Date: Thu, 25 Jun 1998 23:27:46 -0600
Reply-To: kenstich <kenstich@BEWELLNET.COM>
Sender: Vanagon Mailing List <Vanagon@vanagon.com>
From: kenstich <kenstich@BEWELLNET.COM>
Organization: Central Intelligence
Subject: Re: digifant test adaptor improvement suggestions =>> Long
Content-Type: text/plain; charset=us-ascii
Darrell et all,
Very Cool! -> I've fought with my L-Jetronic system trying to troubleshoot
various problems. And, since it has no built-in diagnostics or test points -
*^)(#$!#^&*%, I've ended up exposing the ECU connector contacts by removing
the backshell an pin probing the contacts -> very kludgey at best - I'm very
impressed by your homegrown test adaptor. I'm considering adapting your idea
to my system. Being an aerospace engineer, somewhat familiar with test
equipment and harness design, I thought I'd suggest some potential
improvements and cautions:
1) Suggestion - Certainly, at least the power pins ought to be protected from
shorts to ground by wiring in series resistors (in-line) to prevent shorts
from having a devastating effect. A 1 thousand ohm (1K) or 10K resistor is
about right. With a 1K resistor, the max current from a short is around 14
millliamps (approx 14 volts across 1K - ohms law applies) which will protect
your 26awg wire (and internal Digifant circuits!) nicely. The resistors should
be wired in inside the ECU as one of the failure modes is wire chafing through
the insulation where it penetrates the case. These resistors are called
current limiting or test point protection resistors. In the aerospace
industry, we protect virtually all test points and still manage to fry some
hardware. How much is a replacement Digifant ??? Resistors are cheap - OBTW -
use resistor packs - usually 4 or more resistors in a single tiny plastic
package - resistor power dissipation rating can be very low (i.e. 1/4 watt for
1K & 1/8 watt for 10K resistances) as power dissipation is low.
2) Suggestion - The wiring inside the case is subject to long term
chafing/abrasion (due to vibration) and should be tack bonded "glued" to the
printed wiring board every couple inches. The aerospace industry uses epoxy
compounds - Radio Shack must have something for this purpose. This is also
particularly applicable where the ribbon cable penetrates the ECU case.
3) Caution - One needs to use a DIGITAL multimeter on the test points to
prevent excessive loading of the test points when measuring voltage. Digital
meters have around 10 Megohm of input impedance resulting in virtually no
loading of the circuit being measured. Digital circuits (i.e. injector control
outputs) really need a oscilliscope or digital test equipment to "see" the
signal. Some of the fancier digital multimeters have pulse width duty cycle
measurement - way cool. One exception is the power leads, which instead of a
LED indicator light, an analog voltage meter would be way cool =>> actually, a
6-way rotary switch with one meter would allow one to see any of the power
voltages selectively => a few bucks more than the 6 led's.
4) Alternate design idea - If one could find mating male and female
connectors, one could construct a "Tee" harness that one plugs into the
ECU/Digifant and the vehicle harness that tee's off to a connector to provide
test points while passing the signals/power through to the ECU. This is the
usual method in the aerospace industry (we call them break-out boxes - the
slang term is "sandwich box"). It has the advantage that it can be temporarily
plugged in for troubleshooting and is not limited to one vehicle.
5) Question - Do you run the ribbon cable up to the driver area/dash so that
you can see the measurements while driving ??
Many thanks for sharing your ideas - looks like a winner - Good
explanation/illustrations in your web pages!
Thanks,
Ken Stich
80 Vanagon-L
57 Vintage Oval Bug
85 Prelude
Conifer, CO