A little guide to scavenging toroids, inductors, transformers, and magnet wire.

[Evil Lurker]

I thought I would do a little guide geared towards identifying and scavenging the magnetic components commonly found in electronics. This is because eventually if a person tinkers with electronics long enough to start doing a bit of prototyping... whether it be for a switching converter or joule thief, they will eventually need some magnetic components. The good news is that it is easy as heck to scavenge toroids and transformers essentially for free from any well stocked junk box. The hard part most of the time is simply figuring out how to re-purpose whatever it is to what ever it needs to be.

Take your basic toroid for instance. They are in all sorts of things, power supplies, TV inverter boards, even computer motherboards. The thing with toroids though is that they can be made from various formulations of powdered iron, ferrite, phenolic, or some rolled up brittle metal stuff and/or even be encapsulated in plastic. The reason behind all of the different materials is that each core composition has been specifically formulated to operate at maximum efficiency based upon intended application and specific frequency range.

Take these random toroids I have scavenged for example:



Notice how many of the toroids are colored and some are not. Generally a colored toroid is an indication that it is of the powdered iron type. However there is no firm set industry standard color coding for toroid inductors... a prime example is the large grey ferrite toroid on the left whereas the two slightly smaller toroids next to it are actually powdered iron/metal types.

If you are wondering exactly what is the difference between a ferrite toroid and a powdered iron type it more or less has to do with magnetic permeability and energy storage characteristics. Oversimplified, ferrite types are more suited towards high frequency transformer applications where there is very little holdup of energy in between on/off cycles (such as with the joule thief circuit) and for filtering of high frequency EMI. Powdered iron types are more suited towards lower frequency applications... the powdered metal more or less acts as it's own air gap and helps the core resist going into saturation. These are generally used for PSU output filtering and also find use in continuous mode DC boost coverter circuits. In fact just about all the green/blue and yellow/white toroids for the picture were snagged from the secondary side of various ATX PSUs.

The website below has a lot of great information about ferrite/iron properties. Judging by the application where the toroids were found, that website, and several other references I believe one can safely assume any yellow/white colored toroid consists of type-26 material and a green/blue toroid is type 52. The nice thing about knowing what material a toroid consists of is there are online calculators and programs floating about the internet that allow you to enter in the core size, material, and desired inductance and it will crap out an estimated number of turns of wire needed to achieve a particular inductance... and at the same time some will also do the reverse and let you estimate the inductance of an unknown scavenged toroid based on the number of turns it has.

http://users.catchnet.com.au/~rjandu.../ip_mat_1.html

Also a couple other ways to tell the difference between a ferrite toroid and powdered iron type is that they each have slightly different sizes and weights, ferrite being lighter. For a general idea of what is what, going from left to right the big grey toroid is a FT-140A, the next two are T-130's, the pi filter and two toroids below it are T-106's, then you a T-80,T-68,T-50 so on and so forth. The best list I have seen of standard toroid sizes is also at the same website I mentioned earlier.

http://users.catchnet.com.au/~rjandu...ft_phys_1.html
http://users.catchnet.com.au/~rjandu...ip_phys_1.html

Before I wrap the toroid part of my guide up there are just a few more things I feel should be mentioned. On the bottom part of the picture there are three groups of toroids. Going from left to right are a couple of those weird brittle foil plastic encapsulated type toroids. The group next to it consists of toroids wound as common mode chokes. These were snagged from ATX PSUs from the primary side input filtering part. The white plastic encapsulated ferrite ones were specifically snagged from various Bestec ATX-250-xx's and to be honest i can't recollect where the green ones came from, nor do I know what material they are constructed of. The last group are ferrites are a bunch mainly snagged from power cord filters and what not. Again I don't know too much about them and I'm more or less stuck without a LC meter.

Moving on to transformers, wire, and other inductors...



Toroids are wonderfully efficient with low losses. However if you have to wind more than a few turns on one it quickly becomes a royal pain in the ass without a 5 digit price tag automated winding machine like this:

Luckily there are also other types of inductors that can be more easily put to use and more or less just as easily scavenged. This also includes SMPS transformers which basically for all purposes are nothing more than inductors with multiple windings. The absolute dead simplest to work with are the bobbin wound types shown below and including the green/yellow thing in the picture below. These can be found easier than cow piles in a pasture littering the main circuit boards of CRT TV's and computer monitors. Usually the inductance is printed somewhere on the inductor and it can be lowered simply by removing turns or even completely changed by swapping wire sizes and adding even more turns. Converting one of the bobbin type inductors to a flyback transformer is also just as simple as wrapping a few turns of wire around the outside of the coil.

If one wants to really go all out then just snag a transformer. These take quite a bit of work to recycle but as can be seen I have several of them pictured to show it can indeed be done. Best way to do it is to heat them up in the oven to around 300F for the shortest time possible... just long enough to get the glue holding the ferrite cores together weak enough to gently gently gently pry them apart. And yes it has to be done while the transformer is still 300F. Grab it with a heavy towel, grab each half of the ferrite core with your thumb and index finger and gently gently gently torque the halves in opposite directions while simultaneously pulling them apart. It takes a bit of practice, but as can be seen even a small 5vsb transformer can be taken apart without damage (the bobbins on these things are ridiculously fragile... the pin header rows love to snap off with the slightest excess of pressure). Don't worry about heat damage to the transformer core and bobbin... the plastic used on some of these things can take temperatures well in excess of 500F and as long as the ferrite cores are not exposed to temps above their Curie temperature (like 400F or so) its all good.

Its important that the tape and windings also be removed while the transformer is still hot otherwise the tape is a complete and total bitch to get off. I swear it would not surprise me at all if I were to discover that the tape was at one time considered a candidate for use in bulletproof vests... yes its that tough. Like packing tape on steroids. But with the transformer still hot the adhesive is weakened and it comes off without too much hassle. I prefer to use a rubber hose pick or dental explorer type tool to get the tape started and snip off the wires going to the windings right above the header pins. When all the tape/wire has been removed just take a soldering iron to where the wire is wrapped around the pins and rake it off. Again if one tries to do it any other way they risk breaking off a header pin row... and yes I have snapped off quite a few of them before I finally started sticking transformers in the oven. When that happens the bobbin is ruined and the transformer might as well be tossed in the trash.

Also in the picture are two high voltage flyback transformer cores. I just got the square looking open frame one recently, I hope to get it wound pretty soon. It came out of an old circa 1970's television... they basically don't make them like that anymore. The other rectangular looking flyback core came out of a vintage 1990's set. In the middle of the picture there are two plastic looking thingies. These came off the yoke of a CRT I believe they have something to do with the horizontal deflection of the tube or whatever. The only thing I find interesting about these transformers is that they have a threaded ferrite core in the center of them... essentially a variable inductor. Also the square boxy looking things are SMD inductors... I really like these as I can get them easily from junk satellite reciever boxes I get for free at the local satellite installer/seller's business. They are easily adapted for use on a breadboard by soldering on a couple pins. The trick is to use an IDE header off an old junk motherboard... simply bend a couple of the pins and tack them on. The pins are sqare so they don't wobble around and are spaced exactly the same as the holes on a breadboard.

[Evil Lurker]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

And to show what can be done with a scavenged transformer, I took the ferrite core and bobbin off a common mode choke (the type found on OEM Lite-On PSUs in older Dells and HP's), swapped out the old pins for some new IDE header pins in the positions I wanted, wrapped a few turns of wire around the bottom part of the bobbin and voila... instant joule thief. If you want to know why I used that instead of a toroid... well lets say its a hell of a lot easier to put a 300 turn high voltage secondary winding on that than a toroid. And oh yeah, never put a breadboard in the dishwasher on high temperature cycle, they will warp all to hell.

And last but not least, the wire in the background. The three rolls I got are magnet wire salvaged from 2 degaussing wires I got off a couple old CRT TVs. Yup, that big cable looking wire wrapped in a spiral of electrical tape going around the tube is one single long strand of wire. And, since the wire has spent its entire life coated in tape the enamel is for all intents and purposes good as new, even on a 15 year old set. The smaller wires in front of the big spools are various examples of Litz wire. Litz wire is used in high frequency applications to mitigate the "skin effect" or whatever. Of special interest is the small dark looking wad right in the center. This is Litz wire scavenged from a Panasonic inverter microwave. I'm not for sure how many strands it has in it, but its like quadruple insulated... I figure it would be absolutely perfect for use in a high voltage flyback.

Well thats about all I got for right now. I hope to do an update eventually when I get around to acquiring a LC meter and delve more into practical applications of the components I have scavenged so far. In the meantime maybe this will inspire some people to start doing some research on their own. After all, transformers, toroids, inductors, and wire are all over the place and best of all, free for the taking.

[PeteS in CA]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

For some reason, I don't see the pix you apparently included in your post. That said ...

Oversimplified, ferrite types are more suited towards high frequency transformer applications where there is very little holdup of energy in between on/off cycles (such as with the joule thief circuit) and for filtering of high frequency EMI. Powdered iron types are more suited towards lower frequency applications... the powdered metal more or less acts as it's own air gap and helps the core resist going into saturation. These are generally used for PSU output filtering and also find use in continuous mode DC boost coverter circuits. In fact just about all the green/blue and yellow/white toroids for the picture were snagged from the secondary side of various ATX PSUs.

Ferrite cores work best for high frequency transformers, in which energy is transferred straight from the primary to the secondary (secondaries). For lower frequency (i.e. power line frequncies) transformers, steel laminations are used for the transformer core. Powdered iron (and other powdered ferric alloys), as you said, have an intrinsic distributed air gap, in which energy can be stored. Different formulations of powdered iron have different permeabilities and frequency characteristics. The two you described - yellow/white and green/blue (white and blue being just one side of the toroid), Micrometals -26 and -52 materials respectively - have the same permeability, but -26 starts having excessive losses above 50KHz-100KHz, while -52 is good up to 200KHz-250KHz and has lower losses than -26 below 100KHz. IIRC, Micrometals -4 material (yellow/red) is good for higher frequencies.

As you said, there are no industry standards for color-coding. IIRC, the equivalent for -26 material made by Pyroferric is painted grey/white. Yes, they do have doctors for people who remember this. The grey color of the ferrite toroid you mentioned suggests that it may be from Magnetics Inc., and they use the same paint for a wide variety of ferrite materials. The purpose of the paint is to prevent scratches in the magnet wire varnish and to insulate the core, not to identify the material.

Be careful, though, about your assumptions about transformers. The transformers for flyback topology designs are more like inductors, despite their ferrite cores. Flyback transformers have a significant air gap, so that energy is stored during the switch "On" time, and then discharged into the secondary during the switch "Off" time. How can you discern a flyback transformer from one from a forward or half-bridge or full-bridge converter? Well, if there are 2 or 4 switch devices, it is not a flyback. There are single-switch forward converters, however. Second, if you have just one switch device, look on the O/P side. If you have a large O/P inductor, you have a forward converter. If all you have are small ferrite rod inductors, you have a flyback. Third (if you can't look at the populated PCB), forward or bridge converter transformers' primaries typically have inductances of many hundreds of microhenries or a couple of millihenries, while flyback transformers are likely to be less than a couple hundred microhenries.

[PeteS in CA]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

... and to be honest i can't recollect where the green ones came from, nor do I know what material they are constructed of.

If the cores are a fairly dark green color, they may have come from a Delta Products P/S. Common mode chokes tend to be ferrite, high permeability, with effective frequencies in the low hundreds of KHz range.

Also a couple other ways to tell the difference between a ferrite toroid and powdered iron type is that they each have slightly different sizes and weights, ferrite being lighter. For a general idea of what is what, going from left to right the big grey toroid is a FT-140A, the next two are T-130's, the pi filter and two toroids below it are T-106's, then you a T-80,T-68,T-50 so on and so forth. The best list I have seen of standard toroid sizes is also at the same website I mentioned earlier.

Those core size designations are from Micrometals, and you can find dimension info and winding info (inductance per turns-squared, A(L)) on this page, Specifically, the Power Conversion Products sections.

[PeteS in CA]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

... and to be honest i can't recollect where the green ones came from, nor do I know what material they are constructed of.

If the cores are a fairly dark green color, they may have come from a Delta Products P/S. Common mode chokes tend to be ferrite, high permeability, with effective frequencies in the low hundreds of KHz range.

Also a couple other ways to tell the difference between a ferrite toroid and powdered iron type is that they each have slightly different sizes and weights, ferrite being lighter. For a general idea of what is what, going from left to right the big grey toroid is a FT-140A, the next two are T-130's, the pi filter and two toroids below it are T-106's, then you a T-80,T-68,T-50 so on and so forth. The best list I have seen of standard toroid sizes is also at the same website I mentioned earlier.

Those core size designations are from Micrometals, and you can find dimension info and winding info (inductance per turns-squared, A(L)) on this page, Specifically, the Power Conversion Products sections.

[PeteS in CA]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

Going from left to right are a couple of those weird brittle foil plastic encapsulated type toroids.

These sound like tape wound or bobbin wound cores, of which Magnetics Inc. is a manufacturer. Relatively speaking, they are not cheap. I think those may have come from +3.3V regulator circuits. Another core type that might be used in that circuit would be Metglas (to use a particular vendor's trade name), which is a non-crystalline metallic glass-like material (ferrite is crystalline, IIRC).

However if you have to wind more than a few turns on one it quickly becomes a royal pain in the ass ...

Been there, done that, got a closetful of T-shirts. My "favorite" was winding some ten turns of teflon-insulated AWG #10 wire on a large ferrite toroid quadrifilar!

[domas]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

Hi there... bumping an old topic.

I have a corsair vx550w ATX p/s which will be repurposed to adjustable voltage p/s (will be using 41amp 12v rail)

The thing is, the first one I have made few years ago (thanks badcaps) was really a tl494 feedback pin hack and few extra features. I have made another one where I followed a tutorial for making one 13.8V supply and then redoing it to 2,5-25V adjustable on my own. The tutorial helped me to remove all unnecessary components for unused rails. This time I want to do the same.

Lets get back to topic: The 25V supply I have repurposed included rewinding LC1 output filter / toroidal coil, as it had more rails going through there and needed more turns now as both voltage and output amps have changed

As my new corsair p/s will be done similar way (removing everything not needed) am I allowed to trash -12v, 5V wires and anything else I find on that particular toroid? just to keep it dedicated to 12V rail.

To simplify the question: are magnetic properties of coil changed when it is used for few rails at a time vs being used for 1 rail? If yes, do I have to care about that?

[Longbow]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

It is indeed fun to try building various projects with discarded toroids. Some amount of testing is necessary first, however. Even if the original schematic showing a particular inductor is available, the core information will still not be disclosed. You might surmise what the core material is by its use in a circuit, but that will still be insufficient information in most cases. Companies that make the cores have the required technical data showing permeability vs frequency, as well as saturation flux density. With magnetics, you don't get the same inductance at all frequencies, or at all current levels. Thus, the need for testing unless you are just having fun. See this .pdf for some interesting commentary:

[domas]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

there is a special toroid tester which can test its flux, right?
i have inductance meter, that is a good starting point.

Re: my post above, i have just unwound -12v rail and that was it. no more rails were present there apart from the original +12v rail

[Longbow]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

You can get lots of good information by hooking up a known capacitance in series (or parallel) with the inductor. Run a sweep and observe where the resonant point is. Calculate the inductance value, and then a reactance value at the frequency of interest. Divide the reactance by the coil resistance to get a rough approximation of the inductor's Q. That's a lot of information coming from a simple test.

[domas]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

What is "run a sweep"?
Thanks.

[joshnz]

Re: A little guide to scavenging toroids, inductors, transformers, and magnet wire.

What is "run a sweep"?
Thanks.

frequency sweep