Subject: Coil mount
From: Paul
Date: 10/24/2000, 7:43 AM


Finn Lassen wrote:

I had one of my coils (stock)  mounted on the engine mount plate until
Tracy
pointed out that the cooler you mount them the better (electronics
fails a lot
faster when ambient temp goes up). Don't know if that's an issue here,
but thought
I mention it.


I cannot speak for your specific components, but for electronic (as
opposed to
electrical) components in general, it is the thermal *cycles* more than
the maximum
operating temperature that kills them.  For example, Texas Instruments
commercial grade
(as opposed to military grade) products are tested to 85 degC (although
the data sheets
frequently list 50 or 70 degC as the upper spec limit).  The common
reliability test
(for new product designs) is called the 85/85; that's 85 degC at 85%
relative humidity,
and the devices are tested after 100, 250, 500, and 1000 hours.  No
devices in a new
design should fail under these conditions; if any do then we have to
re-work the design.

Thermal cycles induce mechanical stress on the subcomponents of the
electronics and
assist other phenomena like electro-migration and ion-migration.  This
is because the
lead frame (a copper-nickel alloy), the mounting epoxy, the chip (a
silicon-boron
alloy), the metalization layers (pure aluminum, aluminum-copper, or pure
copper), the
mold compound (a plastic confection), and the bond wires (pure gold) all
have different
coefficients of thermal expansion.

Russell Kent
Member, Group Technical Staff
Texas Instruments, Inc.


Good points by Al and Russell on this subject.  My figure of 135F is a
conservative value.  But keep in mind that this was an ambient air spec.
The devices themselves (specifically the pass transistors in the ignitors)
will be hotter than this because they disipate their own heat.  A generally
accepted figure for silicon device lifespan (not counting infant
mortalities) is that life is doubled for each 10 deg.C reduction in
temperature.   Fortunately, the inherent life of silicon devices is very
long, even at 85 deg.C.

Tracy Crook
rws@altavista.net
www.rotaryaviation.com

 
The AirCraft Rotary Engine NewsLetter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org