Anonymous
11/8/2025, 1:05:47 AM
No.2956048
>>2955845
You’ll want to understand what kind of course you have first. There’s a lot of different core materials for different uses, to the point it may not even be worth trying to characterise a mystery core. But if you salvaged it from a circuit where it was outputting no less power at no lower a frequency, it’s probably fine. You’ll still want to characterise the inductance per square-turns ratio, probably using an oscilloscope and some known-value low-ESR capacitors. X2 film caps work great.
As for winding them, you should calculate your desired inductance, and figure out the number of windings from the aforementioned ratio. Then use your calculated rms current to determine the power dissipation with varying wire gauges, figuring out the length of each winding from the geometry of the toroid and the radius of the wire. If you pick wire too thick, it will be difficult to wind, so sometimes people use multiple thinner wires in parallel, or even flat wires. It may still be hard to fit through the core if there isn’t much room. Wire too thin will get too hot, you can estimate the temperature by estimating the heat dissipation of your inductor to ambient as compared to similarly sized power resistors. There are various enamel coatings for the wire with differing maximum temperatures, but your core may also stop operating properly at elevated temperatures, so beware.
Then with a selected wire size and number of windings, you can spool out the required length of wire and start winding. The full length needs to go all the way through the middle of the core for each winding, assuming you don’t have a bobbin of wire that can fit through the core. It may be easier to wind multiple smaller windings and put them in series.
>>2956035
At least it stopped the DRC bitching at you. Unless it says the component value next to each footprint on the silkscreen, I look at the board layout when populating the board regardless. I guess I could look at a BOM.
You’ll want to understand what kind of course you have first. There’s a lot of different core materials for different uses, to the point it may not even be worth trying to characterise a mystery core. But if you salvaged it from a circuit where it was outputting no less power at no lower a frequency, it’s probably fine. You’ll still want to characterise the inductance per square-turns ratio, probably using an oscilloscope and some known-value low-ESR capacitors. X2 film caps work great.
As for winding them, you should calculate your desired inductance, and figure out the number of windings from the aforementioned ratio. Then use your calculated rms current to determine the power dissipation with varying wire gauges, figuring out the length of each winding from the geometry of the toroid and the radius of the wire. If you pick wire too thick, it will be difficult to wind, so sometimes people use multiple thinner wires in parallel, or even flat wires. It may still be hard to fit through the core if there isn’t much room. Wire too thin will get too hot, you can estimate the temperature by estimating the heat dissipation of your inductor to ambient as compared to similarly sized power resistors. There are various enamel coatings for the wire with differing maximum temperatures, but your core may also stop operating properly at elevated temperatures, so beware.
Then with a selected wire size and number of windings, you can spool out the required length of wire and start winding. The full length needs to go all the way through the middle of the core for each winding, assuming you don’t have a bobbin of wire that can fit through the core. It may be easier to wind multiple smaller windings and put them in series.
>>2956035
At least it stopped the DRC bitching at you. Unless it says the component value next to each footprint on the silkscreen, I look at the board layout when populating the board regardless. I guess I could look at a BOM.