System Level ESD Testing to IEC and other ESD Standards

Document Sample
System Level ESD Testing to IEC and other ESD Standards Powered By Docstoc
					                                      EMC Compliance
The world leader in serving science   Tests vs Reality!

                                         Michael Hopkins
Why we’re here today

 This short presentation is intended to look at EMC Compliance
 test levels for immunity and compare them with what we believe
 reality to be. Particularly those required by MOST product family
                  Electrical Fast Transients (EFT)
                         Radiated Immunity
                           Surge (lightning)
                           Magnetic Fields

The Effects of ESD ? (Not likely, but a great picture!)

The Effects of ESD ? (could be..)

ESD can and does destroy
 device junctions in handling.

At the system level ESD is
 frequently the cause of upset
 as opposed to damage.

A few ESD Standards

  •   European Norms: EN/IEC 61000-4-2
  •   Telcordia GR 1089
  •   ISO Automotive Standards
  •   Other corporate standards – GM, Boeing, HP, etc…

What is Electrostatic Discharge ?

A person or charged object discharging into
 a sensitive electronic component or circuit
 can cause a device to fail or circuit to be

ESD really is a miniature lightning bolt, but at what levels?

What is Electrostatic Discharge ?

The threshold of feeling is 2kV to 4kV.

Everyone can feel 5kV.

15kV is a memorable event!

What are Electrostatic Discharge Levels

 In a normal office or home environment – the environment most
  standards are written around:

  • We are all moving around at a few kV most of the time…
  • We frequently make low voltage discharges to electronic products,
    but rarely do we cause them to fail.
  • In extreme cases (Winter in Calgary) it could be more than just a
    few kV, but rarely more then 10kV.

What are Electrostatic Discharge Levels

   In special case environments:
    •       In a vehicle --- no more than a few kV
    •       Getting OUT of a vehicle – MUCH more than a few kV
            •   Could be 20kV or more!!! Remember q=cv; v=q/c
    •       Satellites
            •   50kV
    •       Helicopters
            •   50kV
    •       Hospital Beds ??

Electrical Fast Transients

   Arcing, which occurs as a switch opens, is both
   conducted and radiated into sensitive electronic

Electrical Fast Transients

IEC 1000-4-4 is the Basic Standard for EFT testing. Although EFT is
 mentioned in ANSI standards, they simple refer to the IEC. Variations on
 the EFT test are also found in standards for power sub-station
 environments, automotive and avionics standards.

Generic and Product Standards in IEC call for EFT testing on all mains,
 signal, and I/O lines
Tests are required to 2kV on the AC mains.

Electrical Fast Transients

 So what are real EFT levels ??
 Kind of depends on who you ask --- there isn’t much published
  data on EFT measurements for general electronics in the real
  world, BUT

   • Anecdotal information suggests:
      • EFT isn’t much of a problem in most light industrial or home environments – measured
        levels tend to be in the 10’s of volts rather than kV’s
      • It may be a problem in heavy industry environments where large power cables are run
        in parallel with signal and data lines.
      • It is definitely a problem in power sub-station environments, but the induced EFT
        doesn’t travel very far

Electrical Fast Transients

  In the power-sub station environment
    • EFT can be significant, especially close to HV switches.
    • EFT events are both radiated and conducted into nearby electronic
      controls (relays) in the sub-station control areas
    • Some measurements show burst frequencies with significant energy
      up to 50MHz resulting from switching operations in both Air
      Insulated (AIS) and Gas Insulated (GIS) substations.

  For this environment, ANSI, IEC and other standards exist that
   require testing to 30MHz and 4kV, which is in line with measured

 Radiated RF Immunity

 This is a big one for many
   • Requires testing in a semi-anechoic
     chamber with wideband amplifiers,
     antennas and other instrumentation, and
     it takes considerable time….

 International specifications call for test
  to done at 3V/m for products to be used
  in light industrial and home
  environments, and 10V/m in heavy
  industry or industrial environments.

Radiated RF Immunity

 Question is are these levels reality??

   • IEC 3V/m or 10V/m
   • For Avionics (DO160 Section 20) test levels can be from 4 to
     several hundred volts/meter, depending on frequency and Category.
   • For Telecom (Telcordia GR1089), it’s the same as IEC
   • The automotive industry tests to 200V/m

   Sounds like a pretty wide range of possibilities, depending on the
    product, the environment and the consequences of a failure.

Radiated RF Immunity

 Just a quick look at each:

   • IEC – general electronic products in office or light industrial
     environments – not typically high exposure areas for RF
   • Avionics – Turns out the terminal area at a large airport is the
     worst place to be for RF susceptibility -- RF transmitters
     everywhere – other aircraft, data circuits, handheld radio’s, cell
     phones, radars --- all concentrated in a small area.
   • Telecom – not much different than the light industrial applications
     --- exposures consistent with your average office or home.
   • Automotive – Exposure to high RF fields may not be very
     common, but with all the electronics in vehicles today and the
     advent of drive-by-wire, the primary issue becomes liability – an
     electronic failure could be catastrophic.

Radiated RF Immunity

 There doesn’t seem to be any such thing as a normal RF
  environment –

 The IEC levels of 3V/m and 10V/m were derived from the
  possibility of someone with a 5W handheld transceiver being in
  close proximity to an electronic product --- mainframe computers
  in process plants were the big issue of the day – and not based
  on radiation from modern cell phones which operate at much
  lower power levels.

Magnetic Field Immunity

 A non-issue for most electronic products
 Required for products that could be affected by AC Magnetic Fields !
   • CRT’s
   • Products using Hall Effect sensors
   • Your product susceptible ?

 The IEC standard for magnetic field immunity, IEC 61000-4-8, has
  been around for some time with no updates and virtually no
  controversy. This standard was recently re-issued with no changes or

Magnetic Field Immunity

 Although this hasn’t been an issue for most electronic products,
  there has been considerable interest in the effects of magnetic
  fields on people (and livestock) over the years, especially related
  to high voltage transmissions lines.

 Although there have been many reports of problems assumed to
  be associated with magnetic fields, there is no medical or
  scientific evidence to support these claims.

Surge Immunity

 Mostly a problem of lightning remnants
  being injected into the power mains or other
 Test levels vary depending on the standard
  being used:
   • IEC is 2kV and 1kA
   • Telcordia is 6kV and up to 10kA
   • DO 160 to 3.2kV and 2kA

Surge Immunity

 What’s reality ??

   • In July 2000, 33,863 cloud to ground lightning
     strikes were observed near Tamp, FL
      • lightning currents ranged from 5kA to 180kA
      • 2,597 were positive strokes and 31,266 were negative
      • Of the positive strokes, 95% were < 30kA and 98% were
        <60kA; 7 > 100kA
      • Of the negative strokes, 82% were <30kA and 98% were
        <60kA; 92 > 100kA
      • Only 2 strikes positive strikes and 5 negative were observed
        above 150kA

      • Taken from Square D Data Bulletin DB03A

Surge Immunity

 Reality for your products:
   • Depends at bit on where you are and what kind of lines we’re talking
     about ---
      • AC Power Mains inside a building
           IEEE Data shows injecting 10kA into standard residential wiring causes multiple flashovers at
            around 6kV ---- which indicates surge voltages INSIDE a residence or light industrial facility
            can’t be much higher.
           ANSI/IEEE C62.41 indicates that in a medium exposure area, one could expect to see 2kV
            surges in unprotected areas once/year and 1kV about 10 times/year.
           All that being said – how many unprotected areas are there these days?
               • Surge arrestors are everywhere!!
               • Your computer, stereo system, telephone answering machine, garage door opener,
           Today, surge voltages are kept pretty low inside a residence or office building but
            surge currents are another matter….

Surge Immunity

 Reality for your products:
   • Depends at bit on where you are and what kind of lines we’re talking
     about ---
      • Telecom lines
           Because these lines are really bundles of individual wires, all of which are closely coupled, the
            surge currents seen in telecom lines are longer in durations and lower in peaks than what’s
            seen on the AC mains.
           Telcordia waveforms stretch out to 1000us
           ITU (CCITT) and IEC use waves that are 700us long
           In the telecom world, it’s also assumed primary protection exists at the building entrance, so
            the voltages used for testing rarely exceed 2kV in the US; ITU makes different assumptions
            and goes up to 6kV and even 7kV for some products. .\.


Shared By: