1. Home
  2. Electronics Design
  3. looking for: extremely ESD senistive componets - for demonstrationg safe ESD practices

looking for: extremely ESD senistive componets - for demonstrationg safe ESD practices

To help back up the trainings of my employees about the potential dangers of ESD, I'm looking for the most ESD sensitive components.

To be of use, the components have to be:

1) cheap (they will be destroyed in the demonstration) 2) easy to obtain (MR heads don't count for this reason) 3) easy to show some form of functionality (MR heads are also not good for this reason)

The things that come to mind are old CMOS 4000 series logic IC's (the newer types seem to be much more robust), small signal MOSFET's (2N700x), or maybe a small signal RF MOSFET.

The intent is to build something like a LED flasher / sequencer, remove the said ESD sensitive component, make sure the air ionization is turned off, people are not wearing the normal ESD protective stuff (smocks, foot straps and wrist straps), or using the conductive chairs or are on conductive paint, etc, and have the said component passed around the room. Hopefully someone will have a nice furry sweater on, and the humidity will be low. If necessary, mount the part on a piece of PCB designed to encourage one person to be touching the ground / power / source pins, while the other person would first grab the end with the gate or I/O pins / pins exposed.

Any suggestions on what is the most sensitive component for this application?

Reply to
Jeff L
Loading thread data ...

Don't underestimate Murphy's Law: If someting is considered to fail, it will not fail.

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
Reply to
Uwe Bonnes

[and demonstrate failure after handling]

There's a gizmo used by shippers called a 'shockwatch' that does this kind of task. Various types are used, like a prestressed glass tube of dye, weight-trapped-by-springs, etc. If the package to which the shockwatch is attached takes a hard bump or fall, the dye stains a card, or the weight dislodges.

So, I'd think in terms of a power MOSFET with gate floating, with a battery/diode to bias D-S to 5V or so, and a fuse in the drain (maybe 0.5 A). If the gate turns hard ON, the fuse blows. This all goes on a plug-in card with card-edge fingers that your testers will be touching or getting near to. Make sure some of the touchable metal connects to the battery +5V, if you don't want to trust in static.

Your test circuit will operate from a higher voltage (like, 12V) and will pass only 100 mA of current in, say, an audio buffer amp application. When in-circuit, the diode reverse biases so the battery isn't part of the active circuit. When out of circuit, leakage keeps the gate turned off (mainly).

Most modern MOSFETs are gate-protected (I well remember the screams from RF designers back in the 1970s when unprotected MOS got dropped from the product lines), but this approach doesn't require the voltage stress to be outside the normal device range, so shouldn't be subject to effective gate-voltage-protection built into the MOSFET. The fuses, though, are gonna need replacement.

Reply to
whit3rd

Two things come to mind. One is anything with an open drain, i.e. a device that does not have protection "diodes" to both rails. To do open drain ESD, the output is generally (but not necessarily) protected with a beefy fet with gate tied to source. [A resistor may be inserted between the protection device and the driver to make sure the protection device takes the abuse.] The breakdown of the protection device is expected to absorb the energy in one direction, and the built-in diodes handle the other direction. The other likely to fail pin would be one that has a very low leakage specification on it, implying a small protection device. You would have to measure before and after leakage.

You can spot ESD damage with a curve tracer if the abuse is sufficient. Believe it or not, a chip can work after ESD damage. For instance, damage to a large output device could be a damaged diode junction (drain to substrate). The output device could drive the damaged junction. The only way you would find out would be additional supply current due to driving the damage.

In real life, you find the damage with an emission microscope or liquid crystals. This require the die be visible.

Analog chips generally have poor ESD compared to digital. [i can't speak for RF, but I'd put them in the analog category.] To complicate matters, it is possible to simply charge the internal nodes of the chip, making it behave as if it was damaged. If you wait a while, the charge drains and the chip returns to normal. Heat speeds this along, and the chip doesn't have to be powered while heated.

How about something similar to ESD damage, but not really damaged. I'm thinking of a fet in series with a LED and resistor. Float the gate of the fet. Perhaps touching this gate will cause the LED to light due to charge transfer. I never tried this, but it could work.

Reply to
miso

I suspect these are old enough to have static sensitivity issues...

formatting link

url may need splicing.

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
Synergetics   3860 West First Street   Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml   email: don@tinaja.com

Please visit my GURU's LAIR web site at http://www.tinaja.com
Reply to
Don Lancaster

1N23s are very sensitive, but fail badly on points 1 and 2 above. I don't know how costly or available the 40673 and 3N140 DG Mosfets are these days, but they were pretty easy to damage.

Failing that, old CMOS should do, the older and cheaper the better.

I did a few demo/tutorials on ESD a while ago, and while I did not damage anything, an electrophorus was great fun and much appreciated. I just used it to fire a neon bulb.

See the IEEE "EMC Education Manual" page 13 for details on it, although I did not need the Teflon sheet specified, polythene worked fine. You can download it from here:

formatting link

Barry

Reply to
Barry Lennox

Sorry, and the other side of the deal is you make sure you give them the tools to do the job.

DNA

Reply to
Genome

laser diodes

Cheers Terry

Reply to
Terry Given

Thanks Everyone!

A special thanks goes to Barry for the link to the "electrophorus" and some sensitive part numbers!

Jeff

Reply to
Jeff L

You will have a lot of fun fooling with it, and determining the best combinations of plastic sheet and cloth. Rabbit skin is supposed to be well up there, but it was hard to find.

A few hardy souls in the class were prepared to take a little zap off it.!

Barry

Reply to
Barry Lennox

I've tried this, & it works fine. You don't even have to touch the gate pin to enable it, just waving your hand nearby can switch it on, which looks impressive as hell.

--
   W  "Some people are alive only because it is illegal to kill them."
 . | ,. w ,      
  \\|/  \\|/              Perna condita delenda est
---^----^---------------------------------------------------------------
Reply to
Lionel

Oh man, that is one nifty document. I'm going to have to build a few of those experiments just for the fun of it.

Thanks for the link, Barry.

--
   W  "Some people are alive only because it is illegal to kill them."
 . | ,. w ,      
  \\|/  \\|/              Perna condita delenda est
---^----^---------------------------------------------------------------
Reply to
Lionel

I did one of John Howard's courses 4 years ago (He's on the IEEE EMC Society Education Committee) and he said they were in the process of updating and expanding that EMC manual. It hasn't happened yet, but it should be very good when it's done. You might want to check their site from time to time.

Barry

Reply to
Barry Lennox

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.