Hi.

I read through a few dozen threads (great forum) and have a problem I cannot resolve.

The car is a non VTEC Honda. I recently bought it and did all the tune-up things, fuel filter, air filter, etc. The PCV is functional. The car does not burn oil.

The car idles rough and surges under power. The fuel mileage is terrible. When I depress the clutch, the RPMs drop to 500 or so then it idles between 1000-400 rpms. Under power, the car stutters at cruise speed. With no load, the engine hesitates and surges.

I cleaned the EGR passages. The ECU appears to be good. I can unplug the MAP (puts it in closed loop) and it idles at 1000 rpms and runs better (still not perfect.) I can then clear the ECU and the problem comes back.

I sprayed all the vacumn lines I could find with TB cleaner, and found no change in behavior.

All four plugs are nearly fouled (though they appear very new.) The car is running very rich.

I found the rad overflow empty and filled it. The engine drank 2/3 of an overflow tank of water. I will keep an eye on this. Bleed the cooling system?

The ECU was changed by the PO. It appears they were tracking down the same problem.

Basically, the car runs terrible, throws no codes, the ECU seems to function properly, the EGR is cleaned. I have ruled out the IACV and FICV because the EGR passages were so clean and the problem happens under power as well.

Help!

 

I added a sig, seeing if it shows up.

 

Have you checked your fuel pressure? If the pressure regulator fails or gets stuck, pressure can go pretty high causing it to run rich, even if the computer compensates by clamping down on the injector pulse width.

Also, if it's like my Fords, there's another failure mode that will make it run rich. Since the regulator is referenced to manifold vacuum, a leak in the diaphragm will let gas pass through the diaphragm, into the vacuum line and ultimately to the intake manifold. The easy way to check this is to pull the vacuum line and see if there's fuel in it. Be careful as you don't want fuel in your face or on any hot engine parts.

O2 and MAP sensors can also affect air/fuel ratio. The engine coolant temperature sensor can come into play too. If it's failed cold and in-range, the ECU could be stuck in warm-up mode, but I wouldn't necessarily expect it to run as bad as you're making it sound. However, it's worth a look.

As for the coolant loss, if it's not leaking externally, there's only one other place for it to go. Anytime you get low on coolant, bleed the system to be safe. When the engine cools, if the overflow isn't full, air can get sucked in.

 

I swapped the ECM for another one, and the car has the same symptoms. If I unplug the MAP, car throws MAP code and has the same symptoms. I unhooked the EGR (for only a couple of minutes), car did not throw code (maybe EGR not off long enough), car showed same symptoms. Removed and cleaned idle screw, same symptoms (someone already did it). Checked vacumn to pressure reg, good vacumn there.

When the key is turned on, the CEL comes on for the 3 seconds the fuel pump runs, then the pump turns off and CEL goes out (according to spec). This leads me away from fuel pressure problem.

Took car to shop and hooked up diag tool. No1 O2 sensor showed 0.59 volts. Unplug sensor, voltage climbs above 3.85v, no code thrown (not off long enough probably), car showed the same symptoms.

Everyplace I look, someone has been there. IACV has been obviously removed.

It appears someone checked everything. Valve cover has been removed and replaced (someone checked valve lash? I think so.) Plug wires in good shape. Plugs brand new (but fouled).

I hope the internet has the answer, because driving this car is now out of the question. Can't afford it. Have used 3/4 tank of fuel for 101 miles.

I've got about 14-16 hours into it and $100 worth of gas. It does up for sale tomorrow. I'm avoiding Hondas in the future. I know most are good cars but I don't want to ever change a Honda fuel filter again.

 

This only tells you that the fuel pump is running and priming the fuel system correctly prior to engine start. It's not an indication that the fuel pressure regulator is working.

 

You're right basket case. Sorry I did not get back, I didn't get email notifications.

I did the fuel pressure test. Of course, we all know there is no schraeder valve on these cars, so I cut the hose to the fuel rail and tee-d it. The fuel pressure test went as noted in the shop manual. High fuel pressure with FPR unplugged, normal pressure with regulator plugged in. Both readings within spec. Eliminated that gremlin.

I got a diagnostic tool from a friend. I need one of these at all times in all of my cars.

Car is throwing no CEL so we looked at all the general readings first. Both O2 sensors show a rich condition, O2-2 (after cat) shows ~0.9v, O2-1 (before cat) shows ~0.6.

I didn't know what to look for so as I drove home I noticed short term fuel trim cycling through very HIGH (+) readings. Up to 46.9%, then drops to zero and climbs again. Weird.

Ok how about the basics? Well I didn't know. Checked MAP and got 21.6"Hg. Now I'm getting somewhere. KOEO MAP is 28.9"Hg (off a little bit, there is a storm here though). General OBD-II diags on this tool do not show car's BARO sensor.

Still, that is only 7.3" of Hg vacuum. There's my problem. Read a little bit about vacuum readings. Have eliminated catalytic conv. so far because the car breathes ok at high RPM, and the idle vacuum sits so steady. Snap throttle and variable RPM tests are inconclusive with this tool, does not react fast enough.

Auto Zone in the morning for an analog vacuum gauge.

Condition becomes worse as engine heats up (totally heat-soaked). I'm making a guess of first intake gasket or cracked intake, or timing belt off a tooth.

Eliminating injector o-rings (all four plugs fouled about the same, though I may grease up the o-rings if I can't get it to run perfectly.) I doubt other vacuum lines because car reacts correctly when I played with ALL of them (and I mean all). Will check anyway.

Yes it's a long post but I do better typing than talking and thinking.

I'm pretty optimistic I am on the right track.

One thing I'm missing and expected from the Honda tuner world... What is the exact formula for Honda's Calculated Load Value? I see the SAE OBD-II spec, which has a LOT of room for discretion built into the formula.

Honda is obviously taking IAT, BARO, and MAP readings into account to estimate current airflow. There is a lot of talk around the internet about what Honda uses for Current Airflow. I'd like to see the formula, and the table of max airflows. Guess I need a demon and software.

I am running 79.8% CLV at idle.

Yes a lot of this is duh but this is my first crack at EFI with diag tools. Fairly interesting.

 

Sounds like progress. Even if you haven't found the problem, you're eliminating possibilities. What diagnostic tool are you using and would you recommend it? I've always borrowed one from the parts store, but I'm considering something that will display live sensor outputs.

What about the engine coolant temperature sensor? What is the output temperature when the engine is cold and when warm?

I can understand the low vacuum being related to an intake leak, but that usually gives you a lean condition, not rich. I'd be interested in what the analog gauge tells you compared to the MAP sensor.

What is CLV?

You mentioned the shop manual. Are you following any of the troubleshooting flow diagrams or are you just winging it?

Keep up the good work. It may seem like a big hassle now, but you'll figure it out. Persistence pays off. Think of this as part of your education. Next time, it'll be much easier.

 

CLV is Calculated Load Value. It's talked about 'round the Honda community, here is as good a definition as exists:

About PID Calculated Load Value » OBD-II Resource

The OBD regulations previously defined CLV as:
(current airflow / peak airflow @sea level) * (BARO @ sea level / BARO) * 100%
Various manufacturers have implemented this calculation in a variety of ways. The following definition, although a little more restrictive, will standardise and improve the accuracy the calculation.
LOAD_PCT = [current airflow] / [(peak airflow at WOT@STP as a function of rpm) * (BARO/29.92) * SQRT(298/(AAT+273))]

Frankly, I hate to repost info that is going to be just another search hit for some poor schmuck looking for Honda's implementation. :lol: But there it is.

Since the MAP is before the throttle butterfly, that's not good vacuum, in any case, car warrants look at analog vacuum gauge.

Hooked autozone loaner tool up to brake booster vacuum (the convenient spot in middle of intake plenum that won't throw a code.) 20-15psi vacuum, fluctuating wildly. Fluctuation increases and becomes more dramatic as throttle increases. Snap throttle shows drop to ~0psi, then an increase to ~23psi on decel.

Classic valve guide according to vacuum gauge lore on internet. In any case, it's showing some type of valve activity, but NOT showing clogged exhaust, intake leak, vacuum accessory leak, etc. All other symptoms ruled out.

Ok, get a compression tester. Set the car up for proper comp test, and all 4 cylinders show increase to 178psi within 5 cranks. Pretty much an illustration of what the shop manual, Haynes manual, and all other info I can find indicates. Motor indicates brand-new condition with this test. It's not a leak-down test, but this test indicates don't try it.

Ok, check the timing belt timing. Was advised could be off a tooth. Car doesn't run like that is the case, really, but let's try it. Timing belt IS lined up correctly, in fact, it's very new, to the poitn that the timing belt I purchased to replace this belt at only $18 was returned. The belt is that new. Both accessory belts are brand new as well.

Put the valve cover back on, returned all my tools, and am ready to go home.

The car is great, very gratified by the compression test. Motor is in top shape and well-maintained.

I'll think outside the box (but probably too far). This car was either a shop car tore down and built up for educational purposes, or owned by a fairly decent shadetree mechanic who tried everything to fix this problem. The car is well-maintained (but everything has been taken apart on it) and competently repaired/inspected.

It's maddening because with about $250, this car would ride, start, stop, handle perfectly, and even have some cosmetic work done, and it would be a top of the line used car with what appears to be a motor worth another 150,000 or 200,000 completely trouble free miles. Instead, it is a car with an unfixable problem that makes it a pain to drive and super expensive.

edit:

Frankly, did not check valve lash out of disgust. Just put the cover back. I couldn't instantly find a .010 and .012 feeler gauge (pretty tight Honda, great motor), so it didn't get done, lol.

I've also sprayed every injector down and got no vacuum leaks.

The only thing I can think is to do a full leak-down test to 100% eliminate the motor as source of vacuum fluctuation. I am 98% sure the motor is fine.

I have followed the troubleshooting procedures in some cases, in others just kind checked it on my own. IACV, MAP, ECU, PCV, etc, were done according to book. I usually went farther than normal in every case, assuming bad indications when there is a good indication, just because this car has been torn apart and I want to look as close as possible at it. And in every case, things come back ok.

One thing I have done: Check the EVAP valve. The solenoid does not appear to a direct 12v signal, i.e. can't hear it activate like you can with other control devices like the IACV (can hear and see the plunger retract on 12v signal). I checked the EVAP signal to ECU. Appears to be no connection to ECU plug on the grounding wire (the wire that actually controls the activation, ECU grounds right?) So, I just hooked a pen tube (pen with guts torn out) up in place of teh EVAP solenoid. The car is supposed to suck EVAP any time it's warm anyway, so I wanted to see. The car still runs the same. So either the EVAP is bad, and the control line is bad, AND the entire EVAP system is bad, or that's not the problem. Occam's Razor says EVAP is not the problem.

 

I could be wrong on this, but it looks to me like the MAP sensor senses pressure downstream of the butterfly. I have a spare intake/throttle body left over from an engine swap and I took a look at it. Also, I had a CEL and got the freeze frame data. MAP reading was 57 KPa, or 8 psi. If the MAP sensor were sensing pressure in front of the butterfly, I would expect closer to atmospheric pressure.

Also, did you get the units correct on the vacuum? 15 psi is about atmospheric pressure (14.7 actually,for a standard day). Less than that is a vacuum and more than that is positive pressure.

Don't give up. You'll get there. Take your time and don't rush it.

 

I think the map is routed to test pressure after the throttle plate. You may want to check the o-ring on the map sensor and maybe check the port for a clog.

You said you did a tuneup, but what maintenance items did you replace?

Can you tell us the wire colors going to the map sensor and the throttle position sensor? Each should have 3 wires.

Also, tell us the wire colors going to the injector resistor. It is near the firewall on the driver's side of the car. It is a metal box with cooling fins on top.

I would probably clean every ground that you can find in the engine bay. If you go to the common diy thread on top of the gen tech help forum, there is a link to the online shop manual thread. In that thread, the honda-tech link has a 94 shop manual that will help a lot. The electrical section has a wire harness and ground location part that shows all the grounds in the engine bay.

Have you checked the timing marks by aligning the crank to TDC, then look at the markings on the cam sprocket? You will have to remove the valve cover to do this.

Maybe try a compression test?