90 Civic - flooding/fuel in intake manifold

Hey all,

I have a 90 civic, 4dr, AT, 1.5L, DPFI, basically everything's stock, except for the open cone filter and air intake tube.

Over the last several months I've had an increasingly frustrating fuel problem. I've replaced both injectors (upper is Standard Motor, lower is Honda, neither is reman.) Also replaced the coolant temp sensor.

The problem is that during startup, too much fuel is spraying into the TB/Intake and it's flooding the engine, and it won't start. I've been running it with the lower (cold start) injector disconnected, as that's the only way to keep it from flooding at startup. However, with cooler weather coming on, it seems like it wants to flood even with the lower injector unplugged.

It's not throwing any codes, except the obvious one for the fuel injector that's unplugged.

Any ideas? Any help is appreciated.

If you're getting too much fuel, check your fuel return line for clogs.. it's the line that comes out of the firewall right next to the brake booster.

Also, check the Fuel Pressure Regulator to make sure it's not stuck shut or blocked.

It should open up when you crank the engine, due to the amount of vacuum the not-running engine applies to the manifold.

Check that the vacuum hose to it isn't disconnected or plugged as well, and spray some carb cleaner down your intake while the engine is hot, then let it sit awhile, then start it back up and drive it a few mins.

Thanks for the ideas... I had also suspected it could have something to do with the pressure regulator, so I took it off and checked to see if I was able to get it to open, and it appears to be opening, but I didn't check the vacuum port on the throttle body and didn't check the return line to see if they were clear.

I'll check them and let you know if I find anything.

not a fix but so you know, you can hold the gas pedal to the floor and start the car, it will put the computer into clear flood mode and cut off fuel untill the engine starts, good to know so you dont become stranded over the issue

could be a wiring issue possibly keeping the injector open all the time. im not sure how to check that on a dpfi, was wondering about that myself that thing must be spraying all the time.? or does it pulse like a mpfi set up

when you let it sit for a while does it fire right up or does it take a while to start? or does it justflood out really quick?

DPFI still pulses, just differently than MPFI does... it seems to stay open nearly all the time.

The only one you'd have to worry about is the cold-run injector (secondary, also referred to as the "floor-it" injector) as far as something in the wiring sticking it open... this is the one that only comes on when you need an extra burst of fuel, you can think of it like the accelerator pump/jet in a carburetor.

If, for some reason, it was sticking open, especially when starting, yes, it would definitely cause problems... If you're running with it off/disconnected, there is no reason it could be the culprit though.

Being frank, you'd be better off doing an MPFI swap than trying to diagnose this problem, but... if you want to keep the DPFI, and actually fix this thing, I'm willing to sit here and figure it out with you to the best of my ability... the choice is yours.

If you want the MPFI swap, we can all point you in the right direction for the information and parts you need. Just ask.

Next things to check on this scenario though, is the wiring to both injectors, it runs with all the sensor wiring through the loom back to the ECU, through the passenger side shock tower terminal clip. The wire colors you're looking for throughout this loom are [solid red], [solid yellow], and 2x [yellow/black stripe].

At the ECU, they correspond to pins: A1, A3 (Primary) A5, A7 (Secondary).
A11/A13 (Those yellow/black stripe wires, also go to main relay.)

The wires at A11 and A13 are powered. Don't let them touch anything unless you want to trace more wiring. If they touch the ECU case, they will burn the ECU fuse under the hood (ask me how I know).

The wires at the ECU at A1,A3,A5,A7 have the same color as the injector they're connected to. A11 and A13 are Yellow with Silver dots.

Check for chafing, chapping, cracks, etc... If you have an Ohm-Meter with long enough leads, you can check from the ECU to the strut tower, then from the strut tower to the injector w/o removing the wiring at all... I don't have one that long though.


If you need further assistance, we're just a post away.

Also its possible that one of the injectors is simply leaking or sticking open. Just because you pulled the plug to the lower injector doesn't mean fuel isn't still pouring through it.

I didn't think of that... I suppose you could test this by... hm... how would you test for that... ok, i know!

What you need:
fuel pressure gauge.

Add the gauge inline with your fuel lines, after the fuel filter.
Disconnect both injectors, and the fuel pressure regulator, turn the key ON, but do not start the engine.
Turn the key to OFF after the fuel pump kicks off.
Immediately, after shutting the key off, check the fuel pressure displayed on the gauge... after a few seconds, check it again... if the injector(s) is/are leaking, the pressure will go down significantly, and quickly.

Too bad you cant just pull the injector out and prime the system..

Id be worried about the pump leaking back to the tank and giving a funny reading. If you can somehow take the throttle body off and look at the injectors while the fuel line is still attached, that would do it.

Good point, you can do that too, but the thing is, if the pump is leaking back into the tank, it's soon to go too... lol.

If the regulator was leaking pressure, that defeats the problem to begin with... it wouldn't be flooding.

You CAN take the TB off and check it that way... for the sake of your eyes, wear safety glasses... 50PSI is alot for fuel to spray into your face at.

You'd have to take it off, flip it upside down, and have someone else turn the key on for you... or, if you ABSOLUTELY don't have anyone around to turn the key on, disconnect the negative battery cable, leave the key in the "on" position, and when you get it upside down, push the neg battery cable back on... you'll know immediately if fuel is leaking past the injectors.

Leaking injectors IS unlikely since they're new, but my Axles are new, and they click... go figure...just because they're new doesn't mean they're correct.

Wow, thanks for all the replies and suggestions...

Here's a little more about what's been done and what I know and don't know...

The injectors aren't leaking. I've done several things to test this and have had a local shop pressure test them while installed, and they're not leaking. That was my first thought, too, because that's what it seemed like was happening.

To start the car when it's cold with the lower injector unplugged, I do have to slowly depress the pedal and sometimes have to floor it if it's cold outside, but I believe this is due to the fact that the top injector isn't spraying atomized fuel directly into the intake, so the only way to get the atomized fuel into the cylinders at cold start, is to open the lower throttle plate by depressing the pedal. Once the car has run for a while and is warm, it restarts first time without having to depress the throttle.

I believe what's happening with the lower injector plugged in, is that something is feeding the ECU a bogus input making it think it needs a ton of fuel, or the lower injector is getting voltage from somewhere it's not supposed to (possible wiring problem) that's causing it to stay open all the time and just spray constantly. The other possibility as Dr0pZ0n3 suggested is that the fuel pressure regulator or the return line could be the culprit, causing the injector to spray too much fuel because of excessive pressure.

There are several sensors that feed the ECU at startup to tell it how to schedule the fuel...the coolant temp sensor, the intake air temp sensor, the atmospheric pressure sensor, the MAP sensor, and the throttle position sensor. Any one of these could be sending the ECU a bogus signal and causing it to schedule way too much fuel to the auxiliary injector (also known as the "cold start" injector). And it wouldn't necessarily throw a code if it was still within the "normal" range expected by the ECU.

Of course, as mentioned, the whole thing could also be the fault of a wiring problem that's putting too much voltage to the lower injector and causing it to spray too much fuel.

As for dumping the DPFI and going to MPFI, I've thought about that, but as far as I know, the ECU relies on inputs from the same sensors to regulate their fuel flow, so if one of them is giving the ECU a bogus signal, the problem wouldn't necessarily go away after the swap. I really appreciate the offer to help me get the MPFI put together, though.

Anyway, after I try Dr0pZ0n3's suggestion about the pressure regulator, if I don't find any problems there, I'm going to try to get some of these sensors at a local junkyard and plug them in one at a time to see if anything changes, since I haven't got the test harness Honda made to test them all with the ECU still connected, and you can't test most of them with just a multimeter.

I know, that's a cheesy way to go, but if I find that one of them fixes the problem, I'll go get a new one. It's just that most of them are pretty costly new, and I don't want to keep throwing a lot of money at it without knowing whether it will fix the problem.

I'll be out of town all next week, and won't be able to do anything past this weekend, and I don't know how much I'll be able to do this weekend on it, but I'll let you know what I come up with if anything.

Thanks again for all the ideas and suggestions...

Although you're correct about sensors controlling the injector and how much fuel it gives, none of them would actually cause it to "flood" your engine.

It's a physical issue, I can almost guarantee it.

Well theres only one single way the injector could be getting voltage its not supposed to, and that is if the signal from the ecu is broken and has grounded out. Or the injector was wired straight to ground, because they are always being fed 12v from the main relay. You can identify a black wire with yellow stripe as the 12v input. The ecu is a switched ground essentially.

either way, the ECU would be throwing an injector code 16.

so it must still have the connection to the ECU too...

Or, it's not actually flooding at all.

So, you don't think it possible that one of the sensors could be sending an inaccurate signal to the ECU causing it to program an excessive amount of fuel to be delivered and cause it to flood the engine?

Like, for instance the Intake Air Temp sensor couldn't be sending the voltage to the ECU that would indicate that it's -30 degrees in the intake, but as far as the ECU is concerned, it receives a voltage and computes a higher amount of fuel to be delivered in response, even though it's inaccurate, yet still in the acceptable range of the sensor?

Either way, that still doesn't rule out the possibility of the pressure regulator, which is what I plan to check first...and currently I'm out of town and can't check it.

Thanks again for the input, it's good to have this kind of communication with guys that are familiar with the electrical details of the fuel system.

Yes, it's possible that the sensors would read incorrectly... problem is that until the engine is started, the ECU is in open loop mode... it doesn't get input from sensors on air temp, engine speed, etc... it sprays the injectors based on the CYP and CKP on MPFI engines, and it just sprays a quick burst of fuel every so often on DPFI engines... there is no other input until the engine actually starts...

Even with the engine running, it's nearly impossible for a sensor to read that far off to cause flooding without also causing a CEL... not to mention that the more fooked up one sensor is, the more the others will help the ECU to compensate... which is why it's so difficult to diagnose fuel injection sometimes.

Say, for instance, the IAT is incorrect, and it casing the ECU to think the air temp is 10* higher than it is.... this means that the ECU richens the fuel mix to prevent detonation, but then when the ECU gets the signal from the O2 that there is too much fuel in the exhaust, it leans it back out... it's called a feedback loop.

This is what would be happening in your case as well.

I still maintain that it's a physical issue.

Hey everyone,

I know I'm not supposed to resurrect old threads, but I wanted to put a closure to this one, as we discovered the problem and I felt like with all the great input and with all the things that were done to try and correct this problem, that I owed the forum a closure to this thread. An admin can lock the thread after I post if you want and sorry for the late posting.

The entire problem was a physical issue as DropZon3 suggested...

Back in April/May time frame my son put aftermarket plugs and wires on the engine. (Bosch Platinum Plugs) After hours of work, much money and months of head-banging, we took the car to a local Honda specialist. He called me up late in the day and told me to put stock NGK plugs in the car and replace the wires with original Honda parts. He wasn't absolutely positive this would fix it, but said in his experience this was the best place to start based on what he could see.

We did as he suggested and the engine fired up first time before I could get my fingers off the key, and it's been starting every day since then with no problems.

So, this is definitely a lesson for anyone thinking about putting another brand of spark plug in their Civic. Stay with the stock NGK plugs. I'm not saying other brands won't work, but I know what happened when we used them.

Thanks again for all the suggestions and again, sorry for posting on an old thread.