Can ESD break only resistor and not LED?

[jkprg]

Hi,

We use cable with standard (THT) LED in series with metal foil resistor 470R/0.4W:

connector -- cable -- resistor -- LED (loop back to the same cable and connector)

Resistor with LED leads are poured by silicon. Cable was tested OK with 12V and current within the spec. But after transport the same test failed finding failed R. Does it sound crazy ESD would break only resistor and not LED? I've got 20 pieces like that from the last batch so it wasn't single random occurance (but no problems during last 3 years). The last batch resistors are definitely out of spec - at least they all have bad(thin)/none lacquer coating and maybe the metal foil is bad/thinner/impurities as well.

One of my friend did a single 2kV(limited eq.) ESD test and the resistor failed visually the same way (only) after dozens of pulses. But it's strange to me LED was always good and its VA characteristic stays same.

Thx for any comments.

[Zero999]

It seems unlikely ESD would cause a 470R resistor to fail, but not the LED. I would have suspected a mechanical failure, rather than ESD.

Did your friend ESD test both the resistor and LED, in series, as per the circuit?

[jkprg]

He did the ESD test as per circuit. After failure detection he measured each part separately. The test was done in university lab according to IEC 61000-4-2 (150 pF/330R) in continues discharges mode 10 Hz. He agreed it's unlikely the circuit was exposed to dozens of 2kV ESD pulses but he thinks it could happen with small number of 10k+V pulses.

Mechanical failure was my first assumption too because wire is soldered directly to resistor. I will probably ask him to do a mechanical stress test.
Thanks

[Medved]

I did encountered problems with 6kV ESD damaging a 100 Ohm series resistor and not the 0.35um CMOS based chip (the chip was actually being the DUT; OK, it was specifically designed to withstand such discharge on that pin).
It was a regular SMD resistor, according to its manufacturer's representative it likely experienced a flash over which is killing for the thin layer construction, there was supposed to be used a special type (I don't remember which, because it was confirmed it is not the chip, so ours, problem, so nothing to fix for us...).

[TimFox]

What is the final state of the failed 470R resistor by itself?  Sometimes, an applied high voltage will leave a track across the surface between the two ends, either metal or conductive residue from insulation.

[Medved]

What is the final state of the failed 470R resistor by itself?  Sometimes, an applied high voltage will leave a track across the surface between the two ends, either metal or conductive residue from insulation.

In my case it ended up open circuit (material eaten away by the consecutive discharges, as the test protocol called for a number consecutive pulses for each polarity)...

[ogden]

Resistor with LED leads are poured by silicon. Cable was tested OK with 12V and current within the spec. But after transport the same test failed finding failed R. Does it sound crazy ESD would break only resistor and not LED?

Indeed it sounds crazy that transport can ESD-zap 20 products in a row. Do everything with product batch but shipping. Then unpack and test. You may get surpsise. Hint: product may be damaged before shipping, not during shipping.

[jkprg]

What is the final state of the failed 470R resistor by itself?

My friend has found some marks (dark spots) on the resistance helix:
Sample 1:
Sample 2:
Some of the resistors end up in infinite resistance state, others have 38K-170K ohms depends on mechanical stress during measurement (170K at no stress).