1991 honda accord tricky start

Hello all. I have a 91 honda accord which I just replaced the distibutor
on because a police man pulled the spark plug wires while the engine was
running. (car was stolen and recovered). But the car will start normal
99.9 percent of the time but every now and then it will crank, fire, and
then die. Try again, nothing. again crank fire die etc. The amout of
starts vary but it will sudenly start normaly. I have inspected the main
relay and it looks normal. I can hear it clicking as well. The only big
change is idle drops faster than I remember but it does not go below idle
limits. Thanks for what ever input you can give me.

 

Thing to listen for (as well as the click) is the fuel pump running for
a couple of secs as the CEL is lit.

 

Even if the main realy looks good, replace, or resolder the
connections.
So far since late August when a friend's 1990 Civic "died" I have seen
10 1989~1992 main realys that where BAD. One looked OK, but when I
measured
the contact resistance, I have built a special test rig, the
resistence was
about 30 Ohms, and would vary as I taped on the relay shell. After
resoldereing
the resistance is less then .1 Ohm. Most have been from wrekced
vehicals on the way to the crunching yard, but this one was from a
classmate of our friendthe classmate had her 1991 Civic die at school.
Our friend carries a spare relay (ECM distributor(with a new ignitor),
rotor, cap etc) and helped her out by swapping the suspect unit with a
repaired one. The engine started and ran great. She had been having
hard starts, especially on hot days, or after a short stop.
She brought the relay to me for study. I have worked in electronics
for 35 years, and I have never seen so repeatable a failure mode as
Honda EFI relays. The solder joints are just BAD. The relay gets no
airflow, and does run very hot, so maybe this is the reason why they
fail.

In any situation where a Honda has a hard start, or starts then dies,
I suspect the relay until proven other wise with a know good unit.
Terry

 

I think that I have found a pattern with the starts. After the initial
start in the morning where the idle is about 1500 and drops to about 500.
every other start the idle drops to 500 right after start even if she has
been sitting for a while. after about the fourth start is when she dies.
If I let the engine turn over about four to five secs the next start she
will fire up normal. Thanks for both of your sugestions. I will take a
look at the relay again. now I just need to resolder the board right. I
do not need to remove it from the connector housing. I will also listen
for the pump during the start. thanks again

 

OK I am getting spark. I do not hear a fuel pump at all, But the car runs
99 percent of the time. Once it has started I have no problems. She never
stalls or stutters while running. And she starts most of the time as well.
Could it be something with the slow or fast idle settings?

 

About the relay, turn the ignition on and listen for a click then
another two seconds after. When soldering the relay, remove the old
solder and put in a fresh quality one. I suspect the idle at 500 rpm
could result from running on three cylinders, initially. If you see
smoke from the tailpipe, probably a flood.

Rick

 

Well all thanks for the help.
I reworked the main relay and I have yet to have the car not start on me
yet. It had become cold in Houston( at least as cold as Houston can
get)and the cold seems to help the situation as well. So thanks for all
your help Rick

 

Excellent!

I have a pal in Houston and chatting at the weekend, I was surprised
to learn of the chillier temps.