Home > Electronics, psu > Power Supply Design Part 1: Unregulated Linear Supplies

Power Supply Design Part 1: Unregulated Linear Supplies

I recently stumbled over a particularly interesting specimen in the family of cheap unregulated power supplies, also lovingly referred to as ‘wallwarts’. Here is the unit in all its prestigious glory:

IMG_20141226_231534

The label seems to claim it’s been certified, but lists no manufacturer or other useful info beyond the useless model number. Inside we find the following:

IMG_20141226_231609

IMG_20141226_231856

What we have here is pretty much the most basic unregulated power supply one can construct, though the bleeder resistor was technically not required. Such luxury. In diagram form we get the following circuit:

psu-linear-unregulated-unsafe
We see the transformer, four diodes (1N4001 or better) forming a bridge rectifier (two extra diodes are cheaper than a center-tapped transformer), the smoothing cap (1,000 uF, 16V) and bleeder resistor (100 Ohm, 1/2W?). 230VAC goes straight into the transformer and is stepped down to the desired voltage.

Now, let’s talk safety. While this circuit will work fine when nothing goes wrong, it is a good idea to consider the two most likely scenarios a circuit like this may encounter in the real world. The first is that of a surge, say from a nearby lightning strike, or an internal short-circuit. The second is when the connected device short-circuits, or its output connector or wires short out. The first scenario results in a massive surge into the adapter, the second will pull more and more power through the circuit until something fails.

With this circuit, the surge or internal short will result in the surge being passed on through the device, into the output and into the connected device. This forms a major electrocution and fire risk. Beyond the circuit failing and cutting off power that way, there are no safety features for this scenario. The same is true for the excessive power draw scenario. Here it’ll keep drawing power until likely something in the circuit blows up, catches on fire or both.

While a transformer in theory electrically isolates a circuit, it has a so-called breakdown voltage at which current will pass straight from the primary into the secondary winding(s), causing a short. During a surge scenario this is likely to happen, depending on the quality of the insulating tape between the windings. One should always consider the scenario where a short forms inside a transformer or related components.

So how to protect against this scenario? There are multiple ways to go about it, but the easiest and cheapest one has to be the humble fuse:

psu-linear-unregulated-fuseWhen the current becomes too much or the voltage too high, the fuse will melt or trip depending on the type of fuse used. One can use thermal fuses if one wants it to be easy to reset: once cooled down they will automatically reset. Regular glass fuses are even cheaper, though probably not as desirable in a closed, maintenance-free unit like a wallwart. There are more options than fuses, of course. One can also look at MOVs, crowbar (zener plus SCR) and clamp (zener plus transistor) overvoltage protection.

At any rate the message should be clear: unregulated linear power supplies are easy and cheap, but one should not skimp on the safeties.

 

Maya

Advertisements
  1. April 6, 2015 at 12:42 AM

    Hi Maya, just a few comments:

    The transformer being intended for mains application may have an internal thermal fuse buried inside with the primary winding. It should also be double-wound with primary and secondary completely isolated from each other by a plastic barrier in addition to the tape and enamel copper wire. Also, the transformer might also not be able to deliver more than a few watts under short circuit conditions. This heat is easily dissipated by the construction and should keep the core well below ~180C at which the enamel breaks down and short circuits the winding.

    Thank you for the great articles covering a wide range of HW and SW topics.
    Marco

    • May 29, 2015 at 8:18 PM

      Thank you. Yes, I did not cover all of the possible situations, nor did I unwrap the coil. I would not be surprise if it had no safeties built-in at all, though. I may do a follow-up article focusing on the functional aspects of this adapter, including maybe some overload tests.

  1. No trackbacks yet.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: