Building Your Own Flyback Transformer

Have you put a scope on the primary to see if it is indeed toggling?

I assume your output is from the rectifier diode and filter cap on the secondary and that one side of the secondary is common to the meter?

Sounds like it ought to have some output if you connected it properly and it is toggling. The toroid material may not be ideal for the frequency you are using, but there still should be some output.

Duty cycle problem?

I have no scope; I thought of using a sound card mic input + resistor, but the frequency's too high.

I'll cut the frequency down by a factor of 10 (to 7 kHz) and listen for a hum

Yep

What's the ideal frequency for this toroid material? Is it really made of ferrite? I'm guessing it was from a noise suppression choke inductor...

Michael

Say... when I wound the secondary, I wound in exactly the same direction, overlapping the primary.

Was I supposed to wind the secondary in the opposite direction?

Michael

Are you running this single-ended? What provision have you made for the half-cycle when the transistor is off - where does the flux go? IOW, how are you resetting the core to prevent it walking away into saturation-land? ;-)

Good Luck! Rich

The flux hits the diode, can't *quite* break it, then thinks, "Hmm, maybe I'd better wait here for a few microseconds until the next pulse arrives."

:-)

I'm a newbie at flyback transformers. (That's why I'm testing mine at low voltage, lest I unwittingly emit subatomic particles in my living room.) Is there a better way to do this, than just have one ultrafast diode + filter cap in the secondary?

Michael

Would use of an ultrafast bridge rectifier avoid the free, all- expenses paid admission to Saturation-Land?

Michael

Well for starters, for flyback you tend to need a core with an air gap !

Go to powerint.com and read all their app notes.

Graham

Read some app notes. Google will help.

Graham

Depends on the material. Anywhere from 20kHz - 1 GHz.

How would we know ? It might be metal powder. Post a pic of it.

Making it completely useless as a power tranformer.

But - for flyback, the biggie is the air gap. That's where the energy is stored.

Graham

Hardly critical if you swap the leads over.

Graham

There is no industry wide standard for color code of toroids.

Your core could be solid ferrite (high permeability, low saturation amp turns) or any one of a number of powdered iron formulas. Based on the heavy wire originally on the core, I suspect it is powdered iron with lots of air gap distributed all around the core between the iron grains.

As I said, different manufacturers use different color codes for their materials. For example:

A powdered iron core could work for a flyback design if the permeability were high enough. This doesn't explain your zero output, though, unless you have a very low permeability core.

Flyback xfmrs need an airgap, you toroid probably doesn't have one.

It matters not a whit. Only time you care is when the phasing is critical to the operation - like the winding is a feedback to your oscillator circuit to keep it running.

Then you just swap the "start" with the "finish" to change the phase - no need to switch winding direction.

Without a scope you're handicapped. Oops, visually challenged. How do you know you are driving the toroid with a signal? You can do things like monitor current or use capacitive coupling and a diode to rectify the AC so your meter can read it, but a scope is lots better.

You also mentioned you only have 3 volts on the oscillator - you have to drive the switching transistor on the toroid with enough voltage and current to get it to saturate - you only need .7 volts with a bipolar transistor, but with enough current to switch. 70 KHZ might be too high for the circuit. Tip31 only has a gain of 25 at one amp collector current.

One easy circuit that almost always works is a blocking oscillator. One transistor with a tapped primary winding. Phase matters - the feedback winding has to be out of phase with the collector.

See

It isn't a high power circuit - but it can be - lower the bias resistance and use a cap around the resistor (drives it into conduction faster and harder). Start with the circuit shown until you get it working then experiment. The frequency will be dependent on a lot of factors. In the 50's we used them to drive loudspeakers and varied the frequency by putting the electrolytic feedback caps in series with the bias resistor and using variable resistors to change the pitch - I made a bike horn out of it and used it to look at resonance in mercury vials.

I wind the toroids with a doubled over piece of magnet wire (bifilar winding) then connect the start to the finish on one coil - this way I may still miscount the turns but the center tap will always be in the center. With ~15 turns of wire and no load on the circuit - the output can be as much as high as 90-100 - and it will fry a 2N4401 most of the time - so keep the led in the circuit; your TIP31 is only good for 40 volts.

no need to remove the wire you already have on, just wind over it

Not at all, they can work quite well _if_ you keep any DC components out of the flux.

Yep. Best would be to drive this core push-pull and capacitively coupled.

Michael, for heavier loads that can be done using a motor driver chip as they can be found in printers, disk drives and such. Make sure the rectification doesn't introduce DC flux and that there is a wee choke right between rectifier and first capacitor to even out the load on the driver.

It does not seem worthwhile to spend time and energy to do experiments on unknown circuit components, especially without the necessary equipment, like a scope or an LCR meter. Either of these can be obtained very inexpensively, especially if you are willing to get an older model and perhaps do a little repair work (which can be educational as well).

You can purchase toroidal cores with full specifications from DigiKey for a few dollars, and then you can wind transformers that will perform predictably. You can also probably get samples on-line.

For the test equipment and the cores, eBay can be your friend, and also check out any local Hamfests that have electronic flea markets.

I would also suggest drawing up your circuit in LTspice, although transformer modeling can be tricky. You could start with a sample circuit for a flyback converter (5v 12v 20mA 1316.asc). It has a transformer model that will work for what you are trying to do. It has a 40 uH primary and a

Paul

Measuring where core saturation occurs is actually possible without a scope. Wind on three windings, 10 turns each or so. Run one to the soundcard output, one to the input. Start some level meter SW, FFT, or similar. Send an increasing amount of DC current through the third winding and see when the audio signal begins to drop off.

I'd have to look it up; I'd probably put "flyback tutorial" or something in google search.

Good Luck! Rich

On Aug 27, 6:15 pm, Joerg wrote: .....

Thanks to all for the replies.

Turns out my astable multivibrator wasn't multivibrating at all - using a 555, I got it to work, kinda. My DMM measured 5VAC across the secondary. Strange, since this is about what I fed it with on the input (6VDC, switched, minus the drop from the TIP31A) - but it could be the high frequency confused my DMM. I was out of alligator clips, else I would have added the ultrafast diode and cap on the output - I measured it raw...

More study needed, when I have more spare time... it's the kids' bedtime now...

Michael

Just don't expect much out of a TIP31. The 555 doesn't have enough muscle to drive it properly. So wear goggles, seriously, because with stuff like this parts can go, in chemistry speak, "exotherm".