×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Schottky diode failure

Schottky diode failure

Schottky diode failure

(OP)
I am having a problem with a 60V Schottky diode.  We are having around a 1% failure rate of the diodes in production.  This appears to be an early break-in type of failure, since we don't have any failures that show up during life testing.  The failure mode is a short.  We have not been able to successfully locate the cause of the failure, but have limited it down to 3 causes:  overvoltage transient, process damage, or inherent defect.  
The function of the diode in the circuit is to act as a barrier diode to block reverse voltage.  The load is a DC/DC boost converter.  After the diode and parallel to the load we have a filter capacitor.  The input range of the circuit is 16VDC up to 42VDC.  The circuit also has an MOV on the input.  
I have been investigating this problem for the last 2 weeks and haven't gotten very far, so I figured I would turn here for some ideas.  
We know that the diode has a avalanche rating of 90V, and confirmed in testing that it can withstand 95V for 1 msec repeating at a 1 Hz rate, but failed at 100V.  
We tried a pilot run of 1000 units with a 100V diode and still had a 1% failure rate.  
We also know that the MOV does not appear to clamp fast enough, so the diode will try to clamp any negative transients before the MOV can react.  
Right now my hunch is that something is happening somewhere in the production process to cause the damage, or else the parts are just bad to start with.  We returned some of the failed parts to the vendor for analysis and are holding our breath until we get an answer from them.  A one percent failure rate seems to point to a quality problem and not an inherent design defect.  If it is indeed a design flaw, then I will probably have to start looking for a new line of work.  

RE: Schottky diode failure

Are they dead shorts, or close?

A potential problem with junction devices subjected to high currents is the migration of the contact metal into the junction, thereby shorting it.

Usually, breakdown problems tend to "soften" the breakdown curve, while high currents cause shorts.

TTFN

RE: Schottky diode failure

If the MOV is not handling the negative transients quick enough then you might try a TVS or zener in conjunction with the MOV (or without the MOV). Have you captured the transients on a scope to know what level you are trying to suppress? I think that should be your first step so you know what your trying to suppress and if the diode can handle it.

RE: Schottky diode failure

(OP)
Sometimes they are dead shorts, other times they are near shorts, less than 20 ohms.  We also have parts that appear to fail but measure normal (~0.2V) as a diode, but the reverse leakage current is through the roof.  

I would be surprised to have the cause of the failure be excessive current, since the part is rated for 42A for 6 msec.  

RE: Schottky diode failure

What is the part and what is the current?
Package type and mounting ?
Try a different brand. Different companies make
the same type of diodes differently.
Any failures at first turn on?
Do you do any type of burn in?
I have had simmilar results, about 1% with in the first
year of service, that was do to silver migration problems.
Changed manufacturer and the problem went away.

RE: Schottky diode failure

High temperature at a lower current can also have the same effect.  The resistance values are small enough that it's very likely to be metal spiking through the junctions.  

It could still be some sort of process problem, aggravated by something happening in your system.

Are the "100V" diodes verified to have as much margin as the "60" diodes?  If so, then either the transient is so large that the reverse breakdown is irrelevant, or it's caused by a forward condition with excessive current.

Have you tried something like paralleling a couple of diodes or possibly try a diode with a MUCH higher current rating?

TTFN

RE: Schottky diode failure

Assuming that this circuit has worked in production before the simple test would be to change to another vendor and see what happens.  This is not an elegant test but it is damned effective in reality.

RE: Schottky diode failure

How much exposure does the diode have to the inductor in the switcher?  L di/dt is a great killer of semiconductors.  Resize the diode based on its coping with that situation and also look at what other inductance is around.

I set up an experiment once to measure transients due to stray inductance, where 4A constant current flowed through a length of 20# hookup wire (4m or 10m long) tossed loosely on the floor and a fast digital CRO with low capacitance probes was connected across the ends of the wire.  When the current was interrupted by disconnecting the power supply by hand, we registered varying voltage transients of up to 270V> Don't recall the exact spike duration, but it was less than 1uS.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources