Author Options:

need help winding toroidal transformer Answered

I'd like to know how to wind a torodial transformer with a toroid that I just happen to have. I would like it to work at 60Hz at 120 volts (US mains) and have an output of 36volt centertap (18-0-18) or maybe a little less (I'm building an op-amp powersuppply that gets regulated to +-12volts) and as many amps as possible (well, safely). The toroid I have is painted a light green with a blue ring on one side. it is 1.25" outerdiameter, 0.75"innerdiameter. It's 0.5" thick. It looks like the bottom left toroid in the picture below, just bigger and no wire. I do not know what gauges to use, how many windings (well, I know the ratio), how many amps it can deliver, how I should wind it, etc.


At 60 Hz you won't be able to wrap the secondary, so never mind the primary, those toroidal core you have might be good enough to build a Switched mode power supply out of that has 120 volt input and 18-0-18 Volt output but the switching frequency would be high then 60 Hz (( some where in the high kHz are to low MHz area And then you only need a few wraps of wire and you can greatly reduce the size of your magnetics.......

That might be a solution, I'll look into switch mode power supplies, cause I need to get this circuit as small as possible.

Ok so your line powering an op amp, what sort of current are you looking for ? there is lots of wal-wart switched mode supplies and there is even small 15-0-15 ones out there the produce 85 watts in a 3 inch by 2 inch by 3/4 inch package.... But if it's just a small op opamp you only need a few mA's and you could use a online supply which is nothing more then an Zener diode cap and resistors, and very small, I built a 5 volt one that fits inside an extension cord end to light it up with blue LED's....

well, here's the whole situation: I have a op-amp in a wein-bridge setup creating a 10MHz sinewave (the opamp needs +-12volts) I need to take that output and amplify that so it has more current and voltage (maybe a total of 25 watts out?) This amplified output will be driving a tank circuit (I'm trying to make a wireless electricity device) I need a power supply to deleiver all of the energy

How's your radio theory? (mine sucks.) You're essentially building a high power radio transmitter, which has technical, safety, and legal issues well beyond those of normal circuits...

well, the lowest radio waves in use are like 29 MHz, and that's for toys. You know that wireless electricity machine that MIT made? It works by resonance (with a tank circuit) and it operates at about 10MHz. saftey issues, all the power that isn't used get's reassorbed into the tank circuit. legal issue... I don't think that's a problem

Ah, your radio knowlege is worse than mine. A bit of searching on wikipedia shows that AM broadcast radio goes from about 0.5MHz to 1.6MHz, with 10MHz being smack in the middle of the frequency range known as "shortwave." In theory, 10MHz is between "bands", but the "most heavily used band" stops at 9.9MHz...

And the safety issue has to do with some of the radiated energy ending up "getting used" to cook your tissues rather than ending up in the target circuit. See "RF burn"; a things that starts to be of concern in the 100kHz range...

is that how Nikola Tesla died,crouched in a corner afraid of the naysayers and the,the wolf?

And there is octaves, 81 of them, so if your not careful you'll annoy more people then you think, and this can include law enforcement, search and rescue personal, and they have the ability to find you, they have found people using low cost DVD players emitting 121.5 MHz which was an octave of is clock speed, much lower then 121.5 MHz and with no shielding it was picked up by search and rescue..... (( radio it's tons of fun and lots of math and shielding )) And as of Feb 1, 2009, Sarsat will no longer uses 121.5 MHz for emergency beacons

Ever wonder why the FCC approves odd thing's like CFL lights ??

Found the FCC frequency allocations, nice downloadable PDF. Exactly at 10,000KHz (10MHz) is a "STANDARD FREQUENCY AND TIME SIGNAL" for "Space Research." Be careful, hit that exactly and NASA might come after you.

Aim a bit higher, from 10,100 to 10,150KHz is "FIXED" "Amateur" so looks open for playing. But you should check that chart, maybe you can find something with a bigger bandwidth to use.

Is this the MIT thing you're talking about? Found the original paper from the one researcher that showed up in the journal 'Science'. It uses magnetic coils tuned to resonate at 9.9MHz, a matched set. The one is driven at that exact frequency which sends power to the other. Oh, it says they used a Colpitts oscillator.

Are you trying to recreate that? 'Cause otherwise it sounds like you're using radio waves to try to transmit power. There's already a company doing that.

And,? there is a wireless tranmission device here on instructables that use's two coils one large, and one small with 2 .01uf caps on one coil to lite leds ect, it's in the 147.7khz

hmmm... I'll look into that colpitts oscillator, it looks much simpilar and easier to build than a wein-bridge. Thanks for all the links and frequency stuff I'm not trying to transfer power with radiowaves (well, I guess I am) but yeah, I guess I'm basically doing what they did in the article... but let it be known I'm not copying them, I'm copying tesla... lol I'm transfering power though electrical resonance...

Then you may be interested in this guy's site, head to the bottom for "Links" then about 3/5 down you'll find the high-voltage stuff including assorted Tesla things. He's built a lot of things, good directions and pics as well as links to programs that will design asst. coils and things for you.

Heh. I want to build one of those, too. It's not really a radio transmitter, since it uses near-field magnetic fields, as I understand it.

Nothing ventured nothing,absolutly notning gained. Fearnot the legal ranglingsBush?Chenny didn't!

A budding Tesla in our midst? Why am I not suprised. I'm sorry I can't help, but your project sounds very interesting, even if it doesn't work.

An op-amp dealing with 10MHz? I thought op-amps basically didn't work at that sort of frequency, from what I remember from my first year. If your op-amp is putting out the sine wave then I guess that's OK, but that raises the question why you need the op-amp if it isn't going to be amplifying... you might want to look up the frequency-gain curve for your amp before you go any further, IIRC it's a logarithmic decrease down to a gain of 1 'round about 10MHz.

I'm getting a couple of samples from texas instrument that can handle up to 12 MHz. The first Op-amp is used to generat the sinewave using a wein bridge circuit. I then need to beef up that signal so that I can drive the tank circuit.

Aren't there better, dedicated circuits for generating pure sine waves at those frequencies?

The easiest way would be to use a Switched mode supply, there are several ready made supplies out there for laptops etc or you could mod one to give you the required outputs... When you need that much power online supplies are not the best option....

Lol, This is what I've been trying to do, But I just settled for a old Switch Mode Power Supply mounted out of the box...

Ok, new question, same (kinda) topic. Is it OK to put components inside of a toroid (such as bridge rectifiers, capacitors, etc.). I'm asking this because size is a factor.

Depends. Try to keep things "in and out." Like if you had a resistor sticking straight up with the far lead going straight down, a "U" shape, roughly centered in a coil laying flat on the board. Induced currents would cancel out, current path in, same current traveling the other direction going out. A resistor merely going straight thru the hole would get an induced current same as a wire would. Plain diodes and bridge rectifiers should be fine, semiconductors like MOSFETS that get triggered by tiny signals would not. Capacitors are a different issue, as magnetic fields tend to make electrons want to move that'd lower the dielectric strength, the exact value of a capacitor would fluctuate. Of course you can try using a strip of steel inside the toroid to block the magnetism, use the multi-tester and a small coil of wire to see if there's anything noticeable.

thanks That's exactly what I needed to know worse comes to worse I'll just put that band of steel


10 years ago

Just to add to the concerns expressed below, that looks like enamelled wire- do you trust enamel insulation on tight coil wraps with 120V? Also the concerns about number of turns, frequency etc. still apply, you'd probably be better off finding a commercial supply. Just for the record... would terrible things happen if you tried to connect two 12V "wall-warts" in series to get a -12V-0-12V supply, if they were running off the same mains supply? I can't think what would, but I can see weird problems with reference voltage, etc: if one produces 12V relative to earth, connected straight to one outputting 0V relative to earth, would that act as a short circuit and let the magic smoke out?

well, the only problem with walwarts is that they arn't usually regulated, for example, a 9volt walwart can actually put out 18volts with no load.


10 years ago

60hz power transformers are pretty big. (The inductor size for a given power goes down as frequency goes up.) 80mm diameter for a 50VA transformer, 110mm for 300VA. Here's a commercial source; you might be able to do a bit better with something custom-made, but you're not going to be able to do MUCH better ('ye canna change te laws o physics'), and getting a power transformer out of a 1.25inch core is probably not possible (I suspect the required numbers of wire of the required wire gage for the primary simply won't fit!)

It could be possible to make a transformer using ultra thin wire, but winding ultra thin wire on the torrid core may be very difficult or maybe not possible...

And "ultra thin wire" is not compatible with "as many amps as possible"...

Yeah, I know that, you probably get no more than a milliamp out a transformer with ultra thin wire...

ehhhh... I guess I'm not going to get as much power as I was thinking out of it...

thanks everybody, looking into what you all said

Something about this project screams danger -- be careful! As westfw said, if you use the correct gauge, you may not be able to make the coils as small as you want them to be.


10 years ago

Since there's no way to upload files (and I've forgotten where I downloaded this...)

...I've uploaded the best doc I have on toroids to my amp 'ible-- PSU section. It's the "Winding Toroids" PDF at the bottom. Lots of practical info on wiring, as well as the theory.

It's an appropriate addition to the instructable, anyway... good luck.

Dooh. That's so simple. (Bonks head with handy CD case...)

Wikipedia time!!

Read about transformers here. Information about frequency is this part. How many amps a wire can safely conduct is the ampacity of it. Since you're likely using the
American Wire Gauge system you can check the chart there, if using enameled transformer wire you can probably use the 90deg C numbers, for smaller than 18AWG you can extrapolate using the resistance numbers (for use-at-your-own-risk personal stuff, not for something to sell). This will also give you the cross-sectional areas you'll need. It'd help to know exactly what the core is to find the magnetic flux density (B), but it looks like ferrite so you should be able to ballpark your numbers from the permeability.

Then comes math. Glorious wonderful math.