eagle files. Not my thing but https://corememoryshield.com/report.pdf

[MichaelPI]

Unfortunately, Eagle is not installed and Kicad does not want to open the files. There is only this description in the BOM: 32x ~1mm hard ferrite cores (with wide hysteresis loop)

I opened the files with the editor to see if there is any unencrypted information hidden, but there was nothing like a PN for the used cores.

So far the best option seems to be the AMOBEAD. They are small and relatively cheap - AB3X2X3W 10x approx. 5.5 € at Digikey.
Should saturate already at approx. 40 A/m - at this size one turn and a current of approx. 350 mA should be sufficient to drive the bead.


 

[Terry Bites]

altium viewer?

[Haenk]

https://www.ebay.de/itm/234348014188

1000pcs for a tenner, including shipping.
As this has been intended for core memory, it should already have the correct characteristics.

(I'm not the seller or in any way affiliated with the seller...)

[MichaelPI]

No, sorry, I don't have Altium Viewer.

Thanks, I have seen the 1000 cores for 10 € - seems like a rather good offer. But most probably I will go ahead with the AMOBEAD, I also find their primary usage for EMC reduction interesting - honestly speaking I have never heard of this type of bead (seems like low frequency (?) noise suppression). I have some (maybe not really logical) issue with stuff that is somehow not specified (no datasheet, unknown part number), even though those cores will almost certainly perform well as core memory, I would like to have at least an idea about the shape of the BH curve. How much current I have to provide and if the assumptions I have made are true.

[Haenk]

No, sorry, I don't have Altium Viewer.

Thanks, I have seen the 1000 cores for 10 € - seems like a rather good offer. But most probably I will go ahead with the AMOBEAD, I also find their primary usage for EMC reduction interesting - honestly speaking I have never heard of this type of bead (seems like low frequency (?) noise suppression). I have some (maybe not really logical) issue with stuff that is somehow not specified (no datasheet, unknown part number), even though those cores will almost certainly perform well as core memory, I would like to have at least an idea about the shape of the BH curve. How much current I have to provide and if the assumptions I have made are true.

They were made for commercial core memory, so I can only assume they do exactly what they were designed for.
Btw. I have also seen core memory driver boards, those might ease the start.
(I will have a look tomorrow if I can find some driver schematics in my service manuals...)

[jonpaul]

Easier to just buy an old core memory you can find   in 1960s..1970s office calculators

J

[daniel10351035]

Not sure if you're still looking...but W701( VITROVAC 6025 F) from VAC might work for you. I was looking for replacements for AMOBEADS and this passed all my tests. I also happened to know some Chinese suppliers who can make this

[wspitts2]

Would you mind saying who the supplier in China is? Thanks!