Wednesday, September 07, 2005

What level of ESD will damage parts?

>> I have currently been tasked with trying to improve the ESD standards
>> within our manufacturing facility.
>> One question which has arisen which I cannot find an answer to is this. Is
>> there a level of ESD at which it is generally accepted that on or above
>> this level, significant damage will occur to electronic parts?
>> We are shortly going to be going through the process in question, and
>> measuring the charges and voltages which are created whilst working. It
>> would help us greatly to know if there is a "safe" level of ESD which can
>> be allowed to occur.

The ESD susceptibility of devices depends on the particular component. Each has an "ESD withstand voltage" determined during QA tests. Virtually all semiconductor components have been tested using "Human Body Model" (HBM), some have also been tested using "Machine Model" (MM) or "Charged Device Model" (CDM). Unfortunately many manufacturers do not make this information readily available to users. As a result we end up relying on guesswork for the susceptibility of components. A rough guide is given in our ESD Guide

The ESD standards such as 61340-5-1 are designed to protect devices down to 100 V HBM. For many processes this will be adequate. Some types of components are more susceptible than this (< 100V HBM) and special measures and care are then required.

It is not easy to assess ESD risk in a process. The usual way is to measure electrostatic fields and potentials in the region of the ESD susceptible parts. The 61340-5-1 standard recommends that electrostatic fields should not exceed 10,000 V/m and potentials (voltages) should not exceed 100 V. Note that a 1,000 V/m field could be a 10,000 V source at 1m, or a 100 V source at 1 cm distance etc. In practice I regard the field criterion as the most useful.

The best approach is to remove all non-essential insulators and electrostatic field sources in the ESD Protected Area (EPA). Any essential insulators or other sources may then be assessed for ESD risk, and appropriate ameliorating action (e.g. use of ionisers) taken.


DUBAN Michel said...

As usual, I agree what you write, but my experience show me like written in normative documents and your note, these documents are done for general application, both ICE 61340-5- 1 or 2 and S20-20
gives rules for ESD sensitivity of components in HBM (Human Body Model) of 100 V.
But many components are more sensitive for someone the treshold is not greater than 10 volts. And the HBM is not the most dangerous. Even when people are connected to ground by a wrist strap or by their shoes on dissipative or conductve grounding system, the CDM (Charge Device Model) or MM (Machine Model), FIM (Field Induce Model) become the most important.

In this case it necessary to limit the charge level and the risk of "contact metal" (as defined by ALBERT Wallash (see his Internet Site). A contact metal is a contact with an impedance less than 10 K ohm. In that case you limit the power dissipated during an ESD. So you limit the ESD risk.

DUBAN Michel said...
This comment has been removed by a blog administrator.