95 Camry Ignition problem hard to start

Scott Dorsey wrote:
>
> Mark > wrote:
> >FYI If you ever test a coil on the bench like this by pulsing 12V into
> >the primary, it was interesting to see that it did not develop a good
> >spark unless I also connected a capacitor on the primary.
>
> This is the coolest damn thing. If you actually look at the waveform on
> a scope, it's a huge damped sine. The coil has got a ferrite core, so it
> really is only effective at high frequencies. You pulse it, and only some
> of the high frequency components of that square wave actually make it through
> the coil.

It sounds like he was using a power supply and not a battery for his
bench test. It should work a little different with a battery.


>
> BUT... The coil inductance combined with the shunt cap make a resonant
> circuit and the resulting damped sine wave makes it through the coil very
> effectively.

Well no not exactly. When its working correctly the spark plug will fire
before it gets 1/4 of the way through the first period of the sinusoid.
If you are seeing something that looks like a sine wave that indicates
the high voltage is not making it to the spark plug. At any rate if he
was looking at the wave form on an o-scope he should have seen the
misfires. It's possible he may have been interpreting the waveform
produced by misfires as a bad dwell or timing issue.

If there is one single thing and O-scope is good for finding, it is the
very problem he says he had.

Pulling a plug wire, sticking a spare plug into it, holding
the plug body to a handy ground, and operating the starter will almost
never reveal this problem. That is because it takes much more voltage
to fire the plug inside the engine when cranking then it does to fire
the plug when it is outside the engine resting against the block or
other ground.


-jim

>
> Kettering figured this whole thing out and it's a hell of an ingenious
> idea.
> --scott
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

At any rate if he
> was looking at the wave form on an o-scope he should have seen the
> misfires. It's possible he may have been interpreting the waveform
> produced by misfires as a bad dwell or timing issue.
>
> If there is one single thing and O-scope is good for finding, it is the
> very problem he says he had.
>
>

Well I'm happy the car is working but I have been wondering about this
very question.

My conclusion (as Tegger is correctly needeling me about) is I don't
think a scope is very good at detecting this particular type of mis-
fire.

When the mis fires happened casued by the carbon track, there was
still a spark it's just that the spark happens at the coil instead of
at the plug. The scope is coupled to the wires which still sees the
almost the same volatge waveform. The scope can't tell which end of
he wire is sparking. My thoughts are that the waveform isn't really
that different, there is still a spark in either case.

Remember, the carbon track didn't actually short or resistivly load
the secondary, it just provied another spark gap and the spark would
jump at that gap instead of at the the plug gap, occasionaly.

But Tegger is right, this is a lesson in starting with the basics..

Mark

Mark wrote:
>
> At any rate if he
> ? was looking at the wave form on an o-scope he should have seen the
> ? misfires. It's possible he may have been interpreting the waveform
> ? produced by misfires as a bad dwell or timing issue.
> ?
> ? If there is one single thing and O-scope is good for finding, it is the
> ? very problem he says he had.
> ?
> ?
>
> Well I'm happy the car is working but I have been wondering about this
> very question.
>
> My conclusion (as Tegger is correctly needeling me about) is I don't
> think a scope is very good at detecting this particular type of mis-
> fire.

Neither one of you know what your talking about.


>
> When the mis fires happened casued by the carbon track, there was
> still a spark it's just that the spark happens at the coil instead of
> at the plug.

Yes but it is going to appear as a different waveform on the scope. Heck
just changing the spark plug gap by a couple thou will change the
waveform. For one thing when it fires at the coil instead of the spark
plug it doesn't have the resistance of the spark plug wires and spark
plug.


> The scope is coupled to the wires which still sees the
> almost the same volatge waveform. The scope can't tell which end of
> he wire is sparking. My thoughts are that the waveform isn't really
> that different, there is still a spark in either case.

You are looking at the primary waveform. that is a little different than
the secondary waveform but you should still be able to see everything
you need to diagnose.

But you need to know what the wave form is supposed to look like.


>
> Remember, the carbon track didn't actually short or resistivly load
> the secondary, it just provied another spark gap and the spark would
> jump at that gap instead of at the the plug gap, occasionaly.

But it only jumps that gap when the voltage exceeds some high threshold.
If it always arced at that breach in the coil you would always have no
spark to any of the plugs.

>
> But Tegger is right, this is a lesson in starting with the basics..

There are ways to find that fault without an O-scope. But the procedure
as he described is not going to work, because it requires much less
voltage to fire the plug when it is sitting on the block at atmospheric
pressure. The voltage will never exceed the threshold needed to take
that path.

-jim

>
> Mark

>
>
> > When the mis fires happened casued by the carbon track, there was
> > still a spark it's just that the spark happens *at the coil instead of
> > at the plug.
>
> Yes but it is going to appear as a different waveform on the scope. Heck
> just changing the spark plug gap by a couple thou will change the
> waveform. For one thing when it fires at the coil instead of the spark
> plug it doesn't have the resistance of the spark plug wires and spark
> plug.
>
> > The scope is coupled to the wires which still sees the
> > almost the same volatge waveform. *The scope can't tell which end of
> > he wire is sparking. *My thoughts are that the waveform isn't really
> > that different, there is still a spark in either case.
>
> You are looking at the primary waveform. that is a little different than
> the secondary waveform but you should still be able to see everything
> you need to diagnose.
>
> But you need to know what the wave form is supposed to look like.
>
>

The scope did show a small difference in the waveform when there was a
mis-fire, but it was a subtle difference and at the time (not knowing
what the answer is) I couldn't tell if that differnce was the cause or
an effect.

If for example, the misfire were due to a fuel problem (or timing
which is what I was stuck on), I would also expect it to effect the
waveform. So just becasue you see a subtle change in the waveform
during a mis-fire that doesn't point to the casue of the change. The
waveform durig the mis fire did NOT look like a weak spark.

All in all it was an educational experience.

Mark

Mark wrote:
>
> >
> >
> > > When the mis fires happened casued by the carbon track, there was
> > > still a spark it's just that the spark happens at the coil instead of
> > > at the plug.
> >
> > Yes but it is going to appear as a different waveform on the scope. Heck
> > just changing the spark plug gap by a couple thou will change the
> > waveform. For one thing when it fires at the coil instead of the spark
> > plug it doesn't have the resistance of the spark plug wires and spark
> > plug.
> >
> > > The scope is coupled to the wires which still sees the
> > > almost the same volatge waveform. The scope can't tell which end of
> > > he wire is sparking. My thoughts are that the waveform isn't really
> > > that different, there is still a spark in either case.
> >
> > You are looking at the primary waveform. that is a little different than
> > the secondary waveform but you should still be able to see everything
> > you need to diagnose.
> >
> > But you need to know what the wave form is supposed to look like.
> >
> >
>
> The scope did show a small difference in the waveform when there was a
> mis-fire, but it was a subtle difference and at the time (not knowing
> what the answer is) I couldn't tell if that differnce was the cause or
> an effect.
>

You should look at the secondary for this sort of diagnosis (probably
best to get or make a inductive pick-up). What you would have seen
looking at the secondary voltage would be a lot less subtle.



> If for example, the misfire were due to a fuel problem (or timing
> which is what I was stuck on), I would also expect it to effect the
> waveform. So just becasue you see a subtle change in the waveform
> during a mis-fire that doesn't point to the casue of the change.

The whole purpose of using a o-scope is to view and interpret the subtle
changes in waveform. From what I gathered you were focusing on the dwell
part of the waveform. The type of problem you had should have caused no
change in that part of the waveform.


> The
> waveform durig the mis fire did NOT look like a weak spark.

From what you described it wasn't a weak spark. It was a spark that
followed a different path when the secondary voltage exceeded a certain
level. I would expect it would misfire when you punched the gas pedal.
Usually the spark following a carbon trail gets worst when its damp and
humid and under heavy load and not when the engine is hot, but I suppose
if there was a crack in the coil insulation it could open up when its
hot.

If your spark plug wires or spark plugs are old you might want to make
sure they weren't the root cause of the problem. If the coil is
constantly forced to produce more voltage than it was designed for it
can lead to breakdown of the insulation.

To test the coil voltage capability you need to force it to jump a wider
gap than usual. As a rule of thumb the coil needs to be able to produce
enough voltage to jump a gap equal to the ordinary spark plug gap times
the compression ratio. So if the spark plug gap is .04" and the CR is
10 the coil need to be able to produce enough voltage to jump a gap
(under normal atmospheric pressure) of 0.4" However don't force the coil
voltage to go any higher than that (by jumping a larger gap) or you may
cause the same problem again.


-jim


>
> All in all it was an educational experience.
>
> Mark

>
> To test the coil voltage capability you need to force it to jump a wider
> gap than usual. As a rule of thumb the coil needs to be able to produce
> enough voltage to jump a gap equal to the ordinary spark plug gap times
> the compression ratio. *So if the spark plug gap is .04" and the CR is
> 10 the coil need to be able to produce enough voltage to jump a gap
> (under normal atmospheric pressure) of 0.4" However don't force the coil
> voltage to go any higher than that (by jumping a larger gap) or you may
> cause the same problem again.
>
> -jim
>
>

thats good info, thanks..

I did play with the bad coil on the bench, with a capacitor on the
primary and pulsing 12V current limited to 5A into the primary, it
would easily arc to the carbon track with no gap connected. If i
connected a 0.04 gap, it would never jump to the carbon track,... but
you just gave me the reason, under compression the gap is in effect
0.4"

So coils should be tested on the bench with a 0.4 to 0.5 inch gap..

thanks

I did check and regap the plugs and check the wires.
car has been running perfectly with the new coil, even in the rain..

Mark

Mark wrote:
>
> >
> > To test the coil voltage capability you need to force it to jump a wider
> > gap than usual. As a rule of thumb the coil needs to be able to produce
> > enough voltage to jump a gap equal to the ordinary spark plug gap times
> > the compression ratio. So if the spark plug gap is .04" and the CR is
> > 10 the coil need to be able to produce enough voltage to jump a gap
> > (under normal atmospheric pressure) of 0.4" However don't force the coil
> > voltage to go any higher than that (by jumping a larger gap) or you may
> > cause the same problem again.
> >
> > -jim
> >
> >
>
> thats good info, thanks..
>
> I did play with the bad coil on the bench, with a capacitor on the
> primary and pulsing 12V current limited to 5A into the primary, it
> would easily arc to the carbon track with no gap connected. If i
> connected a 0.04 gap, it would never jump to the carbon track,... but
> you just gave me the reason, under compression the gap is in effect
> 0.4"

Yes that's the general idea.

Bear in mind that with no gap connected you are pushing the voltage
higher than you really need to test and that has the potential of
creating the very problem you had.

Of course a closed throttle means lower cylinder pressures. It is a
little surprising to me that you never mentioned that the car had a
tendency to stumble or stall when you tried to accelerate hard from full
stop. Or maybe you never do that.




>
> So coils should be tested on the bench with a 0.4 to 0.5 inch gap.

Doesn't really need to be "on the bench"

-jim

>
> thanks
>
> I did check and regap the plugs and check the wires.
> car has been running perfectly with the new coil, even in the rain.


..
>
> Mark

While that is a good rule of thumb, those coils are capable of
throwing a spark a spark
considerably farther.
Ben

> As a rule of thumb the coil needs to be able to produce
> enough voltage to jump a gap equal to the ordinary spark plug gap times
> the compression ratio. *So if the spark plug gap is .04" and the CR is
> 10 the coil need to be able to produce enough voltage to jump a gap
> (under normal atmospheric pressure) of 0.4"

ben91932 wrote:
>
> While that is a good rule of thumb, those coils are capable of
> throwing a spark a spark
> considerably farther.

Sure. But driving the coil to excessively high voltages may cause damage
to the insulation. There is really not much point in testing for a lot
more voltage then the car needs.

-jim

> Ben
>
> > As a rule of thumb the coil needs to be able to produce
> > enough voltage to jump a gap equal to the ordinary spark plug gap times
> > the compression ratio. So if the spark plug gap is .04" and the CR is
> > 10 the coil need to be able to produce enough voltage to jump a gap
> > (under normal atmospheric pressure) of 0.4"

> *Of course a closed throttle means lower cylinder pressures. It is a
> little surprising to me that you never mentioned that the car had a
> tendency to stumble or stall when you tried to accelerate hard from full
> stop. Or maybe you never do that.
>

yeah I was thinking about that too. I didn't notice any miss under
power at first, after a day or two the miss at idle got worse and the
miss a power became very obvious..

Lesson leaned is that spark problems can show at start, idle and under
power.

also I was careful not to pulse the new coil on the bench without a
gap connected the secondary.

I didn't care about more damage to the old coil, it was already
defective, and I wanted to SEE the arc jump the carbon track to assure
myself this was indeed the problem.

Mark