2
\$\begingroup\$

Forgive if it is a silly question, but I couldn't find any answers to this online. I find it hard to understand some ESD related issues. I believe that, typically, Electrostatic Discharges occur when something electrically charged touches or get sufficiently close to another thing that is grounded (I understand that it doesn't necessarily have to be grounded. I'm talking about common ESD related problems in practice). If this "other thing" is an ESD sensitive component, the electric current that passes through it may damage it.

Assume there is an ESD sensitive component, with no electric charge, in an antistatic bag (isolated from ground?). An electrically charged person opens the bag and touches the component with her bare hands. I presume that, since the component is not connected to ground, no current could pass through it, only to it (assuming some electric charge would flow from the person to the component, until they are at the same potential).

Could this current easily damage a component? Or the electric charge necessary for it to be damaged is too high to be a concern?

Maybe my question is somewhat similar to this one, except it asks about a bird killed by a powerline instead of a component damaged by an electric charged person. The answers mention that a very high voltage would be necessary to kill the bird: Can a bird, previously at earth potential, get electrocuted by landing on a powerline at high-enough voltage due to the initial “equalization charge”?

\$\endgroup\$
0
4
\$\begingroup\$

While it is less likely to create hard-faults with Electrostatic Charge in a component that is not connected to any bulk surface to dissipate or equalise the initial impulse, it is still possible.

Say you are negatively charged compared to the chip, that means you have "excess electrons", so when you tough a pin of the device, that pin will be connected to your excess and will "want" to become the same charge. This induces a current. It may be shorter in duration and could be smaller than when the chip is "grounded", but it can be enough to kill it, depending on the chip's innards.

Even the over-charge on one pin compared to another can destabilise some constructs in Chips, especially MOS devices, sometimes even permanently.


Note about "Anti-Static" bags, There's a couple types:

  1. Pink Anti-Static: Not safe at all for ESD sensitive chips
  2. Black Dissipative: Very safe for ESD sensitive devices, but less so than the metal-foil ones.
  3. Metal Foil Conductive: Extremely safe.

The Pink bags are only anti-static: It does not create charge when rubbed by another surface. So they are nice in a shipment with several bags, to prevent static to build up, but a human-body zap can very easily go straight through the plastic to any chip.

The black dissipative actually conducts a bit of electricity, usually a few hundred kOhm to a MOhm per square of resistance, and they will dissipate any charge built up across it and human-body discharges are extremely unlikely to penetrate the bag to the chip, but high level discharges may still affect the chip.

Metal foil conductive: The name says it all, has a very low resistance per square. Some bags have the foil (or a micro-laser-perforated foil to allow some see-through) on the inside, some have an extra layer of dissipative material over it, to protect the metal film. It will be extremely hard to have any kind of zap go through the bag, as the foil will conduct it from one side to the other. Even high intensity discharges will have trouble getting through, with the exception of discharges that will vaporize the foil, since that will make the bag (and chip) explode. And of course a few below that level, but I wanted to conjure the image of an exploding ESD bag.

\$\endgroup\$
3
\$\begingroup\$

I recently did a design involving IEC60601. It was a USB DAQ type device, with a 5kV isolation barrier between USB and the main measurement circuit, and another 2.5kV isolation barrier between the main measurement circuit and some electrodes connected directly to the patient's skin.

We were testing to IEC61000-4-2 Level 4, Critera A, meaning +/- 8kV for 10 hits each, with no effect on the device's operation. In this case, the ESD test gun was earth referenced, and the isolated circuit floating. With no ESD protection devices in the circuit, it would have blown up the amplifier connected to the electrodes. ESD protection devices protected the amplifier, but dumping that huge pulse of current into the ground plane caused problems. Finding the magic bullet to protect the amplifier and absorb the energy took some time.

tl;dr, earth referenced ESD can absolutely damage isolated devices, and isolated ESD can damage earth referenced devices. The accumulated charge doesn't care about the reference at all. It wants to equalize with whatever it comes into contact with.

\$\endgroup\$
1
  • \$\begingroup\$ You, its all about equalisation of charge between two bodies. You do not need a reference point for that to occur \$\endgroup\$
    – JonRB
    Aug 1 '15 at 6:47
0
\$\begingroup\$

It is very easy to damage a circuit with ESD WITHOUT EVER DISCHARGING ANYTHING.

A simple experiment demonstrates this.

  1. Wash your hair and let it dry.
  2. Blow up a balloon.
  3. Rub the balloon against some fur (or your own hair, but the purpose is to show that you can transfer static charges without touching the thing that holds static.
  4. Once you have charged the balloon on the fur, stand in front of a mirror and slowly bring the balloon towards your dry hair.
  5. You will see your hair attracted towards the charged balloon. The charge holder (balloon) has induced an opposite static charge into the hair simply by being near to the hair.

If you -- or anything with a static charge -- comes 'near' another object that is even partially conductive, an opposite charge will be induced in that object. Whether or not that induced charge can damage the device depends on the induced voltage and how much energy is contained in the combined induced charge and voltage.

For TTL circuits this was not usually a problem.

For CMOS circuits in their 12v and 5v incarnations, this induced charge was not typically a problem. Usually direct contact was required for a damaging discharge.

For 3.3 volt, and lower voltage devices, however, this induced charge can and sometimes does damage devices without you ever needing to touch the device. The charge induced on any part of the device will further induce charges on parts near to that and so on in cascades that can be well removed from any part of the device "near" the original induced charge. If the induced charge at any of those locations exceeds the insulation around that location -- air gap, semiconductor material, etc. -- electricity will flow. If the charge is sufficiently energetic, it will damage something.

To prevent this --

  1. Wear a static wrist strap, AND
  2. Make sure that wrist strap is correctly attached to a static-suppressing work area, e.g. an anti-static mat on your table, AND
  3. Make sure the anti-static mat is correctly grounded to the building/work-area ground, AND
  4. Make sure the device ground -- and device power if device is not operating -- is ALSO attached to that anti-static mat, AND
  5. Use chairs and wear clothing that minimize generation of static, AND
  6. REQUIRE anyone who approaches to ALSO put on a static wrist strap and attach to the same anti-static mat BEFORE letting them anywhere near the device(s).

This might seem like overkill. And for inexpensive parts or equipment it probably is. BUT the first time you fry a multi-thousand-dollar piece of equipment without knowing why, it will in hindsight not seem like overkill after all.

You can -- in an emergency DO NOT DO THIS AT HOME I am a registered Professional Engineer who works with electricity and electronics and even I try to never use this method -- put down aluminum foil on your desk, and use an alligator-clip to connect the foil to the ground in an outlet -- VERY HAZARDOUS AND LIKELY TO GET YOU KILLED IF YOU DON'T KNOW EXACTLY WHAT YOU ARE DOING -- and then always make sure you are touching the foil and that you also attach an alligator clip from the foil to a common ground on the electronics.

Other examples of static transfer with no contact:

  1. Bend water -- https://www.wikihow.com/Bend-Water-with-Static-Electricity
  2. Static induction -- https://en.wikipedia.org/wiki/Electrostatic_induction
  3. Pick up things without touching them -- https://www.msichicago.org/science-at-home/hands-on-science/static-cling/
\$\endgroup\$
1
  • 1
    \$\begingroup\$ ESD sensitive components can be damaged by a few volts. Changes in charge occur constantly in our environment. Take a look at a high-resistance voltmeter such as the Fluke 8846 or HP 3458 and you will see that the measuring devices show 10 to 20 volts without anything connected. \$\endgroup\$
    – arnisz
    May 23 at 15:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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