Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Joseph Davis]

http://lists.contesting.com/_rfi/2004-10/msg00063.html

To:  rfi@contesting.com
Subject:  [RFI] COP Ignition systems
From:  wng@daimlerchrysler.com
Date:  Tue, 19 Oct 2004 11:24:27 -0400
List-post:  <mailto:rfi@contesting.com>

After some exhaustive research and testing here in Auburn Hills, we have
concluded that the main issue with COP Ignition systems is the inductive
coupling of primary switching noise  ( >10**7 A/sec) to everything in the
engine compartment and to body sheet metal.

Briefly, wiring design is often done by the wiring vendor for minimum
number of different part numbers in the assembly plant and minimum
material content.  This often leads to a situation that causes the
switched primary circuit to be a large loop, encompassing much of the
engine compartment.  This causes a fairly high efficiency coupling mode
especially to the inside surface of body sheet metal.  Since the net
current in the panels must add up to zero, an opposite current is induced
on the outside surface.  This current gives rise to a normal
(perpendicular to the surface) electric field that, in turn, couples very
efficiently to our antennas!

The best was we have found to remedy this  is to route the
ignition-switched-battery, which supplies the COP (and, incidentally,
injector) switching pulses to a large, low inductance (stacked film) 0.5uF
capacitor located AT THE GROUND POINT for the high-speed switches (the
engine controller ground);  then rerouting the feed to the engine IN THE
HARNESS containing the switched side of the coils and injectors.

This operation provides a near zero area loop for these nasty noise
currents, thus minimizing coupling to the rest of the world.  This is
usually enough to cure AM Broadcast reception issues.  It may not be
sufficient for the high sensitivity HF SSB receiver.  Additional measures,
such as a high u (type 77 or J) ferrite common mode choke on the engine
harness, where it leaves the engine for the controller (when mounted on
the body) keeps both primary and secondary and injector garbage on the
engine where it belongs (No, you can't get rid of it; you still need to
make sparks.).  In extreme cases, an RC snubber may need to be installed
in parallel with each coil primary.  The snubber needs to be the complex
conjugate impedance of the coil primary (220 ohms and 0.033uF comes to
mind for one of our COPs).

These remedies, while they are invasive, will not adversely affect
critical engine control systems (emissions, driveability and fuel economy)
IF implemented exactly as described.

Try Bytemark.com for high initial permeability ferrites and ferrite
technical information.

Bill Gilmore  WB8FPQ
Electromagnetic Compatibility Engineer
Core EMC, Electrical/Electronic Engineering
Powertrain Component EMC / SRT & HEV Platform EMC
(248) 576-5813


 

[dvst8r]

Sweet I can finally get my AM talk radio back... 

[Robb]

Ok, im an electrical retard.  They are just putting caps between the coil drivers and ground?

[dvst8r]

Ok, im an electrical retard.  They are just putting caps between the coil drivers and ground?

That is what i got from it.

[Conceptz-X]

and a choke...


Fuck it, Solid Core FTW if you dont have COP

[Joseph Davis]

Ok, im an electrical retard.  They are just putting caps between the coil drivers and ground?

That is what i got from it.

There's a strategic placement involved. 

Electrodynamics is wacky shit.  Electronics theory tends to exist in a dimensionless ideaspace, like some higher math where you are masturbating with numbers, but electrodynamics brings spatial orientation into play.  The manner in which magnetic fields form, the equal-but-opposed back EMF fields that are inducted into nearby metal/circuitry, is some voodoo shit on anything but the simplest level.  Just because the solution seems simple doesn't mean it isn't based on some complex shit.

[Robb]

Ok, im an electrical retard.  They are just putting caps between the coil drivers and ground?

That is what i got from it.

There's a strategic placement involved. 

Electrodynamics is wacky shit.  Electronics theory tends to exist in a dimensionless ideaspace, like some higher math where you are masturbating with numbers, but electrodynamics brings spatial orientation into play.  The manner in which magnetic fields form, the equal-but-opposed back EMF fields that are inducted into nearby metal/circuitry, is some voodoo shit on anything but the simplest level.  Just because the solution seems simple doesn't mean it isn't based on some complex shit.

I think I understand, the "low pressure area" that is the caps is the manner that prevents EM fields from "resonating" through anything that will accept them in the chassis???

[patsmx5]

Ok I wired up an ignition system every wrong way imaginable. So I finally figured out what you want to do. Basically, you want to put the ignition system on it's own ground plane. It doesn't share it with anything else. Ground the head to this plane, the ignitor, the coil, the 25uF cap, etc. All ignition related shit goes to this plane. Now when the coil fires the energy goes out the coil, to the plugs, sparks, then travels from the ground of the plug to the cylinder head, out of the wire you hooked to the head and back to the ground plane.

You ground this plane to the block. Now the ignition system is on it's own plane and doesn't interfere with other shit.

Like this: http://i231.photobucket.com/albums/ee155/patsmx5/wiring4-11-0915.jpg

[TTC]

I think ill stick with my regular non cop style ignition, this seems like there is to much math.

[Robb]

I think ill stick with my regular non cop style ignition, this seems like there is to much math.

When the coils cost less than wires do, im going with the cop thanks.

[TTC]

I agree, does COP really work that much better.  Most auto manufacturers offer it on their higher end cars.

[kgx]

Ok, im an electrical retard.  They are just putting caps between the coil drivers and ground?

i think what they're doing is installing the cap at the grounding point for the ECU itself, then re-routing the switched 12V feed line through the harness alongside the pulsed lines and out to the coils, rather than having the 12v feed splice into the harness from wherever the closest splice point is to get to a 12V junction.

feel free to correct me if i'm wrong, but i think it's something like this:


by decoupling the power source at the ECU, it takes the chassis out of the RF loop (since before, the chassis was the primary conductor between the battery and the ECU and was therefore on the same signal loop as the coils and subject to all the switching signals and CEMF spikes that go with it), and running the power supply alongside the switched signal lines to the coils makes the RF loop that IS created small and narrow, instead of encompassing the whole chassis.

i hope that's right!

[Joseph Davis]

A semester in high speed/low voltage electronics will show you this!

No, it won't, but a semester's worth of electrodynamics will.  Which is a graduate level course for most disciplines.  While the idea comes down to a filter cap, the reason for and placement of has absolutely nothing to do with mundane high speed/low voltage electronics.

Come on JD you've known this for ages!

What does that have to do with proliferation of said information?  Most people don't know this.

[Joseph Davis]

Samuel L jackson would never stare at me like that.  He'd stand at my shoulder and stare at you like that.