Make Your Own Ferrite to Improve Magnetic Fields




About: I like to improve myself and things I find :) Learning new things every day is next to impossible but I still try - only a working brain can work. I have no special sector to cover, electronics, electrical s...

Update 05/09/2018: I did it again, enjoy!

Update 03/07/2015:
I found the right solution - check the last step!

How did it all start you might wonder, so I will let you know ;)

You might have seen my Instructable for the Simple Induction Heater and with my first one I felt the need to find ways of improving the output power.

The elctronically minded people will already know that ferrite is widely used for small tansformers, choke coils and even antennas.

With this wide variety is is only obvious that you will find a lot of differnt types of ferrite designed for a specific job.

Factors like megnetic flux, core saturation and frequency range are just some of the parameters that can be important for proper performance.

What I use for my needs might not even be considered to be ferrite by a proffesional but it does the job quite good.

This Instructable is a work in progress for me as I will update and complete it with pictures "on the go", so every time I make a new batch or try a new recipe you will see the results here - so keep posted for updates.

If my time permits it I will make the next ferrite during the coming weekend.

It will be plaster based and I shall take some pics and a short video as well.

Please post your comments or improvements and I will update the Instructable accordingly.

Step 1: What Is Ferrite?

Ferrite in the commercial sense is a compressed mix of mainly Iron Oxide and binders.

Depending on the necessary properties, Zink Oxide and even rare earth metals are added.

Usually produced under extreme pressures and heat resulting in a ceramic like finnish.

You can look up on Wikipedia for more detailed info as it is not that relevant for this Instructable.

Step 2: Why Can't I Find Anything to Mould My Own Ferrite?

Quite simple:

There are only a handful manufactorers that have mouldable or machinable ferrite in their catalogue - and they treat their formulas very good!

Also the prices are not really attractive for individuals like you and me.

Even if find a supplier you have to state the purpose and megnetic properties you need or do your own calculations based on the supplied data sheets.

Step 3: Ok, But What Are the Uses of My Homemade Ferrite?

As said in the beginning, I needed something for my induction heater that does not saturate with the extreme magnetic fields involved and also wanted some form of shielding to prevent interference.

With the current mix (at the end of the steps) I accomplished both.

I can't recommend it for the use of specific HF coils or similar as I have no means of checking the properties and magnetic fields with my limited equippment - sorry for that!

But if you make your own electro magnets, induction coils or have the need to generally direct magnetic field lines it might be just what you need.

Also for general shielding of HF frquencies it should work quite well, making it possible to fully seal a circuit in the ferrite so no leaks can happen (or better: should happen).

For example:

A 8mm bolt in my coil with ferrite mantle takes about 90 seconds to get red hot, with a ferrite mantle around the work coil the time is reduced to under 30 seconds.

Step 4: Pro's and Con's

Well, as with all new things they never really perfect ;)

I will start with the negative things first:

* It is very messy to produce, so gloves and outside mixing is highly recomended as you don't want to clean up Iron Oxide dust in your kitchen.

* It is not easy to mix as the Iron Oxide is very fine and tends to ignore all binding at the start of mixing it.

* The mix is not perfect yet, so you might have to test the mix in a small batch to check if it fits your needs.

Now the good bits:

* You can put it into any shape you need.
* It can be sanded or drilled when cured.
* It is a good shield against interference.
* Once mixed it ieasy to handle and form.
* You need no special equippment or expensive ingredients.

* You can easily change the mix to adjust it to your needs.

Step 5: How Is Is Made and What Do I Need?

Tools and things you must have:
* Rubber gloves

* A suitable mixing container - I use the ones to mix plaster

* Spoon, spatula or similar for the mixing

* An area that you clean with a pressure cleaner or that does not mid to get a bit dirty

Things to make it easier:

* Some patience ;)


* Iron Oxide - the blck kind also know as Fe3O4, commonly used to color concrete and quite cheap

* Plaster - Plaster of Paris, wall filler, or some type of resin depending on your needs (this is the binder)

* Water if you use plaster or similar, otherwise check the instruction for your resin

* Zink Oxide and other things you find when searching for Ferrite compositions if you feel the need and get it cheap - totally optional!

How to mix it:

First some explanations;

You want to use as less binder as possible, otherwise the properties of your ferrite might not be as expected.

For high power applications, like an induction coil, you will need a thick layer of ferrite as otherwise the core will saturate or might heat up - it does not harm to use it too thick and you can always add more (around) if saturation is an issue, same for too much as you can sand t down or use a file.

Only mix as much as you can handle within the curing time!

Take a rough estimation of how much in final volume you need for your project and add another 20% be on the save side.
I go for the plaster version as this was my first way of doing it and because it the easiest.

Add the Iron Oxide in your mixing container (put gloves on now if you forgot about them ;) ) followed by about a third in volume of plaster.
Mix well while dry - for bigger batches using a jar with lid saves you a lot of black dust flying around!

Now add the water and keep mixing like you would for mormal plaster so there are no lumps.

Don't worry if you used too much water as you can always add oxide and plaster.

Be aware that this mix dries a bit faster and can produce cracks in thicker layers, so working in small stages with new mixes to build up might be necessary (keep the dry mix and only use with water what you need).

You now can put it into a mould or cover what you need to shield.

Let it set in a moist enviroment to prevent cracks - I simply wrap it in a moist towel (very old one because of the black oxide!).

After about 2 hours you can continue to let it dry in the open.

Final curing time depends on the thickness and humidity!

To for moisture warm it up to about 30° celsius and place into a closed container that was in your freezer to cool down.
If not fully dry you will see a lost of mist condesing on the inside of your container.

Step 6: How to Use It on Moisture Sensitve Stuff or for Sealing in Electronics

There is only one way and this by making sure the moisture won't get to it!

Either seal it first by means of paint, resin and so on, or make a mould for the ferrite and place the part in it once fully cured.

If that is not an option you have to use a castable or mouldable resin instead of the plaster.

For 2K compositions this means you have to add equal amount of Oxide to both components of the mix.

Although Fe3O4 is not that reactive it can affect the curing of 2K resins, so do a small batch for testing first to make sure it cures properly and does not heat up too much during curing.

It is very hard to give proper mixing ratios as the properties of the various resins differ too much but I work my way down from a 50/50 mix until I notice either mixing becomes a problem or the curing is not good enough.

But usually I don't go under mix of 35% resin to 65% of Oxide.

Step 7: How Can You Help?

You can help to perfect the way of making ferrite by posting your feedback and tested recipies.

I will add more pictures with my next batch and add another step for collected mixtures from feedback for an easy reference for everyone here.

Step 8: Update!

I realised by the amounts of hits that I need to rush things a bit to get this instructable complete.
After experimenting with additives to reduce the risk of cracks forming during the drying process I decided to use this variation for a video on how to make the ferrite.

To give you a head start on the new recipe I will sum it up here so you get a clue what happens in the video.

1. Prepare yourself with gloves and everything you need, you will see the things in the video.

2. Don't try to follow the video straight away, watch it at least twice and make use of the pause button if I am too fast in the video.

3. Don't do it inside as the black dust goes everywhere!

4. Ingredients are:

Plaster of paris - or any other modelling plaster you have at hand (Gypsum).

Black Iron Oxide

A bit of wall paper glue or if not available in your area use wood glue - this helps to keep the mix workable for longer and slows down the drying process also the finnished product does not crack as easy.

Water and some tools for the mixing and modelling - whatever works best for you and the amount you make.

5. The glue should be added to the water first and only in small amounts for wall paper glue, wood glue should be added in a volume of about 15% to the water.

I might skip this bit in the video and prepare the water glue mix upfront as it takes a while with wall paper glue.

Make a dry mix of plaster and oxide, I do this with a spoon.

Depending on the mix you will get a pretty good or preety bad result in terms of mechanical stability.

I try to use 3-5 parts of plaster to 4 parts of Oxide, a 50-50 mix seems to work well but I always try to stay under it to have more ferrite than plaster in the mix.

The more oxide you add the better are the magnetic properties of the finnished product but the harder it is to work with and to cure it without cracks.

Add water (already prepared with glue) followed by your dry mix into a suitable mixing container and mix well until there are o lumps, keep the mix thick enough to work with but not so thin it runs off, unless you do a casting.

Once you start mixing the clock is ticking as you only have limited time until the mix goes hard and becomes unusable.

It is best to work with small batches that you can properly apply before the initial drying starts.

Important!! :

You must use clean tools and containers as even small amounts of cured mix will mess up your next batch!

It will cause the mix to go hard much faster and can cause lumps in the mix when it cures faster than the rest.

I try to wash my tools and containers before the plaster goes fully hard.

As you can see in the video I only make a small amount to get the first layer onto the coil.

I do it this way to allow for a better drying as we want to prevent to capture too much moisture before adding the next layer.

You should take your time to let it dry - either naturally or by the use of a de-hydrator.

If there is too much moisture left before you do the final curing in the oven it result in massive cracks.

Best way is to let the finnished product dry out for a few days before putting it into the oven.

If you cover moisture sensitve parts it is best to add a layer of paint as a barrier, for best results give the paint a quick sanding to get a rough surface as this will help the mix to stick.

In case you are in a hurry try the flexible tile adhesive mixes from your hardware store without the glue in the mix.

But do some small scale tests first to check if your mix produces cracks while drying.

Update 03/07/2015:
I experimented a bit more with a lot of different (possible) binders.
Nothing we would find at the local hardware store, the kitchen or local pharmacy (at reasonable prices) worked.
But then it hit me!
Grabbed my last few spoons of black oxide and mixed it with Sodium Silicate - Waterglass.
Of course, me being me, I did not take any pics or videos - shoot me...
Anyway I will try to explain:
Sodium Silicate is another "forgotten" chemical in terms of home use.
Some might still know it from the chemical experiment "Chemical Garden".
In the concentrated liquid form it is somewhere between full cream milk and warm honey in the consistency and glass clear.
Once dried it it goes rock hard - a feature used for repairs on wood, china ware and other things like heat resistant tiles.
If you know "Green Sand Casting" you are already familiar with just adding a tiny bit of water to that mix.
I did the same with the black oxide.
Started with a few tablespoons oxide and added the Sodium Silicate in tiny amounts.
Creates a lot of lumps and small balls, so doing this in a little ball mill might be a good idea (apart from the clean up bit).
Anyway, if you check the videos on grenn sand casting you will see the mix looks almost dry but keeps it shape when pressed - I tried the same but in the end just used a block form and small hammer to compact it.
(This reminds me to mention to get the ferrite mix out of the form after this step - I did not and it was impossible to remove the cured stuff from the form).
After this the testpiece went into the oven for about 90 minutes at full heat - this creates a nice and hard "ferrite".
To get is hard enough to be actually used it is placed into a kiln and is slowly heated to a glowing orange.
After that the cooling was done in the oven, preheated to full.
Oven was turned off once the piece was in and allowed to fully cool down over night.
The result was that
1. I was unable to get the cured ferrite out of the metal box I used.
2. It is so hard that I could not drill into it.
3. It does not break or crack.

I will try to find some more time and black oxide and make a short video of the process.
In the meantime, everyone still following can experiment as the only thing that matters is to just get the oxide moist with the Sodium Silicate so it binds together properly.
During the compacting a bit of excess might be pressed out indicating to use even less sodium silicate for the next mix.
The only downside is that you have to make a small furnace, metal melter or kiln so you can fully harden the mix, which is basically like a ceramic once cured.

Step 9: Update! Maybe the Last One Ever....

I finally made some really decent progress :)
My followers here certainly experimented a lot too but it took my about 2 years to finally get hit by the fence post :)

Let's sum what we still know first:
Iron oxide is good enough for a lot of basic core materials.
A lot of other mineral and metal powders can be added for performance and we are lucky that we can look most them up on Google and Wikipedia.
Making a form is not too hard either...
But so far we struggled to create something really durable.
Not anymore....

Proper ferrite cores are baked ceramic, sadly most of us won't have the equippment, not to mention the knowledge for this.
A suitable binder would be ideal and I think I found one.
How about using something that can be dissolved in water and that dries like glass?
Sodium Silicate :)
You can make it with lye from crystal cat litter but also order it off Ebay or simply buy concrete sealer from your hardware store.
Depending if you use the dry form or get it as a liquid the preperations are slightly different.
Here is the basic way of doing it:

First check how the liquid dries.
So either mix some powder with water or spread your sealer out directly with a brush on some carboard.
You will notice that depending on the concentration the liquid dries differently.
Too much and you end up with a nice looking crystal growth.
Too little and you just get a slightly shiny cover on the cardboard.
Get it just right and you will end with a hard coat, like varnish.
Applying several coats on the previously dried one should build up a glass like cover without and crystals growing or white spots forming.
This is the concentration you want to use for binding the ferrite mix.

Preparing the ferrite is mainly reduced to making sure it is as fine as possible and lump free.
You can put it through a fine sieve if you like but the powder from the hardware or art store is usually good enough.
The one massive problem with iron oxide in powder form is that the particles usually refuse to get wet or accept any "glue" coating".
The simple solution for this problem gave me many headaches...
Well, until I realised how simple it can be done in reality if you stop thinking too complicated LOL
Take a preferably no longer used container or just a soda bottle and fill the amount of ferrite mix you need in it.
Make sure that you can properly close and seal it! You will see why I prefer soda bottles here...
You should never fill to more than 1/4....
Now add some water, preferably destilled or at least demineralised water, please don't use plain tap water.
You will see that nothing really mixes, so add enough water to allow a good shaking.
Ok, I did not say to shake it but now that you did you see it still sucks as a result....
Add a few drops of dishwashing liquid and try again ;)
If in doubt add a teaspoon worth or two.
Suddenly the shaking creates a slurry or if you used to much water a black liquid.
But no more lumps floating around, all the powder is now really wet.

Drying the ferrite...
You want to remove the excess soapy water, so best to let the bottle rest over night.
Now most of the water can be tipped out.
I tried to get the wet ferrite mix out by several ways and apart from cutting the bottle vibrations seem to work well.
A vibration motor from an old game controller or a discarded toy of your girlfriend will do wonders to make the slurry run out the bottles neck.
Spread it out on some baking paper with some paper towels around to soak up the water - or do it outside and let the water run off.
The rest can be flattened a bit and if you need to speed things up let it dry in the oven at 70-80°C - don't let it go to boiling point as this will mess with the soap and we need this soap cover!
Once dry break it up and dry again until you can crush it back to a fine powder in a mortar or discarded coffe grinder - be aware of the dust and do this outside!!

Our soapy ferrite mix can now be used with all water based glues and paints, like PVA wood glue or standard paint from the hardware store.
Downside is that all these things tend to shrink quite badly and only provide a more or less rubbery hardness.
Sodium Silicate however dries literally glass hard!
Of course it would be too easy if the rest is straight forward :(
The big problem we face is also the main feature of sodium silicate: it seals really good.
So making a ferrite rod would be as easy as mixing the ferrite with just enough sodium silicate mix to create a putty like substance, cut the head of syringe off and fill it tightly.
Once pressed out you have a rod you can let dry.
But it will only dry for a few mm from the outside at best, the inside will then be sealed and stay wet forever.
Well, at least many months and than crack the whole thing slowly...
Just so know my level of frustration: For he last step I wasted about 3kg of iron oxide in an endless amount of small batches and experiments...

I found two ways they show at least a feasable aproach to the problem.
Number one: Build up of material.
To prevent moisture being trapped when drying or curing it can help to waste a lot.
Let me explain...
Assuming you want to make a toroid like core of more than 5cm in diameter.
Here I would create a carboard template, just a ring and the two ring walls.
Outer and inner diameter should be just over/under what you need if dimensions are really vital.
Fill a thin layer on the bottom and dry it in the oven at under 90°C.
Add another and do the same.
For this to work the mix should be as dry as possible, look up how to prepare "green sand" and you know what I mean.
Should be possible to form a ball in your hand and if you try to break it open you should get two pieces with a clear breakline.
Be aware the mix dries in air too, so keep it covered a mix it every now and then!
Once you filled enough start to smooth the surface with the last one or two fillings.
Let it all dry again for a few hours in the oven.
Then crank the heat to 180°C for two hours.
This should result in a really hard and durable core that you can now sand down or file down to remove the cardboard.
Using a silicone mold certainly helps if you want to make more than one core.

Number two: Risk it all...
For smaller cores or anything flat enough you can try this way.
Make a mold from thick acrylic, like at least 4mm plate material.
Do it so you have notches to put it all together as you really don't want to glue it together.
The bottom plate should be drilled with lots of tiny holes.
I used a stainless filter mesh as a template and drilled through the grid holes.
You don't want holes bigger than 1mm, preferably under if you can use a laser cutter to create your mold.
These holes are vital to drain the liquid while keeping the mix inside the mold.
Cut some filter paper to cover the bottom and sides of your mold.
For the bottom you want two layers.
With a suitable press lid test if you can really fit it inside without any gaps - it should be as tight as possible without jamming up.
Time for a testrun:
With the filter paper in place and the mold on a surface that allow you drian fill your mold with some lemon or orange pieces - peel first please...
If you press hard enough the juice should run out the drain holes, a bit of juice around the press lid too but no pulp or anything coming out of the top.
In this case you can clean your mess and be happy.
If a lot of material came out the gap around the press lid you need to make a new one and try again.
Once dry and clean add again the filer papers and then fill with your ferrite/sodium silicate mix.
Again make sure it is like good green sand, not lumpy, not runny.
Use a dowel or similar to tap and press the mix for a basic compression and to make sure all coreners a compressed identical - again like making a grees sand cast.
Once full enough add the lid and provide as much pressure as possible.
I usually start by just adding weights on the top until the worst run off is over.
Than depending on size I used a piece of wood and hammer or a vice to press out what I can.
Sometimes it can help to wipe the mold with a cloth that had some oil on it, but most basic forms come apart really easy without sticking.
Take great care removing the core from the mold and don't even try to remove the paper.
The key is now to force the remaining water out without allowing the outside to dry, so if in doub have a spray bottle with destilled water ready to prevent the paper from drying - you need to keep the outside moist until all is done!
For a small core the oven is the prefed option but with the core on a little stand in jar with a bit of water at the bottom.
The steam fill prevent the outside of the core to form a glass layer and the heat will drive ost of the water out of the inner core.
If you have something high like a spaghetti jar or can then elevating the core and using more water is really helpful.
As a number to go from I will give you and example that worked quite well for me:
Core was about 8mm thick and like a toroid with an outer diamter of 9cm.
I used 250ml of water with the core almost 20cm high up in a spaghetti jar.
Like this there was only little room left to the "ceiling" of my oven.
Heat of about 90-120° (my oven sucks down in this range) was applied only from the bottom.
The core got a final spray right before placing the jar into the preheated oven.
After about 10 minutes I opened the oven for a first check on the water level and to let some steam out.
Once all was evaporated I left the oven at the same temp for another 2 hours.
Without allowing the core or oven to cool down much the core was then removed from the jar and placed on an upside down stainless steel sieve.
Temp was increased to 200°C for another 2 hours.
The resulting core had only slight crystal formations on the outside that were easily removed by some fine sand paper.
I could not break it by hand and a drop from about 50cm onto a tile did no visable damge either, even after trying it multiple times.
However a slight hit with a hammer when standing up had no trouble cracking it.
At least the cracks showed the core was fully cured.
Considering the tiny amount of binder in the form of sodium silicate left I call this recipe final for now.

A few words of wisdom:
Although the resulting product is quite durable sodium silicate will dissolve in water and other chemicals!
So a proper coat of durable acrylic paint or a resin coat is a must have if you want to prevent premature failure.
But I think even with that it still is a very neat way of creating a more durable homemade ferrite core.

Possible ways to further improve the curing and hardness....
Sodium silicate has many interesting properties and that it cures like real glass is a nice bonus.
Finding the right oxides, metal powders or even salts to get a ferrite to work as desired requires a lot of digging and experimenting.
Best option to start is by searching for the compositions listed for well know cores.
You might not get the exact ratios but a good understanding what materials work best for a given frequency and saturation.
So again I leave the experiments with this to my trusted followers and hope they will pick up where I tried to provide the start points.
A Rodin coil with a custom ferrite toroid maybe? ;)
Sodium silicate forms a really hard crystal structure with cement.
Adding only a tiny amount of dust fine portland cement and even less lye resulted in a sodium silicate solution that create even more durable coatings.
Problem is finding the right balance to prevent unwanted crystal bloom while curing and to prevent the cement powder from settling down to quickly.
In tests ended with a really thick and really corrosive sodium silicate solution a small amount of cement and quite a bit of lye to get it all more in the PH neutral range.
Results showed great potential but I do not like the requirement for full body and face protection and the risks that come with handling corrosive solutions and mixing them.

Other great uses I found while experimenting....
An almost forgotten fact is that sodium silicate was and still is widely used to seal cracked chimneys.
So if you noticed a problem in your old brick fire place in your favourite hide out cabin then make a mix of cement with about 50% of the water replaced by cement sealer and your next winter will see the smoke go out the chimney intead through cracks inside your cabin.
Although I used this to repair cracks in my crappy homemade foundry...
As you now know sodium silicate forms a glass like surface when cured.
I did a little experiment with a piece of wood.
Using a vacuum chamber I fully soaked it with concrete sealer.
Placed it in the oven to fully dry (I stopped when the scale showed no further weight reduction).
Then I sanded down the massive layer of crystals on the outside.
The thing was not only rock hard but also looked really great in areas I tried to polish.
Water and most chemicals did not affect it in any way.
And well, trying to burn it in my drum did not work either.
It cracked, it charred but it was like putting a brick in the fire...
But the best thing I tried was doing the impossible.
Ever tried to repair a cracked piece of china ware?
You know the feeling you get right now...
Yes, we all dropped something made from ceramic that caused a lot of troubles with a loved one...
I misused an old dinner plate that was already the last one left of the set.
Ooops, dropped it on the tiles....
Well, no I did not try to get all the tiny pieces together, only what was required to get plate back togehter without holes or missing parts.
Should have just tried a tap a hammer to get two or three pieces but I got that though once I thought dropping is the best way...
But it showed that if you waste enough time aligning it all on a plate in your oven you can bake the plate back togehter piece by piece.
Some final coats on both sides to get a smotth finnish and fill missing small pieces and the plate enven survives the dishwasher...



    • Pets Challenge

      Pets Challenge
    • Colors of the Rainbow Contest

      Colors of the Rainbow Contest
    • Sensors Contest

      Sensors Contest

    37 Discussions


    20 days ago on Step 9

    Hi dear friend
    Good time
    have a nice day
    I want to make a ferrite model
    Thanks to your training .... Thank you for that
    This ferrite is a full tube with a diameter of 6 mm, 8 mm and 10 mm
    Used in the gas industry's droplet industry (Igniter)
    In this case, I wanted to give you the necessary guidance and advice
    So I can make it
    Thanks for helping you, my dear friend
    Contact my email so I can send you a photo of the imported model and how to use it
    Thank you very much for your kindness
    Thank you for sure to reply to my letter
    Thank you


    Question 1 year ago on Step 7

    I made a short ferrite rod using your recipe. I let it dry a few days but didn't heat it or anything. It was much lighter than the commercial ferrite rod on which it was based. I put a enamelled copper wire coil on both rods and only got 18uH for the one I made versus 200uH for the commercial rod. What can I do to get a comparable inductance for my diy rods. Do I need to compress them or add other ingredients perhaps?

    1 answer

    Answer 6 months ago

    It is impossible 200uH but 250,300 or 400uH is very ok with ferrite powdered powder with hammer, adding iron filing with iron and little iron oxide, and binder + lacquer + hardener + thinner, these are used for varnishing of cars or furniture etc.

    11 months ago

    Bismuth oxide, iron oxide and zinc oxide might work in the right proportions, as bismuth has interesting magnetic properties.

    I have some sodium silicate here so might try the second formula mentioned.

    Denis AugustoL

    2 years ago

    I believe if I just fill a thin plastic tube with iron oxide I will have a nice core, without needing to use any resin, won´t I? Do you see any problem with that?

    3 replies
    spark masterDenis AugustoL

    Reply 1 year ago

    That is exactly what I am going to try! some commenter say it will work others say nope. Won't know until you try it for your application!


    Downunder35mDenis AugustoL

    Reply 2 years ago

    If things would be that easy...
    Problem is density and with the oxide it really needs to be as fine as possible.
    You might get away for some time if you add a tiny bit of water with gum arabic as a binder and compacting it really hard.
    Main problem is movement once a strong magnetic field is applied, that is why proper cores are usually like ceramic.

    Denis AugustoLDownunder35m

    Reply 2 years ago

    Makes a lot of sense...didn´t think about the movement of the particles...:)


    4 years ago on Step 8

    Looks like I'm the first commenter. I am wanting to use this material to make Toroids for Joule Thieves and also to wind Rodin Coils. Now Rodin Coils are presently wound on Air Cores but they have a lot of properties not yet discovered and work needs to be done on ferrite toroids. It is just that Toroides of the Size I am talking off become rather expensive. This looks to be a possible alternative to the financial disaster independent research already is.

    5 replies

    For project like yours I recommend to experiment with resin instead of plaster or similar as a binder.

    For Joule Thiefs I actually prefer to use ferrite beads as they can be found in old printers and amplifiers.

    An basically every motor inside you will find small ferrite toroids perfect for a Joule Thief, check my Forever Light for some examples.
    But back to your big size ferrite cores...

    If you get a 50-50 resin it would be perfect for the mix but usually you will only find the stuff (at reasonable costs) that needs only a few drops of hardener/activator.

    In any case you want to add as much black iron oxide as possible, so you need a resin with a quite long curing time - don't go for anything with working time under one hour!!

    I experimented on a small scale with 2K glue and had quite good results, so it might works fine for your smaller projects too.

    Be aware that you won't be able to get same same density as commercial ferrite meterials which and compressed and cured by heat that turn the binder in ceramic.

    So where you would usually use an air gap to prevent core saturation you will most likely see that it is not necessary as the home made ferrite simply can't reach the flux densities.

    Another option for a quick check, especially for rodin coils would be to fill a plastic doughnut just with the Iron oxide powder - far from perfect but would show you if there is a noticable difference in flux and magnetic field properties.

    As a "last resort" option for a very quick mix you can also use tree resin, like from pine trees.

    Collect what you can find and try to get the dry stuff not the honey like goo that is fresh.

    Measure by volume how much ferrite you would need.

    Heat the resin slowly until it melts and add a little amount of powdered charcoal - this makes your binder.

    Add the ferrite slowly making sue you keep enough heat and stir well.

    When you notice you can't bind more ferrite add a bit more resin - as I said you want as little binder as possible.

    Once you got enough use it while it is still hot, once it cools it will be rock hard!

    For this to work properly on a doughnut you will need to make a plaster cast for two halfs - top and bottom.

    Pre-heat the cast and fill with your resin mix, scrape of any excess while still hot.

    Do the same for the other half but leave the excess and press the two parts together while the second cast is still hot - if you can keep both hot to assitst the binding process.

    Once cooled down sand it clean and try it out.

    I am actually considering using a device similar to a centrifical extractor to build the toroids and the material will be dried with heat and under centrifical force. This should remove all possibility of cracking. As to the binder I am actually considering Epson Salt as it can be melted at low temperature and will mix readily with the Ferrite oxide. Once hard it is like rock and will conduct electricity as well as the ferrite so there should be no interference with the flow of the created fields.


    Reply 1 year ago

    If it conducts electricity, then you'll have issues with eddy currents.


    1 year ago

    Very Good Research.... I am trying to make Big toroidal transformer for metal melting furnace with gouging rod. Using sodium silicate as binder, can you please guide me through the procedure of heating (temperature ?) and releasing from the mold(steel,plastic?),

    1 reply

    Reply 1 year ago

    It is best to do some testing with fine sand first to save you the mess and money for the oxide ;)
    I started by dissolving the silicate in hot water.
    What no longer dissolved was filtered out with a coffee filter.
    Don't use boiling hot water as the conentration can goo too high and crystals will already form on the water when it cools - if it happens add a little bit of water and mix until the crystal disappear.
    Now add the sand/oxide into your form - a silicone mould works great as it is easy to clean and flexible, which is a big plus when you want to get the thing out.
    To help with the wetting you can add a drop of dishwashing liquid to the water/silicate solution.
    When adding the solution to the mould it should go through the sand/ferrite, if in doubt help by mixing it with a toothpick r stirring rod.
    With very fine oxide it might be necessary to mix it all in a container first until you see all oxide is wet.
    For this it is best to add the solution in very small amounts and to break down lumps when mixing.
    The curing is done in two steps.
    Or three if you have too much time ;)
    First air dry, but you can skip this if you got the excess solution off the surface.
    Second oven dry at below 90° Celsius to dry the mix and prevent boiling.
    This will take a bit, especially if the form is quite deep.
    Again it pays off to start with sand first and to use it for reference.
    E.g.: If the sand mix was fully dry and hard after 3 hours you be safe and add another 30 minutes to dry your ferrite.
    Last and worst step:
    With a silicone form you can crank the temp to just under 200° C to start the actually curing of the silicate.
    But it should not be required as the water trapped in the crystal structure will slowly evaporate at temperatures of just around 120° C.
    Going higher can help forming crystal structures with a higher density and strenght but at the risk of breaking the crystal by boiling the water out of them.
    Try and test what works best for your core but don't go over 150° C until the core is dry enough.

    The resulting core should be fairly stable but is hygroscopic, so youneed to add a coat of paint.
    For obvious reasons it is best to compact the mix once fully wet, you can use a wooden dowel, spoon handle or whatever suits here.
    A loose mix will be very porous and unstable when handled plus of course of very poor magnectic properties.
    You want just enough binder in a highly compressed mix to keep it all together.
    For really demanding jobs, like when winding wire on the core without a bobbin is required:
    Glass can help a lot for ring cores if you know how to cut glass tubes or bottles.
    Downside is the thickness and pain of making glass rings, so only really usable for massive sized cores or those that won't rquire a coil wound on them.
    I found that a wrap with thin strips of fine fibreglass and coating it with resin gives the best compormise for protection and added strength.


    2 years ago

    hi there, nice post!. im thinking of a toroidal choke (like 10 cm radius x 3cm depth) to use it with my welder ( plan to add rectifier) as lc circuit. I was thinking about polystyrene and acetone liquid (method 4), would u guys have any advice would it be better to use plaster?? thanks in advance

    1 reply

    Reply 2 years ago

    Nice timing, I am in the need of something big in a few weeks too.
    Unless your welder is an inverter type it might be the easiest to use the core from an old microwave transformer as the choke.
    Works only for the primary though dow the wire sizes.
    As for plaster I am only happy with it for high frequency use and where big coils are involved.
    Simple reason: You need too much plaster to get really high gains in inductivity plus it only works for materials that won't react with water.
    Iron dust for example will rust before curing unless you have the extreme expensive micro encapsuled stuff.
    Polysterene and acetone works quite good but is hard to fully cure in a timely way and without too much deforemation on bigger cores.

    Here is how I currently handle bigger things that require half decent qualities:
    Fine iron powder is moisturised to slightly rust - you want to mix and air frequently until you get enven rust build up without the product turning into rust.
    Now all is treated with phosphoric acid to convert the red iron oxide into the black oxide type.
    This will act as an insulator is important to get right.
    If you have too much iron with no rust you won't be too happy!
    A quick wash and even quicker hot air dry is required for the next step and should be done before the acid dries on the product.
    Mix now about 25% by volume with black iron oxide, the finer the powder the better.
    As the iron powder is quite heavy you are best off using a ball mill without balls for the mixing.
    Now for the hard part:
    Prepare some fibreglass resin by adding MEK or Acetone so the resin is less thick.
    It should flow like warm honey and you should be able to see a drop on a flat surface to grow in size and become flat.
    I prefer MEK (PVC pipe primer) over acetone as I can smell it better to confirm the core is fully cured - plus it is the base product to make the primer for the resin.
    Your mixed powder should now be wetted with some MEK, just enough to make clump like green sand for casting.
    If you can form it and breaks almost clean it is good.
    Be quick now as the stuff evaporates fast!
    Add the powder mix into a suitable container, icecreame boxes or jogurt tubs work great as the resin won't bind with it (easy cleanup once cured).
    Add the activator to your resin.
    For this you want to use the recommended amount for the volume you had before thinning it plus about 20% extra.
    Mix the powder and resin as good as you can - what you mess up here will mess with you later!
    You want to add resin until the mix is plyable without looking dry or creating puddles of resin on the top.
    Think of it like green sand that you want to make a mould with.
    When you tap it to compress a slight amount of resin should be coming out to have it perfect.
    For abovious reasons you casting form should be a bit higher than required ;)
    Put the mix into the form until filled about half way.
    Tap and compress the mix to drive the resin out then add more to fill it up.
    Compress again until you feel the bottom building up.
    I found the thick end of chopsticks work great for this.
    Your form should be about 10% higher than what you need and be filled right up.
    This allows to scrape the resin off so that when done compressing you only have the mix in the form without an additional layer of just resin.
    Once the leftover resin is rock hard it is time to sand the top off and if you want to cover the finnished product with some acrylic paint.
    Make sure to warm up so you can check if the MEK is really gone - if you can still smell it leave a few more days to fully evaporate otherwise your paint will never cure ;)
    Only problem with this method is stability.
    The iron oxide fills the voids between the iron particles and the resin only the gaps left over.
    Having some dry oxide in the mix will certainly turn bad here as the resin won't bind to it.
    If you have access to a 3D printer you can create the mold from ABS which you seal ABS juice.
    That way the stuff binds very well to the mold and you get added insulation and strength.
    Feel free to also print a ring to cover the top once fully cured.


    2 years ago

    Not really too much time at the moment due to work commitments but I thought I post some info on stuff I tested recently.

    1. Flexible ferrite...
    In some cases it is simply not feasable to create ceramics as it involves temps too high for a normal oven.
    There are other options I wil mention further down though....
    From my experience a pure iron oxide ferrite here will be too weak for most applications.
    For example an increase from 1.4mH with an air core to just 2.2mH with a flexible iron oxide core simple does not justify the work.
    An even mix of black ironoxide, NiZn (Nickel-Zink) and MnZn (Manganese-Zink) gives far better results.
    Binder is mostly silicone caulk (50% by volume of the overall mix) and gum arabic (dissolved in water to give 2% by volume) - the rest is ferrite powders.
    Of course you want them as fine as possible - ball mill comes to mind here...
    To get the silicone liquid enough for proper mixing I added an equal amount of Methyl Ethyl Ketone and used a spoon to squish and mix until there were no more lumps.
    Lighter fluid seems to work too but MEK is available for cheap in the plumbng section as the primer for PVC glue.
    Once done here add the powder(s) and mix properly until you have an even mix.
    If you want faster setting times then now add one or two teaspoons of corn starch to the prepared gum arabic and water mix.
    Fill your form and let rest for a few days so all can cure properly.
    The result is still quite weak in comparision to real ferrite but good enough for shielding (as a paing when still liquid) or for some experiments with magnetic fields.
    Works great to "freeze" magnetic field lines into some rubbery fun project too.

    2. "Cold sintered" ferrite...
    I needed a way to least keep my ferrite together until I had it all installed in the bobbins or coil enclosures.
    But anything clay related or even ceramic meant high temp ovens and press forms - too much for now.
    So how do you get the small particles to bind?
    Except for iron powder Sodium Silicate worked quite well for me.
    Downside is still that you need to create a from and use some pressure to get the excess water content out but at least it does not involve tons of pressure like for sintering.
    Key is to create an over-saturated solution of the silicate.
    This means while the water is close to boiling point you disslove silicate until nothing dissolves any further.
    You can check if it is enough by taking the heat off, with the cooling of the water crystals will form on the surface.
    Re-heat to dissolve them as otherwise you have problems with the next step.
    Get a second, heat proof pot or similar and add the amount of ferrite powder required for your form.
    Now add just enough of the hot silicate mix to get a quite wet mix but not so much that it all ends up swimming.
    Fill your form with the mix before it cools down , for bigger forms it pays off to pre-heat the ferrite powder in the oven a bit.
    Press out the excess water content - for this you need a quite tight fit.
    Use a compressor to blow off the excess (outside please!!!) or some paper towels.
    Place form (if possible with the cmapls and all attached) in the oven at 120°C for at least two hours.
    Let cool in the closed oven.
    The thing will still be wet on the inside and very fragile, so it pays off to prepare the form with some release agent like silicone grease.
    Once out of the form heat again for sevearl hours but below 80°C as otherwise the crystals form too fast and the form crumbles apart - if have the time let air dry for 2 weeks or so.
    You check with a digital scale - note the initial and following weights, once there is no more weight loss it is dry.

    3. The quick and dirty one.
    If you need something simple like a ferrite rod you can create a suitable paper roll template and fill it with a mix of molten wax and ferrite powder.
    The mix should be in such a way that when you let it rest for a moment little to no wax should form a puddle on the top.
    If you still have dry, crumbling lumps you need more wax though.
    Fill in your paper form and let cool down - make sure to have some cm in additional length as the wax will shrink in the center.
    Once cooled down cut to the prefered length and remove the excess paper.
    Be aware that due to the nature of the binder this is nothing for permanent use unless in constand cold enviroments.

    4. Plastic ferrite....
    I tried glue sticks in the beginning and let me just say it is a huge mess!!!
    The powders clump and won't really mix properly, the heat control is a nightmare at the best of times and the mess it makes....
    I needed something easy to make, easy to handle and that is at least strong enough to hold the ferrite together.
    If you have a 3D printer you know what I mean, if not: It is ABS plastic dissolved in acetone.
    Natural, colorless ABS works best as some colors are added as pigments rather than dyes.
    Use an old jar with a properly fitting lid and you can prepare quite a bit and keep it stored in a cool place.
    I make my mix so that is almost like honey in consistency.
    The ferrte poweder should be wetted with acetone prior to adding - this will help preventing lumps and makes the mixing far easier.
    The working time you have depends on the temperature and wether or not you can close your mix airtight.
    This is also the downside of this method.
    Bigger forms can take weeks to fully cure and harden as there is little to no vaopration happening once the outside is hardened.
    Instead the remiaing acetone works it way out by sweeping through.
    Like in a cell membrane it will try to equalise and so the entire thing is weak until fully cured.
    Even if it feel already hard on the outside the inside can be still soft.
    A good way to check is to put the core into a closed yar and to warm it up to about 40°C for about 2 hours or more if you want.
    When you open the yar and smell acetone it is not fully cured.

    I hope this gives you some more ideas to tinker and discuss here ;)
    Thanks for following!


    2 years ago

    Awesome! As someone who spends a large percentage of their time on electronics projects, I often find I need more ferrite than I can afford. The torroids required for the winding of inductors and transformers are actually very expensive, especially for an individual on a strict budget. This is a great solution, as the materials required are inexpensive and the mix can be molded in ways that you could never find pre-made.

    I have not yet made any of this ferrite, but I do have plans to create a giant homemade flyback transformer (the ones I ripped from two virtually identical televisions sucked). Alas, I have browsed the online stores time and time again, and every single time I look for a giant flyback transformer core nothing pops up. So, I guess my two paragraphs of commenting are really asking if I can use a custom mold to create a giant transformer core. I know I can definitely mold the stuff into the shape I want, but will it work for such a high-voltage application?