From: Ken Cantrell ([email protected])
Date: Fri Nov 03 2000 - 09:54:07 PST
Jeff,
Do you have a radio in the unit also? The situation changes dramatically
when you have radios in little plastic boxes.
Zeners are too slow. You need a fast TVS that can respond in <=100 ps to
stop precursor events. There are a number of vendors that make these parts.
Keep the TVS capacitance to less than 200 pF. Lower the better. Placement
is critical. You may already have a solution and don't know it. The fast
TVS should be mounted as close as possible to the charge pins, with as much
surface are between the pins and the TVS as possible (3:1 width/length ratio
minimum if the enclosure allows that). If it's greater than 250 mils, it
probably won't eliminate the problem because of the . Post battery, feed
thru caps are also very good since you can achieve attenuations in the 50dB
range. These are most effective for DC applications (in series with the pos
batt terminal). These should be stuffed on the board. The ferrite clamp
around the ribbon cable only reduced the conducted current problem (common
mode), which indicates that you also have radiated coupling.
An alternative to direct contact discharge, and the first rule of thumb for
plastic battery powered boxes is: don't let it get in. Recess the pins 125
to 250 mils so that the user can't touch them with his/her finger which will
change the requirement to direct air discharge. The recess has to be deep
enough so that air arcing won't occur at the highest level of test. This is
usually in the above mentioned range.
Many times you have to employ all of the strategies mentioned, fast TVS,
ferrite clamp, in line ferrite for particulary troublesome lines (will only
work for a conducted problem), and filtering capacitance. If this doesn't
get you there, an often overlooked solution (or another piece of the
solution) is to simply increase the contact surface area inside the box
(copper tape)at the pin connectors (for engineering eval/proof of concept)
or metalization. Metalization has come a long way over the last 5 years,
and is now cost effective and reliable. Talk with your Mechanical Engineers
on the latest methods.
If you haven't metalized the inside of the enclosure and/or installed an EMI
gasket for the mating halves, these are also effective options. Pay special
attention to the mechanical mating of the two halves. The metalization
cannot have a bleed through path to the outside world. Empircally confirm
the resistivity of the metalization, both on the first articles and
subsequent lots. Process control at the plating vendor is critical.
Confirm the resistivity of the housing plastic also. Make sure that there
is not a graphite element in the plastic formulation.
Ken
-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of Jeff Reeve
Sent: Thursday, November 02, 2000 6:25 PM
To: [email protected]
Subject: [SI-LIST] : Unit failing ESD testing
SI List,
ESD related situation (maybe due to EMI resulting from ESD?). I have a
battery powered PCB surrounded by a plastic enclosure, on the rear of the
enclosure are several pins that are used to charge the battery when the
device is placed into a charger (similar to a cell phone). Contact points
are 12V and GND on the rear of the unit. I would like to operate up to 8kV
contact discharge to the unit (at the GND/12V contact) before the device
exhibits failures. Presently, the device functions up to 4kV and then begins
to exhibit failures (it resets itself and continues to operate). I have
tried a very fast Zener across 12V and it did slightly improve things but
not enough. There is a 1.5" ribbon cable from the battery PCB to the Digital
PCB. I clamped a ferrite across the cable and that also seemed to help (good
to 6kV) but that too was not enough. Maybe I have not given enough detail?
But I ask the question: What are common means to improve immunity to ESD
contact discharge? Perhaps someone is willing to share their experiences or
pointers to papers?
Many thanks in advance for your sharing...this is a awesome list!
Jeff
**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected]. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****
**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected]. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****
This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:29:59 PDT