What are the Signs of Diode Degradation and Aging?

[tim9000]

Hi,
Does anyone know if there are such characteristics that are tell tale signs of diode degradation? I'm not sure if diodes fail short or open, I'd have thought open, but I'm not sure.
I was wondering if as a diode gets old the forward bias voltage might increase? Or the diode power loss per forward conduction Amp might increase? Something to indicate where the diode is on it's lifespan.

Cheers!

 

[davenn]

Does anyone know if there are such characteristics that are tell tale signs of diode degradation?

good Q, I'm not sure

I'm not sure if diodes fail short or open, I'd have thought open, but I'm not sure.

I have seen both

D

 

[Windadct]

At the Diode level - they tend to be relatively large semiconductor structures and do not really age or wear out when kept within spec, I guess size really has nothing to do with it, I can not think of an wear-out mechanisms for any semiconductors. Failures may include, separation of the lead from the crystal, environmental breakdown causing migrating moisture into the diode, separation of the crystal from the case / housing causing increase in thermal resistance ( thermal failure) - and then just misuse over voltage, over current / over heating.

Some of these you can test for, some you can not. Thermal performance is pretty straightforward. IN all cases good baseline data and a consistent measurement technique are critical. Initial test should be "in place" and datasheet values are not so good to reference to a field measurement.

I work whit large devices, every case is different.

 

[tim9000]

At the Diode level - they tend to be relatively large semiconductor structures and do not really age or wear out when kept within spec, I guess size really has nothing to do with it, I can not think of an wear-out mechanisms for any semiconductors. Failures may include, separation of the lead from the crystal, environmental breakdown causing migrating moisture into the diode, separation of the crystal from the case / housing causing increase in thermal resistance ( thermal failure) - and then just misuse over voltage, over current / over heating.

Some of these you can test for, some you can not. Thermal performance is pretty straightforward. IN all cases good baseline data and a consistent measurement technique are critical. Initial test should be "in place" and datasheet values are not so good to reference to a field measurement.

I work whit large devices, every case is different.

Specifically relevant to me atm is the: https://www.google.com.au/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&cad=rja&uact=8&ved=0ahUKEwjO3vuQh9fQAhWQNpQKHXvoAoUQFggmMAI&url=https://www.semikron.com/index.php?id=228&tx_smkdownloads_downloaddownload%5BnameWithArticlenumber%5D=SKN_130_02235290&usg=AFQjCNHHc0xdkxgexff4wuBK_SDIYKOZWQ&sig2=-Na6CSG0bOrvisAKDkYkCA

semikron 130 diode.

I understand what you're saying about the crystal being sensitive to physical trauma separating it from the metal leads, or getting moisture in the junction or contact surfaces. However, regarding the wearout mechanisms: a lighting bloke told me that they [LEDs] experience something like electrolysis; so would have a lifespan even within spec (admittedly this is anecdotal hear-say), but I didn't learn in that detail back doing semiconductor physics at uni. I want to got out and test these diodes forward bias voltage, leakage current and maybe breakdown voltage out in the field, to see which ones will need replacing, as a preventative maintenance.
These diodes might see a few amps through them for a few seconds every 15 minutes (maybe 6A peak), they're housed in boxes on big heat-sinks. And they do wear out (over a long period, I expect), I assume not just from fault conditions, which would be very rare; so generally they're used under-spec.
...it's all very mysterious to me, and I haven't had the experience to make my own observations over a long period, yet.Cheers

 

[nsaspook]

I have noticed several cases of CTR (current transfer ratio) reduction in 20+ year continuous service optocoupler devices to the point where electrical interfaces no longer work (repaired by simple opto chip replacement) due to what we suspect to be internal LED dimming in a industrial environment. Maybe it's the high stress and elevated temperature of this one application but the quoted lifetimes seem to be optimistic.
http://machinedesign.com/news/gauging-led-lifetime-optocouplers
http://www.vishay.com/docs/83741/83741.pdf

One device we investigated for bulk equipment ageing replacement was the obsolete HP R8060 with the modern equivalent ASSR-1611.
http://www.avagotech.com/products/o...lastic/other/solid-state-relay/assr-1611-001e

 

[Windadct]

My point is that the "Diode" is not really what fails, wears out. It is the application and life of the packaging that leads to their death. Stud diode are very rugged, will be hard to wear one out in the traditional sense.

Have you had failures? A 130A stud diode will carry 6 amps, without a heatsink...also what is the application voltage, relative the Vrrm of the Diode.

The Semikron Handbook is very good, while a lot of discussion on lifetime / reliability for IGBT modules, the Stud diode reliability is not discussed, but some of the details on construction are interesting ( well to some people..)