Hi guys,
Just a couple notes here about standard rectifier diodes in rectifier applications...
Rectifier circuits are rather unusual circuits even though they seem so simple. The reason is because of the behavior of the diodes under stress with various kinds of post filtering is rather complex. We are dealing with very low impedance devices so there are problems that can come up.
Take a 1 amp diode and pump an already rectified sine wave though it and you get some normal power heating. Double the current but halve the time (half wave rectification) and in basic power heating theory you get to use that diode for twice its rated current. Looking closer however, the diode voltage goes up somewhat and so it causes more power heating than the basic theory would predict. This means we'd have to derate the diode by about 20 percent of twice the rated current (half wave) which brings us to 1.6 amps in this half wave sine wave application, but this is with an actual half wave sine, or a diode in a full wave bridge with resistive load only.
Problems set in however when we add capacitive filtering. Suddenly the diode current is not sine anymore, sometimes no where near sine, but more like a short pulse of current. With only a small amount of capacitor ESR, we could end up with as much as three times the power heating in that diode even in bridge configuration. That's pretty amazing if you ask me, but there is a saving grace built in to most low voltage rectifier circuits: the transformer. The transformer has primary and secondary resistance and even some inductance, which lowers the peak current through the diode and so saves the diode in the end from overheating.
The exact current peak is very hard to calculate without knowing a lot about the transformer, but many transformers have significant equivalent resistance and so brings down the peak current quite a bit, at least in most low voltage low power circuits. In higher power circuits however, we have to be very careful because often the minor resistances are not as large as with the lower power circuits because we dont want excessive inefficiencies to eat up too much of the available input power. We do want to be able to use the diodes to full capacity though, and in rectifier circuits we can sometimes use the diode at a higher than rated current. It depends mostly on the external circuit impedances.
So what's the best medicine here then with all these variables?
It appears that the best is to take a couple measurements and see just what your diodes are doing. Often in a bridge you only have to measure one diode to see what the others are doing too. A measurement of current and voltage on the scope is a good idea, followed by some simple calculations to calculate the power heating. Based on the package size you should be able to determine if you've got a safe workable design or not.
It's interesting that if many of our low voltage circuits that we use every day did not have lossy transformers in them (wall warts) we would probably be burning up diodes on a regular basis