multiple secondaries on a toroid

[awashi]

not talking about 2x30v or so but more like 2x30V + 2x25V + 2x18V + 2x12V
would pulling current from more than one secondary at once change the voltage i get from them?
how would i have to wind it? stacking the windings problably isn't a very good idea, can i use the whole toroid to wind cable around? do the dual secondaries have to be on opposite positions?

thanks!

 

[richie00boy]

With each winding whatever it's type you should aim to evenly cover/spread out over the core.

The regulation is a function of load, so if it's fully loaded then you stop drawing on one winding, the other windings will go up a little bit.

Remeber you should stay within the core VA limit.

 

[Eva]

The core doesn't directly impose a VA limit, thermal capability of the windings is actually the limiting factor for power output. However, the maximum amound of magnet wire that may be used is limited by core size and shape.

These are some useful recommendations:

- Spread each winding around the entire toroid, this improves coupling and reduces leakage inductance. This recommendation may be omitted for low VA windings like the ones used to generate +15V/-15V for op-amps

- Use bifilar windings to obtain symmetrical +/- output voltages in order to prevent transformer saturation at high loads. This may be omitted for low VA windings

- Place the higher VA windings closer to the internal primary. If several high VA windings are required, then it's a good idea to split the primary in two equal parts and to sandwitch the secondaries betweem the two primary halves (internal half of the primary + secondaries + external half of the primary).

For your winding requirements I recommend tapping, this dramatically reduces the required amount of magnet wire. I suggest arranging it in a way such that an entire lap around the toroid produces 6 or 12V. Using two toroids may make things easier.

 

[AndrewT]

Hi Eva,
your opening statement "core doesn't directly impose a VA limit" kind of implies that the core does have some effect on the VA rating.
Could you explain further?
I do not agree that only the windings control the VA limit.

 

[mzzj]

"core doesn't directly impose a VA limit" kind of implies that the core does have some effect on the VA rating.
Could you explain further?
I do not agree that only the windings control the VA limit.

Transformer core loss is not related to output power, but copper loss is directly related. Core loss stays just the same in transformer whether it is 0% load or 110% load. So the limiting factor is how thick copper you can wind around toroid(or what is practical)

In most? commercial products copper/steel ratio is balanced for minimum cost, ie you can choose large core wich needs lots of turns, or you can choose big core with lesser turns and thus less copper.

In SMPS you can draw huge powers from silly small toroids, D=30mm toroid is capable of 1kW out if water cooled

 

[richie00boy]

But surely the core material determines the mangetic flux capability, hence why cheap cores can mean that two identically sized and wound toroids can have very different VA ratings.

SMPS transformers are much smaller because they operate at much higher frequencies.

 

[Eva]

The magnetic flux induced in the core is not dependent on load current, it only depends on the volts/(turn*Hz) product (input voltage, turn count and frequency).

In other words, the energy transfer process between windings, also known as magnetic coupling, is completely independent of the core. It happens also without core in the same exact way.

However, a core is required in order to add some magnetizing inductance and keep primary impedance high at low frequencies. Otherwise, leakage current flowing through the primary would be much bigger than load current.

For example: Try winding a bifilar air cored inductor with two identical windings, you will discover that both windings are coupled, and that energy transfer is possible with no core at all. The bad news are that the primary is an air-cored inductor, and shows low impedance and big size.

 

[AndrewT]

Hi,
are you saying that a smaller iron core could be rewound with thicker primary wire then rewind the secondary again with thicker wire and you end up with a higher VA rating than the smaller core used to have?
What would happen to the VA rating if you only beefed up the secondary? Primary overheating?

ps. I have an air-cored transformer driving my piezo tweeters but it comes after the 6kHz passive xover.

 

[mzzj]

Hi,
are you saying that a smaller iron core could be rewound with thicker primary wire then rewind the secondary again with thicker wire and you end up with a higher VA rating than the smaller core used to have?
What would happen to the VA rating if you only beefed up the secondary? Primary overheating?

Yeap, you end up with higher VA rating. Double your wire area and VA rating apprx. doubles.

With beefed up secondary you can squeese a bit higher VA rating, because your secondary is now dissipating less than before so primary can run with little higher dissipation. And if you squeese too hard you end up with fried trafo.

 

[Eva]

AndrewT:

Yes, that's the point. However, if you take a toroid and try to do 700 turns or so by hand with a very thick magnet wire, you will discover how exhausting and frustrating is such a task.

Obviously, manufacturers use huge iron cores in order to obtain higher VA ratings with less coper and less winding costs. Remember that automated toroid winding processes have its limitations, the central hole of the toroid must end free, without substantial filling. Note also that iron is much easier to find in Nature than copper, being the latter a semi-precious metal.

 

[djQUAN]

since we're on the topic on tranny VA rating, I bought a toroid used as an autotranny (rated as 1kVA but will be 500VA with isolated secondaries?) which just has two 110V windings in series, I added dual secondaries of 33VAC with #15 which is much thicker than what most 500VA's use. (I've seen most use #16?)

does that mean I could get a bit more power from that than my expected 500VA?

 

[Eva]

Your VA rating would probably be somewhere between 500VA and 1000VA.

 

[djQUAN]

I just measured the winding resistances. I got 0.9 ohm on each 110V primary and 0.1 ohm on each 33V secondary.

I think that's good?