OT: bridge rectifier failure mode.

Nov 06, 2015 11 Replies
Saved reading
Jump to replies (11)

I've just had a bridge rectifier fail with two of the four diodes dead short. one of those 1" square blocks with a spade tag in each corner. Type GBPC1510. Rated 15 amp 1000v piv used to rectify 100v AC 50hz day in day out at average of say 5 amps peaking at 10 from time to time. Feeding a reservoir capacitor and non inductive load. On a heatsink cooled with a fan mainly for the benefit of other things inside the case.



I can't think why it should have died



Do these give up of old age? possibly about 3 years service - no date code on it.



TIA Bob


Don't think they expire with age. I've got a few power amps here, and the oldest one in daily use is 1976.

They can. Temperature has a big influence. Generally, each 10C rise halves the life of an electrical assembly.

Bare in mind current only flows in the rectifier for the short period of time required to top up the reservoir capacitor since last half-cycle. This means the current flow in the rectifier is much higher than the average current draw from the PSU output, albeit for a short time. However, I²R heating is high due to this, and the larger the reservoir capacitor and smaller the ripple, the higher the I²R heating is. Poor connections to the bridge rectifier can make this worse, conducting heat into the diodes via the terminals. Push-on spade connectors can cause resistance heating on the terminals, as can a soldered joint which has run hot for some time and turned into a dry joint. If the wires have insulation on them, the state of the insulation may hint if the wires have been running hot for a while. If you have spade connectors, when you replace the bridge rectifier, close up the connectors with pliers a little before connecting so that the connections are extremely tight to assemble (or cut off and solder).

no. Military & space apps pay extra for higher reliability chips, but most failures are down to imperfect design, especially with consumer goods, where price always rules.

NT

I've just had a PA amplifier fail. My diagnosis is a burnt out primary on the mains transformer. I'm told that it cannot be repaired. And a replacement unit is not available. I wonder why? Mind you, it was't expensive & I'm being given my money back.

Naked dreams?

More rubbish disinformatio9n semiconductors degrade with age. At high temepratures life degrades to fractions of a second from tens of years at say 50C. Running kit at high (125C) junction temps will mean premature failure. Needless to say many commercial apps do this.

Another cause of failure is a voltage spike. That will gibe secondary breakdown on the diode(s) generally making it/them go short.

Not strictly true. Most ICs with extended temperature range will be the same IC as those sold commercially but with additional testing.

On occasions an alternative hermetic package will be used which may also have a better thermal expansion match to silicon.

When you say 100VAC, I presume this was powered from a transformer?

Just that there can be an awful lot of inductively stored energy in a such a transformer which has to go somewhere if you disconnect the primary at the wrong time.

I'm also wondering what reservoir capacitor can cope with say 10% ripple at 10A? Would it be 10,000uF at 200V rating?

I'm afraid they do.

You can play about at the tap end or go right down to the other end, but you're always somewhere in the bathtub.

Everything will eventually fail.

My point was an adequately rated and mounted bridge rectifier isn't the most likely first one on most things. Based on repairing a large amount of assorted electronics.

25a amp replacement fitted today; only pennies more than to replace the original 15 amp one and the same footprint so we will see how that lasts.

Bob

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required