6
\$\begingroup\$

While it's easy to find ESD ratings for ICs (bipolar, CMOS, and any mix) and to some extent for MOS transistors (for the inherent reasons), I don't find ESD rating for many common diodes (left apart TVS and Zener devices, given their use).

Let's say I'm looking for ESD rating of an S1J or S1M diode: no manufacturer seems to detail it. Is it because being massive, high-voltage parts (operating up to 1kV DC) they already are over an "interesting" value?

Again, if I look for the same rating of a BAS21 part, just a 250V diode, I can't find it from common suppliers.

Even going down to an humble BC847 bipolar NPN, no details on ESD rating, looking in a few data sheets.

If any of these parts is handled wrong, or connected to the outside port of a device, could easily come in contact with dangerous voltages. I'd like to know more on these ratings and why they aren't listed.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ All these components have various voltage limits specified in their data sheets so are you talking about "ESD" from the human body model perspective or something else? If your question relates to handling procedures then please say. \$\endgroup\$
    – Andy aka
    Commented Aug 9, 2018 at 11:17
  • \$\begingroup\$ As with a comment below, my original question is maybe unclear. Take it as: why an ESD suppression diode has declared limits in its data sheet, while a standard diode doesn't? The HBM and MM are possibly different, but what I mean is that there is no rating at all in their data sheet: I know that a 1.5KE100A operates around 100V IEC-61000-4-2, what for an 1N4002 that's 100V rated? \$\endgroup\$
    – LuC
    Commented Aug 10, 2018 at 9:41
  • \$\begingroup\$ I was wondering if you ever got an answer to your question. I have 2 diode products that I can't seem to get answers from manufacturers if they are considered ESD Sensitive or not (my question is more in regards to handling of these items at our facility). \$\endgroup\$ Commented Mar 8, 2023 at 22:02
  • \$\begingroup\$ @Stealthgirl well unfortunately I found no answer to that. They clearly are sensitive, as any other junction device, what's unclear is their IEC-61000-4-2 level. As a cautionary statement, you can consider them to be level 1, but it's quite obvious that a BAT54 is quite different from a BAV23 or a S1M... \$\endgroup\$
    – LuC
    Commented Mar 9, 2023 at 9:49

3 Answers 3

9
\$\begingroup\$

ICs are ESD protected by connecting diodes from I/O to rails.

How would you protect a diode? By using another diode? The diode has two possibilities when hit with an overload. (a) It conducts in the forward direction. Many diodes have a peak current specification, which tends to be orders of magnitude higher than the sort of diodes you can fit into an IC for protection. (b) It blocks in the reverse direction. You already have a specification for the peak reverse voltage it can take without breaking down.

Transistors are similar. If you hit a junction in reverse with an ESD spike, you already have in the data sheet the voltage limit for that junction, above which it may break down. A forward junction will tend to protect itself as for a diode.

Generally, the common 'cooking grade' transistors (like BC847) are fairly robust, built with big processes. It's ICs with small junctions that need explicit protection, and FETs with a very high impedance gate. You can buy microwave diodes and transistors that are very susceptible to ESD, as they have to be built with tiny junctions to work at high frequency.

\$\endgroup\$
4
  • \$\begingroup\$ yeah, while reverse currrent ESD does damage ordinary transistors the damage is not enough to measure unless repeated many many times. \$\endgroup\$ Commented Aug 9, 2018 at 10:52
  • 2
    \$\begingroup\$ "The diode has two possibilities when hit with an overload." Three possibilities: if the reverse voltage is high enough, it'll go into reverse breakdown mode and start conducting electricity again. \$\endgroup\$
    – nick012000
    Commented Aug 9, 2018 at 12:01
  • \$\begingroup\$ I'm sorry to disagree to some points. First, you can protect a standard diode with an avalanche/TVS diode, that's a fact. Second, BJT ain't that "robust", just figure of a common base connected BC847, where you expose the emitter terminal to the outer world; its breadown voltage is quite low, around 5-6 V maybe, but there's more (see next answer) \$\endgroup\$
    – LuC
    Commented Aug 9, 2018 at 16:13
  • \$\begingroup\$ BTW, about junction size: any large size blue/white LED is very easy to be broken with any tiny reverse current, even if it's tens of Watts rated (when directly polarized). For that reason they are normally protected by an internal Zener/TVS \$\endgroup\$
    – LuC
    Commented Aug 9, 2018 at 16:26
3
\$\begingroup\$

Ordinary junctions (think diodes and BJTs) are damaged by heat caused by the current rather than the current alone, so they can dissipate a fair amount of energy before dying. Common diodes an BJTs can dissipate a few millijoules, which is about what a ESD shock delivers.

ESD can easily damage a junction which is reverse-biased in the first place. This is because the initial avalanche current produced by ESD will be sustained by the reverse voltage applied to it, delivery far more energy than the initial ESD discharge.

\$\endgroup\$
1
  • \$\begingroup\$ A high enough energy can overheat local points in a P-N junction, causing it to fail way before the whole junction has been heated. Usual ESD damages aren't about large areas in the junction. There are many examples from X-ray inspection of faulty devices, taken in the automotive field (where there are large investments in security), and reported by IPC. Some of them make the devices work apparently fine for months, and then abruptely turn to some fault. \$\endgroup\$
    – LuC
    Commented Aug 9, 2018 at 16:21
0
\$\begingroup\$

ESD circuits survive the ESD event by guiding the ESD charge flows to be deep in the silicon, so the BULK is heated up.

Diodes naturally have large volumes to be heated up, at least when forward biased.

\$\endgroup\$
1
  • \$\begingroup\$ The main question still holds: if a TVS diode, which is designed for ESD suppression, has a limit defined under IEC-61000 tests, why a standard diode (which isn't defined for ESD suppression) has no limits declared in its datasheet? Maybe I should have posed the original posting like this, my fault \$\endgroup\$
    – LuC
    Commented Aug 10, 2018 at 9:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.