Short circuiting a power supply vs. transformer

Question

When you short circuit a transformer, it heats up and might blow up due to the high current flowing. In lab a transformer or actually multiple transformers got broken this way, one ended up with a hole in it. The teacher explained that it should have better protection inside, like a fuse.

I just wondered, why is it then safe to short circuit a power supply. We use a device that is a voltage source and also a current source and when using it as a current source or setting the current to certain limit, we have to short circuit the supply first. Why is that then safe in comparison to the transformer? Has it only to do with the supply having proper protection build in it compared to the transformers used or is there another reason why it is safe to short circuit a source / power supply and not safe to short circuit a transformer?

Answer 1

Has it only to do with the supply having proper protection build in it compared to the transformers used or is there another reason why it is safe to short circuit a source / power supply and not safe to short circuit a transformer?

Yes, a bench power supply (and even a lot of in-line power supplies) are far more sophisticated than a basic transformer. They can be sophisticated enough to self-protect without destruction.

Answer 2

Power supplies are today almost always switched mode power supplies. That's because it's the only possible way to create a power supply with stable voltage as a function of load, low cost, low weight, high efficiency, perfect power factor (sinusoidal current input) and ripple-free output.

Old transformer-based power supplies almost always had huge variation of voltage as a function of load, and if they didn't they just used a linear regulator to dump the excess voltage as heat, becoming extremely inefficient. They weighed a lot for a very low current capability. The power factor was crap due to large capacitor current surges, and if they didn't have a linear regulator then the capacitor ripple was huge.

A switched mode power supply converting AC to low-voltage DC can be for example forward converter or flyback converter.

They create the required voltage by rapidly switching current through an inductor via a MOSFET.

In a forward converter, current through the output inductor is going through the load. At any time, the controller has the option to also supply more current to the inductor via a high-frequency switched-DC transformer.

In a flyback converter, current is alternately flowing from input to inductor (energizing it), or from inductor to load (de-energizing it). The inductor has two isolated windings.

By adjusting the duty cycle, it's possible to adjust the current and voltage through the load. This is the basis of voltage and current limits at most lab power supplies.

To limit voltage, you need to measure it -- nothing else, the rest is in the control algorithm. Same for current, to limit it, you need to measure it, and the rest is in the control algorithm.

It's probably very hard to find a switched mode power supply with absolutely no form of overcurrent protection. It's so cheap to add there that it's almost always there.

Also, do note that short circuiting a SMPS lab power supply isn't advisable. For example, I have a 30V 10A power supply. I have tested short-circuiting it at voltages of below 10V, and at few amperes of current. It creates a magnificent spark show when you remove the short-circuited leads. Not only that, but the metal in the short-circuited leads is damaged, because essentially what you're doing is using the lab power supply as a welding power supply.

I suspect that at 30 volts, and 10 amperes, the spark show would be so huge that you have to wear eye protection or else some of the sparks can fly into your eye.

You don't want to destroy your power supply leads by creating spot welds into them! I have, and won't do it again.

Of course by short circuiting before turning the AC side on, adjusting the current, and then turning the AC side off without removing the leads, and only then removing the leads you can eliminate the spot welds and the spark show.

Answer 3

Most bench power supply used in labs have some kind of protection circuit in the form of a current limiter. That is when the output is shorted or just overloaded, the current won't reach levels that can damage the power supply, but the current will be limited to some set level which is somewhat higher than the maximum current the power supply can provide in normal operation.

There are more advanced protection circuits that can reduce the output current wrt the maximum allowable in case of gross overload (these are called protection circuits with current foldback).

This is true also for power supplies that cannot work as current sources.

In addition, power supplies intended to be used not only as voltage sources, but also as current sources have a current limiter whose threshold can be set by the user. In some purely analog supplies this can be done by shorting the output and setting the desired current level using the appropriate knob, usually while watching the current put out by the supply on the built-in ammeter of the instrument.

Modern digitally-controlled bench power supplies, be they linear (not very efficient and heavy, but very quiet) or switch-mode (extremely efficient and lightweight, but producing a noisy output), have the capability to set the current limit using a keyboard or a digital encoder knob on their control panel without shorting the output.

Anyway bench power supplies have multiple levels of protection against abuse. In addition to the current limiter, there is most often an internal fuse on the mains line that protects against gross failures of the internal circuitry (e.g. an internal fault that causes an anomalous current absorption from the mains that can start a fire otherwise). Moreover over-temperature protection is quite common, even in very old analog supplies.