I have a 92 dodge dakota with the symptom that when the engine gets warm
it loses spark. I have the manual and have followed the diagnostic
procedure to the point where I need to test the output from the ECM. On
this model, pin 19 of the ECM controls the ground to the ignition coil
to fire the plugs at the proper time. The manual says something to
diagnose this that I can't understand. They say to make a little jumper
with a 33uf capacitor that is used to ground pin 19 and ground the coil.
When the ground is removed, if a spark is seen at the coil cable (I
assume they mean secondary), they say the ECM is bad.

Can someone explain the logic of what I'm trying to do. I understand
what the ECM is supposed to do, but I don't understand how to know if
it's doing it, even with this little jumper gizmo they describe. I'm
fairly electronics literate, but not a wiz.

There are no stored error codes at any point in the failure that I have
been able to observe. All help appreciated
thanks

Mark Leininger writes:

I have a 92 dodge dakota with the symptom that when the engine gets
warm it loses spark. I have the manual and have followed the
diagnostic procedure to the point where I need to test the output from
the ECM. On this model, pin 19 of the ECM controls the ground to the
ignition coil to fire the plugs at the proper time. The manual says
something to diagnose this that I can't understand. They say to make a
little jumper with a 33uf capacitor that is used to ground pin 19 and
ground the coil. When the ground is removed, if a spark is seen at the
coil cable (I assume they mean secondary), they say the ECM is bad.

Can someone explain the logic of what I'm trying to do. I understand
what the ECM is supposed to do, but I don't understand how to know if
it's doing it, even with this little jumper gizmo they describe. I'm
fairly electronics literate, but not a wiz.

There are no stored error codes at any point in the failure that I
have been able to observe. All help appreciated
thanks

You are basically pulsing the ignition coil by hand. It would seem that
the implication is that if there is a spark (I assume on a dummy plug),
then the ignition system is OK but the ECM is not pulsing it.

--- sam | Sci.Electronics.Repair FAQ Mirror: http://repairfaq.ece.drexel.edu/
Repair | Main Table of Contents: http://repairfaq.ece.drexel.edu/REPAIR/
+Lasers | Sam's Laser FAQ: http://repairfaq.ece.drexel.edu/sam/lasersam.htm
| Mirror Sites: http://repairfaq.ece.drexel.edu/REPAIR/F_mirror.html

Note: These links are hopefully temporary until we can sort out the excessive
traffic on Repairfaq.org.

Important: Anything sent to the email address in the message header is ignored.
To contact me, please use the feedback form on the S.E.R FAQ Web sites.

Take it from someone who knows how to troubleshoot, don't use the chart. I'm
not talking about incompetence here, but most of those charts were written
before there were any real world failures in that particular system.

Unless you have and know how to use a scope this could be difficult. I remember
the secondary crank sensor in my Dad's Olds was a bear to isolate for sure.

You're on a 92, so it has a computer and active control of everything. You need
to know what type of pickup it uses. Magnetic ones once bad usually stay bad,
on the other hand photointerrupters and Hall effect sensors are semiconductor
devices and are prone to thermal failures. Sometimes as the heat subsides, so
does the fault.

They say ECM usually they mean the computer, but not always. If you're
grounding the coil and looking for spark you are only testing the coil. If the
thing has a pickup, an ECM and a spark (power) module seperate they all need to
be tested. The thing that drives the coil is not the ECM unless the spark
module is integral to the ECM. Some manufacturers call it an ESC, while what
the ECM does is to process the inputs from the MAP and/or MAF, TPS, O2 or
Lambda, and whatever other sensors it may have. From this it determines how to
drive either the solenoid in the carburator (sic), or the duty cycle of the
fuel injector(s). The ECM also controls the spark module, in two ways.

1. The bypass line.
This is used for starting and will change states at anywhere from 150-350
RPM depending on a preset value. Once in run mode the computer (ECM) takes over
timing. It may advance or retard the pulses to the spark module or use a DC
voltage.

2. Timing.
As stated above, understand that it is not the spark module ignoring the
computer, it is the computer TELLING the spark module to ignore it. It doesn't
have any data yet. If it shows signs of life when cranking but dies right away
it could be something seriously screwing up the timing, but that is not the
first thing to check.

I don't know what they're talking about with this capacitor deal, but
disconnect the coil before you test it, screw their stooooopid method. Hook it
to a spark plug, get 12 volts and just "scratch" the wires to the primary.
Incedentally, the coil and the plug must both be firmly grounded. I'd put the
test plug in a jumper cable to ground, and as far as the coil goes, nowhere.
Leave it mounted and it'll still be grounded. Work smart.

Remember all you're testing is the coil, to say it's the ECM might be post hoc
ergo propter hoc, which means it followed ______ therefore it was caused by
______. (I didn't study Latin so this is the best I can do)

This is one of those old wisdomisms that warn not to jump to conclusions.

Since there are other components in the system other than the coil and the
"ECM" you cannot assume that since the coil is good tthe "ECM" is bad. In
another case, I'm not sure on your particular system, but if the bypass line is
pulled high by the "ECM" then you might just have a blown fuse.

However if what they call the "ECM" is in the distributor or under the coils it
may well be at fault. This is the power device, like the amp in your stereo, it
gets hot, and hotter yet where it is. Devices of this particular nature are
prone to those thermal problems and the book may well be right. It's only the
reasoning in question to me. I've learned not to take everything at face value.

There has to be a crank or cam position sensor somewhere. This needs to be
looked into before replacing more expensive components. Yes econimics IS a
factor. If it seems like the pussy way out, well so what.

Amsel Rothchild, one of the richest men in the world equated economies to
electronic formulae, mostly inductance and capacitance, while resistance and
conductance were different, the "reactive" functions of the formulae were (I
guess) successfully applied to reactions in the stock market etc.. He doesn't
work for me. Actually he considered banks and the governments they controlled
to be like transistors. Able to change their resistance or switch on and off.
The public, meaning the public stock exchange, personal saving, purchases and
investments by "consumers", he considered the reactive factor.

I'm not in that particular biz, I think all science is related, Boltz' law is
analoguous with Ohm's law. mechanics, a transformer is a lever, or a hydaulic
equation between the sizes of two different pistons, or which particular gears
are engaged in your transmission right now. (heads up) Yeah, people with their
laptop on in their car, driving. Of course they're not thinking about whether
to change the $8 part or the $80 part.

But I digress. Suffice it to say to try the cheaper part first if you can't be
sure. There are times when you got data, not pulses or a voltage to measure.

Yes, two components (usually) are what make the spark during cranking. Does it
have spark during cranking when it's hot ? (failing) Just because it won't
start. . . . . . . . If it doesn't have spark while cranking, find out if it's
a semiconductor type sensor.

If the sensor is a semiconductor (not a coil) I'll give it 60% chance of being
the problem. If there's no spark DURING THE FAILURE I'll give it 80%+.

JURB

On Sun, 17 Oct 2004 18:51:10 -0500, Mark Leininger
wrote:

I have a 92 dodge dakota with the symptom that when the engine gets warm
it loses spark. I have the manual and have followed the diagnostic
procedure to the point where I need to test the output from the ECM. On
this model, pin 19 of the ECM controls the ground to the ignition coil
to fire the plugs at the proper time. The manual says something to
diagnose this that I can't understand. They say to make a little jumper
with a 33uf capacitor that is used to ground pin 19 and ground the coil.
When the ground is removed, if a spark is seen at the coil cable (I
assume they mean secondary), they say the ECM is bad.

Can someone explain the logic of what I'm trying to do. I understand
what the ECM is supposed to do, but I don't understand how to know if
it's doing it, even with this little jumper gizmo they describe. I'm
fairly electronics literate, but not a wiz.

There are no stored error codes at any point in the failure that I have
been able to observe. All help appreciated
thanks


You're just manually triggering the coil to verify the wiring to the
coil and the coil itself aren't bad. Not a great test, really. An
oscope on on the ecm output would be your best way to look at this. A
multimeter could be useful if it showed duty cycle, but even ac
voltage would be an indicator that something is pulsing.

Personally, for intermittent problems when it's warm I would suspect a
bad coil. Next culprit would be the pickup sensor. Either a crank or
distributor. Not sure what your Dodge has. If the rpm guage is
dropping to zero when the engine is still turning, I would lean toward
a sensor problem.

The Jeep engines are notorious for bad crank position sensors causing
intermittent no spark problems when warm. No codes are usually set
when this happens.


-Chris

Thanks for all the thoughts. To elaborate, this model uses the engine
control module (there is no separate ignition module) to control the
coil firing by interrupting the ground. The ECM gets signals from a Hall
effect device (they call it a sync signal generator) in the distributor,
and a CPS. I don't believe the other signals to the ECM (MAP sensor,
coolant sensor) would cause it to not fire the coil at all, they would
just cause it to run really crappy.

I think I have proven that the coil can't be the problem because of the
following test. I jumpered the coil primary leads to the coil so that I
could put a meter across them. When running properly I read 14v at the
positive side compared to battery ground, and about 1v when measured to
coil ground, which gives an idea what fraction of time the ECM is
telling the coil to fire. In failure mode I measure 12v at the postiive
coil compared to battery ground and 0v compared to coil ground. This
tells me the coil is getting 12v but is not being grounded by the ECM,
so it never fires.

So I think I'm at the point where either something is preventing the ECM
from firing the coil, or the ECM itself is bad. I do have a scope, but
it's real dusty and I'd rather not take time to start relearning how to
use it. Is it possible with a meter to tell while the engine is cranking
(not running) if the correct signal is being output from the CPS? Could
I compare the output during failure mode with the output during working
mode and conclude if it's producing any output?

thanks

Mark Leininger wrote:

I have a 92 dodge dakota with the symptom that when the engine gets warm
it loses spark. I have the manual and have followed the diagnostic
procedure to the point where I need to test the output from the ECM. On
this model, pin 19 of the ECM controls the ground to the ignition coil
to fire the plugs at the proper time. The manual says something to
diagnose this that I can't understand. They say to make a little jumper
with a 33uf capacitor that is used to ground pin 19 and ground the coil.
When the ground is removed, if a spark is seen at the coil cable (I
assume they mean secondary), they say the ECM is bad.

Can someone explain the logic of what I'm trying to do. I understand
what the ECM is supposed to do, but I don't understand how to know if
it's doing it, even with this little jumper gizmo they describe. I'm
fairly electronics literate, but not a wiz.

There are no stored error codes at any point in the failure that I have
been able to observe. All help appreciated
thanks

Forget the manual. Touch a logic probe to the coil primary terminals
when the no start is present:
Key on-engine off: high both sides
Key on-engine crank: high one side, pull low pulsing other side
If no, either the trigger is missing to the 'puter or the 'puter is bad.
JR

Mark Leininger wrote:
I have a 92 dodge dakota with the symptom that when the engine gets warm
it loses spark. I have the manual and have followed the diagnostic
procedure to the point where I need to test the output from the ECM. On
this model, pin 19 of the ECM controls the ground to the ignition coil
to fire the plugs at the proper time. The manual says something to
diagnose this that I can't understand. They say to make a little jumper
with a 33uf capacitor that is used to ground pin 19 and ground the coil.
When the ground is removed, if a spark is seen at the coil cable (I
assume they mean secondary), they say the ECM is bad.

Can someone explain the logic of what I'm trying to do. I understand
what the ECM is supposed to do, but I don't understand how to know if
it's doing it, even with this little jumper gizmo they describe. I'm
fairly electronics literate, but not a wiz.

There are no stored error codes at any point in the failure that I have
been able to observe. All help appreciated
thanks

You could probably look for an ac voltage on the hall effect sensor,
but it wouldn't tell you if the signal was too weak. On the Jeeps I
mentioned, often cleaning the conectors to the sensor fixed the
problem.



On Mon, 18 Oct 2004 07:17:25 -0500, Mark Leininger
wrote:

Thanks for all the thoughts. To elaborate, this model uses the engine
control module (there is no separate ignition module) to control the
coil firing by interrupting the ground. The ECM gets signals from a Hall
effect device (they call it a sync signal generator) in the distributor,
and a CPS. I don't believe the other signals to the ECM (MAP sensor,
coolant sensor) would cause it to not fire the coil at all, they would
just cause it to run really crappy.

I think I have proven that the coil can't be the problem because of the
following test. I jumpered the coil primary leads to the coil so that I
could put a meter across them. When running properly I read 14v at the
positive side compared to battery ground, and about 1v when measured to
coil ground, which gives an idea what fraction of time the ECM is
telling the coil to fire. In failure mode I measure 12v at the postiive
coil compared to battery ground and 0v compared to coil ground. This
tells me the coil is getting 12v but is not being grounded by the ECM,
so it never fires.

So I think I'm at the point where either something is preventing the ECM
from firing the coil, or the ECM itself is bad. I do have a scope, but
it's real dusty and I'd rather not take time to start relearning how to
use it. Is it possible with a meter to tell while the engine is cranking
(not running) if the correct signal is being output from the CPS? Could
I compare the output during failure mode with the output during working
mode and conclude if it's producing any output?

thanks

Mark Leininger wrote:

I have a 92 dodge dakota with the symptom that when the engine gets warm
it loses spark. I have the manual and have followed the diagnostic
procedure to the point where I need to test the output from the ECM. On
this model, pin 19 of the ECM controls the ground to the ignition coil
to fire the plugs at the proper time. The manual says something to
diagnose this that I can't understand. They say to make a little jumper
with a 33uf capacitor that is used to ground pin 19 and ground the coil.
When the ground is removed, if a spark is seen at the coil cable (I
assume they mean secondary), they say the ECM is bad.

Can someone explain the logic of what I'm trying to do. I understand
what the ECM is supposed to do, but I don't understand how to know if
it's doing it, even with this little jumper gizmo they describe. I'm
fairly electronics literate, but not a wiz.

There are no stored error codes at any point in the failure that I have
been able to observe. All help appreciated
thanks

Change the hall effect device and you may get lucky. I've seen this cause
this type of problem with Chryslers before. I'd also suggest you post this
in rec.auto.makers.chrysler as there are many people there that can provide
advice from their experience.

Bob

"Mark Leininger" wrote in message
...
Thanks for all the thoughts. To elaborate, this model uses the engine
control module (there is no separate ignition module) to control the
coil firing by interrupting the ground. The ECM gets signals from a Hall
effect device (they call it a sync signal generator) in the distributor,
and a CPS.

I posted on alt.auto.trucks.dodge but have found responses here to be
more helpful in the analytic process. The truck forum was useful for
finding all the common things that other people had experienced. I'll
take another wack at it this weekend.
thanks all


Bob Shuman wrote:
Change the hall effect device and you may get lucky. I've seen this cause
this type of problem with Chryslers before. I'd also suggest you post this
in rec.auto.makers.chrysler as there are many people there that can provide
advice from their experience.

Bob

"Mark Leininger" wrote in message
...

Thanks for all the thoughts. To elaborate, this model uses the engine
control module (there is no separate ignition module) to control the
coil firing by interrupting the ground. The ECM gets signals from a Hall
effect device (they call it a sync signal generator) in the distributor,
and a CPS.