Automotive Forums  Where Car Enthusiasts Go to Discuss Cars Online

1994 Honda Civic EX, Vtec 1.6L, Manual,  154,000 miles

After a rain, the car will not initially start.  You turn the key on and the engine check light is on and the car will not start.  All electric circuits are fine, everything turns on, the engine will crank, just won't start.  After a 5-15 mins leaving the key on, the engine light will go off and then you hear the fuel pump turn on.  It will then start fine.  

If you turn off the car, but leave the key on, then the wait period for the fuel pump goes down.  If you continue this key on/off process (without ever starting the car), the wait period continues to go down.  If you leave the key off for awhile, then try again, then you start waiting more for it to turn on.

Seems like there is something warming up with the key on.   This problem will occur for days after it is has rained.

I had a auto repair place look at it and they ruled out any problem with the fuel pump, it does not get any power until after the wait period.  Also ruled out the ignition system and master relay, and most of the electrical system.  

do you have the wiring diagram????

here is a copy of it...

http://repairguide.autozone.com/znetrgs/repair_guide_content/en_us/images/0900c152/80/06/1b/2a/large/0900c15280061b2a.gif

how did you rule out the main relay and the ignition switch...

i want to know.. if you car like almost all hondas have had a car alarm installed....   sometimes the connections made are less than desirable..   the relays are cheep... and fail...


where is the fuel injection computer on your car????  on the right side behind the glove compartment ??? or under the right side passenger seat?????  

take it out...   check it for water... i have found several hondas with the ecm under the seat...  flooded... destroyed ... i have found one that was just short of destroyed... and it worked after some drying out..

if you have a large magnifying lens...  you can remove the circuit board from the case carefully... and examine the solder connections for thermal damage..   look for tiny circular cracks in the individual solder connections..  this is a serious possibility as it describes your problem exactly...  i have fixed many cars with a soldering pencil... i use a quality version... not a cheep one...  weller....  cheep soldering pencils.. can have voltage at the tip... this will blow up what ever you are soldering.

rosin core solder...  SN62... is good.. for computer circuits...   don't forget a fan to blow the fumes to the side .. you don't want to breathe them... bad.. very bad..   i find marking the bad ones with a dot from a sharpie makes them easier to find when soldering..   lead free solder is really hard to repair circuit boards with... almost impossible...

be sure to keep solder strings cleaned up.. it does not take much solder to screw up a joint.. and to overlap onto the next joint..

VCR shops and radio repair shops are good at this... but if they want to take on this..????   recording studio techs do this almost weekly...  at least the good ones do  

Yes, we had a wiring diagram.  I took it to a local shop and tested many of the electrical endpoints, did some testing on the ignition system, replaced the master relay, etc.  He did not test 100%, but he did gave it a good shot.

The fuel pump is under the rear seat. He tested the wires going into fuel pump and they didn't get any power until after this delay.  So, the problem appears to before it.

The car alarm is after market.  We also disconnected this and didn't make any difference.

It is very strange, it is related to something to when just the key is on (not started).  If you leave the key on for a bit, then something happens so that it can start.  If you turn off, then turn the key on again, then this period gets shorter. It continues to get shorter for each test as the key is left.  Sometimes the delay is only a few seconds.  If you then leave the key off, like even for 10 mins or, then the delay time gets longer again.  

Could something be warming up?   I am not much of an auto repair person at all

Thanks for your response!!  

does the fuel pump get power when you are cranking the motor??????

its only supposed to get power for the first 2 seconds after you turn on the key...

and then again... while cranking...

then  all the time.. once the engine is running...

there are trouble shooting charts in the honda factory manuals...

and online...

check for pump power when you turn on the key.. for the first 2 seconds.. then while you are cranking..

post what you find..    it still could be in the computer.. if its under the seat.. dipped in water.. check... check for solder joints.....

it might also be possible that the ecm is not receiving crank signal and rpm data...

its on the wiring diagram...  

No, the fuel pump does not get any power when cranking.  It does not get any power until after the delay when the check engine light goes on (and at the same time, you hear the fuel pump turn on).

I will mention the ecm and rpm data to the auto repair guy.  Thanks!

if there is no power to the fuel pump until after this delay, would this may you think that it is not the fuel pump, but something before it.  I guess it some reason the computer detected some type of short in the fuel pump, then maybe it wouldn't send power.

But, weird thing if it was really some type of water short issue (or similar), why does leaving the key on for 10 minutes or so cause things to work then.  Next time, you try the key, it may be only a minute.  Weird  

did you look at the wiring diagram i posted???

does the  yellow and black wire that feeds the injectors and the heated oxygen sensor get power when you turn on the key?? instantly???

they will need to check to see if fuse 18 gets power during cranking..

see if fuse 31 gets power all the time...

with air bags... see if fuse 24 gets power when you turn the key on.. and while cranking...  or fuse 12 without air bags..

see if fuse 32 gets power all the time..

print the wiring diagram i posted,,, it is usually easier to read than the honda versions... that go from page to page to page...

you can do this with a test light...

post the results...

this is not rocket science.. it is basic honda trouble shooting.. and not everybody understands it...

print the text from this thread... so you can read it on paper ..at the car..

and how did they verify the main relay was good...  did they put in a brand new one...   perhaps it was bad.... stuff happens..   i have even found that people have swapped internals and returned them for refunds.. passing the trouble to the next tech...  

Thanks for the great input.  I may try to take it in on Thursday, we had a big rain yesterday, so hopefully it will still have the problem then.

He had 2 extra main relays, they appeared to have been used, but believes they were ok.  He tried both.

Thanks again, will let you know the results.  

Hi, just letting you know that I haven't forgot your help.  Been having some trouble getting it into the shop when it is messed up and at the same time when they can look at it.  We did look at it some today, but it was working ok.  We sorta tracked down all the stuff that you suggested to be tested once it messes up again. Looking at the diagram, we understand your logic in the things to be tested.

We may get some rain this weekend, so possibly can inspect it again next week.  Sorta wondering if it is the injectors or oxygen sensors.

Thanks again,  

read up on honda relays...

http://www.marklamond.co.uk/tech-honda/pgm-fi/main-relay/main-relay.htm

http://www.tegger.com/hondafaq/mainrelay.html




here is a really great one...

honda owners might really want to copy and paste this link

http://www.techautorepairs.com/mainrelaydefine.ht  

I was able to do some tests today.  The local mechanic gave me the fuse block diagrams and we have the ECM out for testing.  Here are the results...

1) yellow/black is ok
2) fuse 18 never gets power, even after the delay and when the car eventually starts

3) fuse 31 power all the time
4) fuse 24 gets power fine
5) fuse 32 power all the time

Thanks!  

Sure kinda sounds like it could be the main relay.  I am going to pull it and do some additional testing on some of the wires.  Will let you  know.  

I detached the main relay and did testing on it.  Here is what I found out...

1) Yellow/Blk has power fine
2) Yellow/White has power fine
3) Blue/White never gets power
4) Green/Yellow - has power immediately. After the delay, it goes to 0 volts for a couple of seconds (at the same time you hear the fuel pump turn on), then it goes back to 12 volts
5) Yellow / Green - does not turn on until after the delay

So, where does Fuse 18 Blue/White come from, is this the problem? I tested fuse 18 and it is fine.

The fuel pump relay does eventually turn on, but I don't see any input voltage that has this delay.  I guess it eventually gets power from the Yellow/Blk.  So, is the delay from here, but the real problem that the Blue/White does not have immediate power?

CORRECTION!!! Updated! Blue/white and Fuse 18 are now getting power during cranking.  Unless I just did not measure correctly (could swear I tried several times) before.  But, the delay is still there.  Maybe cleaning the main relay circuit board, dunno, confused.  It goes to 10 volts.  But, still have the delay.

Thx  

what is strange.. is that the fuel pump relay actually gets powered up without the engine cranking.. after sitting there are while...

i am going to look at the wiring diagram...

i will be back...


1) Yellow/Blk has power fine OK!
2) Yellow/White has power fine OK!


3) Blue/White never gets power..  (((  this blue and white only get power to the circuit when the KEY is held in the cranking position...  only when the key is in the cranking position.. all the way forward against the spring...)))



trace  the blue and white wire...    it goes to the fuse box fuse 18 in this diagram...does the fuse get power during cranking.. at the fuse box.. ???.. and to the ignition switch... i don't have the diagram for the rest of the wiring in front of me...   the honda i worked on today.. dim headlights on one side.. ended up having a missing fuse..   why were they dim instead of totally out????


4) Green/Yellow - has power immediately. After the delay, it goes to 0 volts for a couple of seconds (at the same time you hear the fuel pump turn on), then it goes back to 12 volts   HMM!!!!   if you see the way its wired in the relay.. when the black and yellow wire get power.. it closes the relay .. this should keep the hold in windings on the fuel pump relay in...  but the blue and white relay wire has to pull the relay in first..   this allows the relay to drop the fuel pump if power is lost..


5) Yellow / Green - does not turn on until after the delay

I want to know.. if you pulled the relay out of its case.. and examined the solder joints on the circuit board .. see any with circular cracks????  see any that need resoldering..

the slow reaction you are seeing is usually do to a bad solder joint..  

>>>3) Blue/White
I tested during cranking.  I also tested back at fuse 18, kinda hard to follow the wires from or to there.

>>> main relay...
I did pull the cover off the main relay.  It had some slight corrosion, but solder joints appeared to be ok.  I cleaned it off, but no change.

It does have an after market alarm system.  The local guy did temporarily disconnect it and still had the issue.  It is connected back up.  Curious if this could be affecting the blue/white?

The local guy did swap out with 2 other relays and same issue.  They were used ones I believe, but I wouldn't think to see the same problem with both of them if it was the relay.  But, who knows  

CORRECTION!!! Updated! Blue/white and Fuse 18 are now getting power during cranking.  Unless I just did not measure correctly (could swear I tried several times) before.  But, the delay is still there.  Maybe cleaning the main relay circuit board, dunno, confused.  It goes to 10 volts.  But, still have the delay.

So, sound like a main relay problem?  

ok... when the wires that carry current pass through the solder they heat up the solder... expanding it ever so slightly..

but since the solder is on a circuit board..   the circuit board works as a heat sink.. pulling heat out of the outer section of the solder lump...

so the inside expands..  the outside not as much.. the inside pushes the outside larger... then the inside shrinks back..

leaving a TUBE of solder in the middle of the connection..

this happens on so many electronic items...   even surface mount items...   i spend hours resoldering circuit boards...

if i  had a solder pot.. it would be easier..   but i can totally control the solder..  i do use a large arm circular lamp with a lens so i can see through it..  a fan to blow the fumes away.. you don't want to breath them...


i want to say.. you are really into this.. this thread will help many many people who are going to read it in the years to come...

that is one of the reasons i do this..