Modern power supplies tend to be more robust than those of a few years back. Not long ago it was not uncommon to hear of power supply failure being the major cause of equipment malfunction.
During the early years of the Switch Mode Power Supply, current limiting was built into the power supply to merely protect the switching transistor and nothing else. Now, it is almost unheard of for a power supply to fail when operated within the limits as stated on the packaging. Take laptop chargers for instance. Thousands of them are in use at any one time during the day. When last did you know of one failing? I thought so...
But then comes the bigger pieces of equipment. There lies a whole different story. Power supplies still remain the weakest link in the whole chain. Apart from the already covered rectifiers, the next most common failure of power supplies is the components that exist in the high-voltage portion i.e. between the rectifiers and transformer, typically referred to as the "switching circuit".
There is, generally speaking, only one cause of failure.... over voltage. The reason is the running voltage after rectification is pushed beyond the component limits. This problem is not only restricted to active semiconductor components, but passive ones such as resistors too. Normal carbon and metal film resistors are only designed to handle between 250 and 350 volts. It does not take long to see that our previously stated working voltage is already in excess of the maximum found on many resistor ranges.
This is not said with any malice but American designs are notorious for failing when exported to Europe. This possibly comes from a naive thinking that 230VAC is the limit. In the UK this is subject to -6..+10% variation, and often exceeds that! An investigation into the high number of power supplies fitted to a hi-tech piece of equipment failing revealed the devices were being subjected to over 250VAC and a resistor in the power supply could not handle the voltage across it.
But this is not a power quality issue, but rather a design one. Well, if a power supply is marked at 250VAC max, then this is the maximum the manufacturer is expecting to have the power supply run on. With the standard being 230VAC this would mean he has built in a 20V safety margin. In electronics that is a lot. Even kettles are marked as 220-240V which means even they are not expecting the wall socket to stretch to far beyond the 230V mark.
Whether or not this is actually a design problem the fact still remains - high voltage is the primary cause of failure in mains powered switch-mode power supplies. No, we have not contradicted ourselves. It is still the current that 'kills' the components, it was the voltage that caused the current to start flowing in the first place when the breakdown was reached.
But Under-Voltage is also a killer of power supplies. As said earlier, most circuitry can accept when the voltage falls below that where the circuitry can maintain regulation. However, if the circuitry were to be sensitive to this e.g. the biasing being upset and causes high current drains, then the low voltage condition could very well lead to failure in the parts being put under this strain.
Although this condition can exist in switch-mode power supplies, it is far less pronounced and is usually well below possible operating voltage (typical cutoffs are at least 40% of operating, and as low as 85VAC in 110..240V input capable units). However, don't be fooled. A switch-mode power supply generates some very interesting properties when operated at below its cut-off!
In some domestic equipment, which has a narrower voltage operating range, a fall in voltage will lead to a rise in current and may start to strain the switching transistor. This is either through the heat the insufficiently cooled transistor now needs to dissapate, or simply the current approaches the maximum for which the transistor was designed. Either (or both) of these factors can drastically reduce the working life of the part from many years to merely a few weeks, or months at best.
Usually such effects are not too noticed if the voltage dip is short, but there is another effect that can occur if the dip is just long enough! We investigate this next.