‘89 CRX won’t start - frustration

It doesn't work that way. It fails on or off when the car sits for that
long. Especially, parking overnight the solder joint normally restores
itself. In fact, the OP hears the 2 second after the ignition is on.

 

Why not just spray some starter fluid into the air intake to see if it
wants to start?
If it then wants to start, you know it most likely is not air or
ignition..

 

Three quick checks I'd do on old OEM wires: (1) Run the car
in the dark and look for sparks coming from the wires. (2)
Run the car at idle with the hood up. Spray the wires with
misting water. Listen for changes in the idle speed. That's
a tip off (according to Tegger's site) that they need
replacement. Seems reasonable to me. (3) Check the wire
resistance. Should be under 15,000 ohms each.

Old wires are said to be harder on the ignition coil, too.
So it pays to replace them (with new OEM ones).

You sound plenty expert in your automotive knowledge. I hope
you nail this one, soon! Should be good.

 

That's a good idea, except the car won't run...

 

Oops.

 

UPDATE
Thank you very much for your support. Sometimes just knowing that somebody
else is just as puzzled gives you energy to keep investigating…
Today the car did finally start but unfortunately without revealing any
clues as to what the problem is, so I’m sure that this no start problem
will reappear sooner or later.

Here is what I did today Monday in the order I did it:

1. First I tried to crank the engine without doing anything, just in case

Cranked engine - No start.

2. Then CLEANED SPARK PLUGS: I took out all 4 spark plugs and cleaned them
with wire brush and adjusted the gap (the gap had increased to about 1.2mm,
I reset it to 0.9mm). The plugs did not seem fouled with oil or gasoline
although there was a faint smell of gasoline on them, normal I assume
although I have not tried to debug such problems before. Plug electrode
appearance looked normal to me just a light gray very few deposits which I
removed with wire brush. Cleaned the wire ends and spark plug hole and
reinstalled the spark plugs.
Cranked engine – NO START – perhaps an occasional very weak firing as
usual.

3. Then SPRAYED STARTER FLUID INTO INTAKE MANIFOLD: Took out the air hose
at the intake manifold, opened the throttle butterfly to make sure the
started fluid would go in, and then sprayed starter fluid directly into
the intake manifold for 3-4 seconds. I could not see inside the intake
manifold but I opened the throttle butterfly and inserted a screwdriver
into it, to make sure there were no other obstructions, all was fine.
Cranked engine - NO START - perhaps an occasional very weak firing as
usual.

4. REPEATED SPRAYING STARTER FLUID INTO INTAKE for another 3-4 seconds
just to make sure,
Cranked engine - NO START -perhaps an occasional very weak firing as
usual.

5. CLEANED THE DISTRIBUTOR: Since I had already checked for spark a couple
of days ago, I removed the plastic black distributor housing, it did not
seem dirty inside but cleaned it anyway with some alcohol.
Cranked engine - NO START -perhaps an occasional very weak firing as
usual.

6. INSISTED CRANKING: In frustration, since I was occasionally hearing
weak firing I insisted cranking the engine. Whenever I fully depressed the
accelerator there seemed to be more firing. After two 20second cranks it
seemed that the firing had increased. The firing seemed smooth (that is,
it did not seem that just a few cylinders were firing it just seemed like
weak uniform firing but I could be wrong). At the end of the third 20
second crank it started firing more and I kept cranking more until the
engine finally RELUCTANTLY STARTED somewhat erratically first for about 5
seconds and then normal. Quite a bit of smoke came out of the exhaust but
that could have been the started fluid. Once started the high idle was
normal, engine noise normal, revving up normal, no smoke from exhaust. I
let it idle for a couple of minutes and then turned it off.
I waited 1 minute, re-cranked and it started normally.
Turned engine off. Waited 15 minutes cranked- started normally.
Took car for test drive all was normal. Accelerated at full throttle going
35mph on a steep hill power was normal no missing or other signs of fuel
starvation.
Turned engine off, waited 30 minutes cranked – started normally.
Turned engine off, waited 1 hour cranked – started normally.

With the engine idling I did the SPRAY BOTTLE TEST sprayed spark plug
wires and distributor until they were drenched, no change in idling or
other engine behavior.

With the engine idling I slowly removed each one of the spark plug wires
until the cylinder I was unplugging stopped firing. I noticed that the
spark must be STRONG because it seemed I had to remove the wire by about ½
inch from the spark plug before any one cylinder stopped firing (although
the engine was able to kept idling on 3 cylinders).

The reluctant slow way in which the engine started and the smoke from the
exhaust would point to flooding/fouling but, as I described in step 2, I
had taken out all 4 spark plugs and they seemed clean, besides I cleaned
them thoroughly in step 2 and it still would not start.

Meanwhile throughout this process I could always hear the fuel pump
running for 2 seconds as soon as I turned the ignition.
As I had said in my original posting I had already tested the fuel
pressure last Thursday and I had also tested for spark.

Soooo… I am puzzled. I’m sure the problem will return sooner or later and
I’m running out of diagnostic tests. You are probably starting to think
that I am making all this up to puzzle you…
I will take out the main relay and re-solder it, seems like an easy job,
not much to loose by doing it just in case…

 

For future reference, doesn't this risk overloading the
ignition coil or something?

See
http://www.tegger.com/hondafaq/startproblems.html#checkspark
.. Plus some guy a few months ago did something like you
describe and burned out his ignition coil.

Right, I'm not confident it's the main relay, but if you've
really got the time, eliminate it. Unfortunately it's a bit
of a bear to get out. Most folks say leave the casing
bracketed to the wall and just slide out the guts.

Certainly seems like the symptoms point to some odd flooding
condition. Did you dump a bottle of Chevron Techron into the
near empty fuel tank (and then fill the tank) yet?

Be interesting to hear whether the car starts tomorrow...

 

Yes, I was a little concerned about the coil but compared to this
frustrating problem changing the coil did not seem that bad. Besides, as I
tried this on the first cylinder I noticed that even when the wire is
pulled away from the spark, the spark was still discharging from the wire
end to the engine (inside the spark pug hole). The bad thing would be if
the spark had absolutely nowhere to discharge and discharged through the
coil insulation inside the coil. Had that been the case I would have had
second thoughts about repeating the procedure on all cylinders.

 

I understand. You ensured you had a ground. That's the key.

 

Now that you've gotten it to start, it occurred to me that I've seen
something like this on a Subaru once where the thermosensor was
defective and adjusted the mixture such that the car just would not
start.
Subarus get into this mode after starting repeatedly where the mixture
goes to a mid range for debug purposes (one can also short two pins
together).

It does the exact same thing you are describing: not starting for a
long time and then suddenly starting. After the car cools down, the
problem would re-appear.

This information was just "for what it is worth" - Subies are quite
different than Hondas but just for giggles check the thermosensor...

Remco

 

something like this on a Subaru once where the thermosensor was
defective and adjusted the mixture such that the car just would not
start.
Subarus get into this mode after starting repeatedly where the mixture
goes to a mid range for debug purposes (one can also short two pins
together).<<<<<<

ThomasE:

Indeed, when this no start problem appeared last year as a rare
intermittent problem I was also having an intermittent code 06 “Bad
temperature sensor code” but the two did not seem related. The 06 code
would cause the check engine light to go on and would have some effect on
the idle (idle was somewhat erratic) and gas consumption which makes sense
because if the computer cannot tell the engine temperature it cannot
regulate well the fuel mixture. I tested the temperature sensor resistance
with engine cold and engine hot and it was within spec. I also tested for
electrical continuity and did not find anything wrong. However after
getting the 06 code 3-4 times in the span of 2 months and suspecting it
may have something to do with the no-start problem I decided to replace
the thermo sensor and indeed the 06 code never appeared again. At that
time, when I got the no-start problem, I did try to disconnect the thermo
sensor just to make sure that the computer got the message that the thermo
sensor was bad and thus force the computer into default operation and at
least give me a mixture that would start the car but the car again would
not start. Once the engine started, disconnecting the thermo sensor
resulted in an 06 code so the computer seemed to be detecting it
correctly.

In any case the 06 code problem went away since I replaced the thermo
sensor about 9 months ago. Besides, in this morning’s tests, if the
mixture were too low wouldn’t the engine do something at least for a
couple of seconds once I sprayed starter fluid into the intake manifold?
Also this intermittent no start problem does not seem to be related to
whether the engine is hot or cold. The no start problem has manifested
itself say 6-8 times total, sometimes with warm engine sometimes with cold
engine sometimes with warm weather sometimes with cold, I cannot find any
correlation whatsoever and I consider the 06 temp sensor problem cured
since I replaced the thermo sensor 9 months ago. There are no codes
related to the no-start problem any more.

 

<SNIP>

The puff of smoke proves it was totally flooded. You've got an injector
that's dripping overnight. It empties the contents of your pressurized
fuel rail into the engine. Holding the pedal to the floor and cranking
it is the (only easy) way to get it going. The cure is to change the
injector or to get it un-gunked with Techron or a few tanks of good
Gasohol.

Your owner's manual tells you how to start a flooded Honda.
Pedal-to-the-metal and crank it.

You can use google groups to confirm.

'Curly'

 

listen, idiot, sparking is a function of air pressure and voltage. at
atmospheric pressure, the gap is too small to reliably spark, for
reasons i'll spare you. [look it up if you dare]. /so/, as i stated in
the part you so helpfully snipped, take an /old/ plug and bend the
electrode. then you get a big fat [visible] spark if the ignition is
functioning properly. that's why ignorant automotive amateurs like
snap-on make tools like this:
http://buy1.snapon.com/catalog/item.asp?P65=&tool=all&item_ID=11293&group_ID=1470&store=snapon-store&dir=catalog

and you hold onto it while testing too, right?

duh. wtf bother testing upstream if you don't need to? if it works
downstream first, the upstream test is entirely pointless and your time
is wasted.

but then again, i'm wasting my time arguing with an idiot, so who am i
to talk.

you're killfiled. bye.

 

and check the oil for gas dilution smell!

 

Not sure if the car would attempt to start.. The subie definitely did
not want to start so it definitely did something to the air/fuel
mixture or timing.
It wasn't so much that it needed to cool down for the problem to
re-occur, but almost seemed time related as if the ECU had a timer
built in. Of course, this is a subaru - totally different car.

Well, thought I'd bring it up as a possible, just in case.
Carry on..

 

The CRX started again this evening after sitting for about 5 hours.

The injector flooding theory makes sense except that when I took out all 4
plugs they did not seem fouled. But then again I have not done this many
times so I don’t know how the spark plugs are supposed to look.

I’ll make the injector flooding theory the most likely one and wait for
the next occurrence. So next time it happens I’ll keep a set of 4 new
spark plugs and when the car does not start again I’ll put in the 4 new
plugs and it should start. I’m thinking that no matter how flooded it is
placing 4 new dry spark plugs should start the combustion that burns off
the flooding (or is it not that the case?).

Also if it is indeed flooding I must have more than one injector that is
leaking since the engine should be able to start even on 3 cylinders (am I
right on this one?)

 

I wanted to thank you all for all your replies. Your suggestions are
helping me get closer to the problem and learn about other potential
problems, like the fact that even if my main relay is ok for now I should
sooner or later expect problems from it on a CRX that old, unless I
resolder it which I may do anyway in the next few days.

And next time I’ll know better how to test for spark. Being a physicist I
should have figured out that the spark is more difficult under compression
rather than at atmospheric pressure. It wasn’t until it was pointed out
that it clicked…because sparking in a gas is the ionization of gas
molecules. When a charged particle inside the gas (an electron I assume)
gains enough speed in the electric field between the electrodes to hit
another gas molecule and make it ionioze then there’s a chain reaction
which leads to the spark, much like lightning during a storm. So as the
air gets thinner the gas molecules are further apart and the charged
particle travels more distance and thus gains more speed, on average,
before colliding with another gas molecule. The more violent the collision
the more likely that the gas molecule that is hit will ionize giving out
another electron. Now there’s two electrons being accelerated which now
hit two new molecules etc. and thus the chain reaction…

Thanks again to all of you. I’m sure I’ll repost this no start puzzle
sooner or later…

-ThomasE-

 

They'll just look "wet". Pour a little gas into a small open container
and dip the end of a plug in it - that's how plugs will look in a
"flooded" engine.

Not necessarily - part of the problem is the excess gas pooling in the
cylinder/on the piston crown, so there's a chance of the clean plugs
getting re-fouled just as quickly. The problem is not entirely with the
plugs getting wet, but there's also FAR too rich an air/fuel mixture in
the combustion chamber.

Gasoline (or anything else, for that matter) won't burn unless there's
the proper proportion of oxidizer present (air, in this case) - if you
drop a lit match into a bucket of gasoline, it will likely just go out,
as once it's in the liquid, there's no air present: it's the vapors
coming off the surface of the gas that actually burn.

Cranking with the pedal floored allows larger amounts of air into the
cylinders, helping evaporate the excess fuel more quickly, but it's
still not an immediate fix, just helps accelerate the process. Cranking
a bit with the plugs out may help clear out the excess fuel as well, but
be careful not to produce any open flame near the front fo the engine
(NO SMOKING!).

 

I'd never denied that it sparks better with a larger gap at atmosphere.
The point is, there is no use bending it when there is spark no
matter how small the gap.

Ah... it seems you're not a pro enough but will have to rely or copy
a $12 jig to determine the coil's voltage output.

This is a pointless argument. Either you mistype or lost it. A faulty
distributor cap, arm doesn't require only a test downstream of the bad
part to determine if the arm, cap is at fault. This is another point that says
you don't know how to troubleshoot electrical circuits. You're somewhat
killfiled.

Apparently, you'd just ran out of useful thought.

 

I see,
Now that you mentioned the flooding theory could explain also why on some
instances when I had the no-start problem leaving the car alone for a day
or so fixed the problem and the CRX starts as if nothing happened, perhaps
the gas flooding dissipates?...