Rep Power: 0

Hi Everyone,

This forum has been a tremendous help to me in the past, especially with diagnosing a VTEC solenoid issue with my 1999 Accord. I'm hopeful that you can help me figure out an issue with my Civic. I read the no start sticky, but I didn't see my specific issue. A few days ago, my Civic started up, ran rough for a few seconds, then the engine died. I could smell a lot of gas, and the only light still on was the battery light. I started it up again, and it ran completely fine for a few days. No DTCs were set.

Today, the car was running fine, but I had to shut it off to run into a store. When I came out, the engine would crank, but it wouldn't start. I had to crank it multiple times, and once it finally started, everything was fine again. Again, no DTCs were set. Had the battery and alternator tested just to rule it out, and they are fine. I verified that I can hear both the fuel pump and the relay turning on and off, and the CEL goes off after the relay turns off, but I guess that doesn't rule out a failure of either since the car was starting when I tested them.

I'm having some trouble diagnosing this just because my normal fuel, spark, compression rundown would cause the car to crank no start every time and have issues once it started as well. I'm reading some things that it could be related to the fuel pump or the relay, but I'm not sure it is either since I can hear them both running. I have a service manual for it at home, but I'm guessing it won't be much help with this. Thanks for any guidance you can give.

 
Rep Power: 517

When it won't start, is it flooding or starving for fuel?


I'd use a scanner that can read live engine data.
I'd watch for what the computer thinks the coolant temperature is reading when it won't start. Is the temp reading correct for the situation??


A flaky temp sensor can flood or starve the engine for fuel at seemingly random times, but as long as it stays running it might not be noticed.
Starting (or not starting) is when the issue becomes a big problem.

As long as the temp reading is plausible to the computer, there won't be any error codes. Meaning, it can be waaay the f*** off and still not set codes.

 
Rep Power: 0

Thanks for your response. I'll try to check the coolant temperature when it happens again. Is the temperature gauge on the cluster controlled by the coolant temp sensor, or this there a separate sensor for that? If not, the gauge seems normal.

 
Rep Power: 517

Yes, in a way.
The temp sensor tells the PCM what the actual temperature is.

The PCM buffers that data and sends it to the gauge cluster to run the temp gauge.
What you see is not necessarily what is actually happening.
Needle sweep is not linear with real temp data.
The gauge stands in the middle when engine temp is anywhere near a normal range, and there can be about 75 degrees change in temperature with no real change in needle position.

Make sense?




Getting live data on my scanner is just one of many paths I may choose to take, depending on how it I think it acts when I see it.

 
Rep Power: 0

Got it, thank you. My normal tool can't read coolant temp, so I have a tool on order that can. In the mean time, is there anything else I should try? Is it worth it to test the fuel pressure, spark, or compression? I'm guessing fuel would be the most likely of the three listed, but I'm not sure if any are worth testing for a starting issue.

 
Rep Power: 517

Sure. Figure out this:
When it won't start, pull a spark plug. Is it wet or dry?

If it is wet, that's one answer (flooded).
If it is dry, grab a spray can of carb and throttle cleaner and see if the engine will fire on that. That would indicate a lack of fuel.


If it's a lack of fuel, then track down why. Is it a flaky temp reading? Is the pump not running? What?






I fixed an older CRV the other day, it would not start after a 10 minute stop at the grocery store. Sometimes.
It would start if you hold the gas pedal to the floor and crank the daylights out of it.
I fugured out the temp sensor was reading about 150 degrees colder than actual temp when it wouldn't start.

 
Rep Power: 0

I may have to look into sensor failure with the car -- I'm getting some weird driveability problems, again. Symptoms are varying power from cold esp, and the stalling problem is back. Yes all connectors and wiring is fine, and no jiggling of connectors magically fixed the problem. Spent my get-out-stall-free card, I guess.

I thought when you held the pedal to the floor in any FI car, it tells the PCM to cut fuel -- this is the proc to clear a flooded engine in case of ingestion or a fuel injector problem (think it's the same on motos as well)...

 
Rep Power: 517

Yes for most, but this feature is not included in all fuel injection systems.

One that stands out in my mind as not having a "clear flood" mode was early 2nd gen RX-7s that loved to flood......although once an owner had it flooded there was almost no way they were going to revive it on their own.

 
Rep Power: 0

Does it apply to the D17 and K24?

 
Rep Power: 517

Mash your gas pedal to the floor and then crank it.

EDIT: If it doesn't start then it has it. Even if it does start it might still have it.
I'm pretty sure Honda uses it across the board but I don't have printed info to back that up.
The owners manual starting procedures might give a clue though.

 
Rep Power: 0

On pg. 177 of the OM, it does say to clear flooding with 100% throttle, for those wondering (as will any modern FI Honda... any modern FI car or powersports engine, actually). The logic cuts fuel and spark at 100% throttle when not already running.

BTW... there's no way that CR-V could've started with the pedal to the wood, as even with a bit of fuel left from a flooding, there's no spark to ignite it.

I think the feature is also present on a 2nd-gen rotary... as long as it's MY89+. Otherwise you have to pull the EGI fuse and leave the throttle alone, which works on all the FI rotaries... but you need only mash the pedal for '89+. Good thing too, as without an open throttle, it'd take forever to clear the engine... which on a pre-'89, does take several start attempts of a few seconds each, which is a lot if your battery isn't fresh.

The way to clear a carbed engine of flooding:

-- fix the reason why it's flooding (clear crap holding open the float needle)

-- WOT

-- crank for a few seconds -- it should now catch. Repeat until it starts and idles.

I dunno how many sad carbed-bike cruisers I've come up to on my (maniacally-fast) street rides on the Oly Peninsula... found they'd been trying to start it for several minutes... cranked their bike's grip to WOT... and had it catch in literally one second. On an FI bike, crank WOT for a second or two, rest a second or two, start at half-throttle. So many beers and free breakfasts at the Breakwater Inn!

This is is so the air from a fully-open throttle can lean out super-rich fumes and dry off plug electrodes. It'd make sense that FI standards industry-wide would converge to make WOT a must-have when clearing a flood condition... but apparently that didn't apply to every FI engine until the '90s.

 
Rep Power: 517

Of course it did. They all will. Clear flood mode ONLY cuts fuel.

The spark is present at WOT+cranking, same as with a carb. Check it yourself on your own car if you don't believe me.



89 sounds about right for when they changed it......2nd gen started with 86 M.Y., they were horrible about flooding. No clear flood mode.

ALL of their rotor motors were horrible about flooding if the driver didn't know WTF they were doing though.
I used to feed ATF down the intake to restore compression. Makes spectacular amounts of smoke once it does run again!

Many starters were burned up by owners trying to start those engines too.

 
Rep Power: 0

Interesting -- my Escort cut both, so clear-flood standards don't apply to all manufacturers, then.

The justification for cutting both in that Escort, was in case of water ingestion from minor fording, to save the coil... apparently Ford thought the Escorts would be crossing streams more likely than an injector would stick open ... but even so that's an odd match. Could be they just adopted their truck standards to their cars? Who knows with Ford, not like they document half the stuff they implement...

 
Rep Power: 517

Interesting. I wish I could verify that first hand. I don't recall that as a strategy on EEC-IV or EEC-V.



Well, THAT sounds optimistic.


Most people mad enough to run them into the lake weren't gonna pull them back out of the lake.