transformer and non-linear load

[csdave]

hi all,
as per my other threads, I am at my first experiences with designing small power supplies for my circuits and I feel I need to understand transformer ratings a little better.

Say I have a transformer that is rated 6V @ 3VA.

My understanding is that if I connect a purely resistive load to this transformer, it will be able to provide up to 500mA(rms) at which value, the voltage will be of about 6V (rms).

With reactive loads, things get more complex, but the key is that the apparent power should not exceed 3VA.

So far so good, right?

Here however, where I start getting confused. With a sinusoidal waveform, apparent power is the product of RMS voltage and RMS current, but what happens when the load is such that the voltage/current waveforms are no longer nicely sine shaped?

Consider the same transformer feeding a full-bridge rectifier with a filter capacitor at its output. What will happen is that the capacitor will draw high bursts of current when it charges. The biggest will be when it charges from the empty state at the beginning. But there will also be bursts at each half-cycle of the initial waveform if the load circuit is drawing enough current to significantly discharge the cap.

- LTSpice allows me to calculate the RMS current required by the circuit, but how should I go about calculating this myself?

- Is it correct to say that once I have the RMS current, then the apparent power is RMS current times the (RMS) Voltage of the transformer?

- Is the RMS current the only thing that can cause problems? Since we no longer have a sine wave, the same RMS current can be obtained with a variety of peak current values. How do I compute the peak current in practice? How large a peak can I assume the transformer to handle?

thanks

Davide

 

[FvM]

- Is it correct to say that once I have the RMS current, then the apparent power is RMS current times the (RMS) Voltage of the transformer?

Yes, I would prefer to say nominal voltage, because the current rating doesn't depend on the actual voltage. 6V @ 3VA simply means, that the current is rated 500 mA rms.

- LTSpice allows me to calculate the RMS current required by the circuit, but how should I go about calculating this myself?

You'll find some rules of thumb that suggest a current form factor for different rectifier topologies, assuming reasonable dimensioning. I remember a factor of 1.6 (Irms/Idc) for full wave rectifiers.

I guess, that your LTSpice simulations give a too high Irms, because you most likely didn't put in realistic values for resistive and inductive transformer impedances. See some representative numbers for small transformers in a previous thread. Transformer impedance is also the reason, that peak currents won't be very high with a real setup.

https://www.edaboard.com/threads/212803/

 

[chuckey]

Its all about conduction angle and ripple. Suppose you have 10% ripple, the the voltage on your reservoir capacitor falls from Vpeak to .9 Vpeak. The current pulse to charge the capacitor runs from .9 Vpeak to Vpeak, during time T. There is then no current flowing for some period, call this X. To find out the true numbers, you have to integrate the funny charging current pulse over the time T knowing that it is equal to load current over T+X. An approximation is (X + T)/T X the output current.
Frank

 

[csdave]

Thanks to both!
But from a design perspective, once I find the actual numbers... what tells me what can be sustained by the transformer? In principle, my 500mA RMS can be achieved with a huge peak over a very short duty cycle.

 

[FvM]

In principle, my 500mA RMS can be achieved with a huge peak over a very short duty cycle.

In principle. For a technical discussion, you should consider the current waveform that can be achieved in a particular circuit. Mostly, the current levels aren't "huge".

 

[csdave]

In principle. For a technical discussion, you should consider the current waveform that can be achieved in a particular circuit. Mostly, the current levels aren't "huge".

mmm and how do I determine if a transformer can provide the current waveform I need then?

 

[FvM]

The answer has been discussed in detail for the rectifier/filter capacitor problem. Generally speaking, you have to analyze the circuit.

 

[csdave]

mmm not sure I get it. Where has it been discussed? I am interested in learning more about this, but I am not sure where to start. Do you have any pointers?

thanks

 

[FvM]

I mean the edaboard thread linked in post #2.

 

[csdave]

I mean the edaboard thread linked in post #2.

I thought you might mean that, but I don't remember discussing at what peak-height the waveform may cause problems. I am not saying that the problem will happen in my circuit, but I am interested in knowing

 

[FvM]

at what peak-height the waveform may cause problems

Current RMS is the only criterion, there's no such thing as a critical peak current. The main point is to understand howver, that peak current will be effectively limited by the transformer impedance.