Toroidal transformer secondary winding overhead

[made2hack]

Hello all,

I am winding an isolation transformer (230V / 230V). I would like to know what are the considerations / calculations for the overhead on the secondary winding.

In other words, how high do I want the No-Load voltage to be? 240V? 250V? I want the secondary winding to be limited to around 2 Amps max out.

And do I simply multiply the Turns per Volt number by an extra 10% ?

Do you need any other information? Let me know.

Thanks,

[bktemp]

The  overhead voltage if necessary for compensating the voltage drop of the transformer due to the copper resistance and leakage inductance.
So, you need to calculate/estimate the voltage drop at full load and then increase the number of turns to compensate for that.

[made2hack]

Would you happen to know how to calculate that?

I mean I get the copper resistance, how do i calculate the inductance leakage?

[Zero999]

The usual way is to wind it to give the rated voltage, under full load. Unfortunately, that makes it output a much higher voltage, when unloaded. This was not acceptable for my application, so when I did this, I designed the transformer to output slightly under the rated voltage, when fully loaded.

I ignored the hysteresis and eddy losses (I believe this appear as a resistor in parallel with the primary, so don't cause the voltage to drop much, only increase the current draw) and only took account for the copper losses, which is just the resistance of the windings. In an isolation transformer (1:1 ratio), the copper loss looks like a resistor with a value equal to both the primary and secondary winding resistances in series.

What's the VA rating of the transformer? One thing you can do is look at the datasheets from transformers of a similar size.

[MrAl]

Would you happen to know how to calculate that?

I mean I get the copper resistance, how do i calculate the inductance leakage?

Hi,

Probably the best way is to wind the two windings at the same time as one bifilar winding.  You can then test it and add turns as needed.  The leakage inductance is not easy to calculate so a test will be a good idea, even if you did calculate it.  The resistance of course is not too hard to estimate.

This is if you dont need the highest class of galvanic isolation.  If you do need that, then you have to do one layer at a time with plenty of tape between or separate the windings entirely, which leads to more leakage inductance unfortunately.

[bktemp]

For a typical toroidal transformer the leakage inductance is insignificant, so I would ignore it and only use the copper resistance for the calculation.

[MrAl]

For a typical toroidal transformer the leakage inductance is insignificant, so I would ignore it and only use the copper resistance for the calculation.

Hi,

A test is always a good idea however, especially if the windings are separated entirely.

[bktemp]

For a typical toroidal transformer the leakage inductance is insignificant, so I would ignore it and only use the copper resistance for the calculation.

Hi,

A test is always a good idea however, especially if the windings are separated entirely.

I did many measurements and the additional impedance of the output due to the leakage inductance was almost unmeasurable compared to the copper resistance.
I have never seen completely seperated windings on a toroidal transformer except for some very special types. They are always wound on top of each other.
For a conventional transformer leakage inductance is much higher, especially when both windings are wound side by side instead of on top of each other.

[Zero999]

Probably the best way is to wind the two windings at the same time as one bifilar winding.  You can then test it and add turns as needed.

*snip*

This is if you dont need the highest class of galvanic isolation.

Unless you use properly rated mains cable for the bifilar winding (not just ordinary magnet wire) then you won't get any proper galvanic isolation.

For a typical toroidal transformer the leakage inductance is insignificant, so I would ignore it and only use the copper resistance for the calculation.

Hi,

A test is always a good idea however, especially if the windings are separated entirely.

I did many measurements and the additional impedance of the output due to the leakage inductance was almost unmeasurable compared to the copper resistance.
I have never seen completely seperated windings on a toroidal transformer except for some very special types. They are always wound on top of each other.
For a conventional transformer leakage inductance is much higher, especially when both windings are wound side by side instead of on top of each other.

Yes, even for a traditional E-core transformer with the windings on separate bobbins, the voltage drop due to leakage inductance, is still much lower than that caused by the resistance.

[MagicSmoker]

Would you happen to know how to calculate that?

I mean I get the copper resistance, how do i calculate the inductance leakage?

Leakage inductance is strictly a function of winding geometry and while there are equations to estimate it, it usually isn't worth calculating, especially for a toroid, which has the lowest leakage of any shape.

For a practical example, I pulled a 150W 120V/60Hz toroidal mains transformer out of my junk pile which has a magnetizing inductance of 3.3H and a leakage inductance of 1.3mH. This amount of leakage inductance presents an AC reactance of 0.5 Ohms, which at the full load primary current of 1.25A results in an additional voltage drop of 0.625V. Meanwhile, the actual primary winding resistance is 5.6 Ohm, which results in about 7V of drop. The leakage inductance is on the better side of average for a mains frequency toroid while the winding resistance is a bit worse than average; usually total regulation is around 3-5% (ie - voltage drops 3-5% from no load to full load).

[Kleinstein]

If in doubt, one could have separate windings for the compensation. So one could choose between an exact 1:1 and compensated for nominal load and maybe even a little high voltage or intentionally lower voltage.

Leakage induction is usually relatively small.