backinblue92

iTrader: (1)

Alright everyone. I've failed smog 4 times now. I've pinpointed the problem to a simple rich mixture. But after my diagnosis, I'm having issues troubleshooting it. I'll save you all the mumbo jumbo of the numbers because they don't matter except for that at 15mph under 50% load I'm running rich. At 25+ and cursing speed, I'm not. Now I've done all the standard tuneup stuff and no marked difference. That's not the problem. I need to dig deeper into why I'm running rich basically only when under load it appears.

I've checked my egr system. Apply vacuum at udle and the car near stalls.
I've replaced the o2 sensor with a bosch, and an ngk coming tomorrow.
I've checked my coolant temp sensor, and my fans do work.
I have no vacuum leaks.
My WOT compression is 250 250 205 250, dry and wet, but a leakdown test showed no significant compression loss via valve or head gasket in the 205 cylinder. So I'm assuming that cylinder just has less carbon buildup than the others, as 250 is generally high for an h22.

So I believe my rich condition under low speed load would have to be fuel-related. That would make sense, right? I've changed the fuel filter and o2 sensor, but it seems there is another sensor that is telling the ecm to richen the crap out of my mix.

Where do I go from here? Test fuel pressure regulator? Can I pull the fuel rail with injectors attached and crank ignition to observe injector performance? I did find an exhaust leak pre-o2 the other day, and did weld it shut. Also reset ecu (which is non-chipped, obd2 p13) on the off chance it needs to relearn a portion of the fuel mapping?

Basically, I've obviously got a problem to fix, but I'd like to learn as much as I can along the way. Hours of searching has turned up a lot of ideas but unfortunately most don't ever report back in a thread with the solution to their problem.

Any ideas how to diagnose the issue further will be met with great emthusiam. I'm ready to get this fixed and move on. Thank you.

98vtec

How do you know you are rich?

Do you have a wideband to come to this conclusion? Looking at the O2 STFT?

backinblue92

iTrader: (1)

I came to this by looking at the smog numbers, which have higher than normal CO and higher HC together at the 15mph test but not 25. It points to a rich mix, which borrowing an obd2 scanner to check fuel trims confirmed.

Also I pulled out my ecu to double check it wasn't chipped that I wasn't aware of and it's not. I wish I could give you more info but I'm doing all the research I can myself to learn about why this condition exists

holmesnmanny

It's likely you have an exhaust leak somewhere from the exhaust manifold to the oxygen sensor.

You can use seafoam on a cold start to try to find it, looking anywhere from the exhaust manifold gasket to the oxygen sensor, looking for white smoke leaking out.

backinblue92

iTrader: (1)

I have already blown compressed air through the exhaust and found that the outer shell on my downpipe was sucking in air. It was welded shut and the ecu reset yet the problem has persisted.

minilogoguy18

No DTC's?

How advanced is the scan tool you're using? Are you able to look at MAP in line graph with a 0.8-1.1V scale at idle? Also, check throttle angle just because if it's slightly off it can cause this, 0.45V closed 4.5V WOT.

backinblue92

iTrader: (1)

No DTCs, that's the thing that's bugging me. Nothing seems out of the ordinary, though we are looking at a first gen obd2 system. So there's flaws that later obd2 systems improved upon.

The scan tool is not as advanced as I believe I will need to source this problem, as it displays a simple rich or lean work (which flips and changes so much during normal operation but stays steadily at rich under load). Which brings me to a thought- the smog test is done at 15mph/50% load and 25mph. Engines run rich under acceleration to prevent detonation, would this be taken into account? I'm using my scan tool to assess the conditions but the condition is already that I will be rich under load as is. Is that making sense? Under load and acceleration are essentially the same thing, no? Maybe I should go back and see what my actual idle numbers are reading too, to get a baseline.

My multimeter will be able to check throttle angle. My o2 heater circuit and coolant temp sensor is good, which already has been checked. I'm starting to also wonder, along with a more advanced scan tool that can really break down values to digestive information, if I should try to track down a different obd2 p13 from a friend if possible. There could be an ecu fault, though I've smogged the car just as is previously 2 times.

minilogoguy18

Current OBD2 is far more ahead than anyone on these forums realize so yeah, you're working with a car that still uses the lean/rich switch to try to balance it out and even with the Honda scan tool the PCM is very limited on what data it can give. Newer cars will even go as far as showing an actual AFR, at least in the Honda dealer used laptops.

The multimeter unfortunately wont be able to keep up with how small of a fluctuation the MAP sensor can detect.

The PCM does richen the mixture under load which is calculated using mainly MAP and TPS input but the input taken from the HO2 sensor over rides other sensor input as far as evaluating short term fuel trim. Speaking of which, if you can monitor ST and LT fuel trims do that, let the car run at idle, at operating temp for a while and see what values you get. A value less than 1 is rich, greater than 1 is lean, the numbers stand for a multiplication of the base injector duty cycle, showing that the PCM is commanding the injector duty cycle up or down. Another thing, most people don't think so but the secondary HO2 sensor can adjust fuel trim, while it's main purpose is to monitor catalyst performance the main reason for OBD is to reduce the cars emissions, the PCM doesn't give a flying F about the engines performance, it's only programmed to keep the emissions in check and to keep the catalyst from failing.

To get a fuel trim DTC, the PCM needs to see either 0.85 or 1.15 LT fuel trim values or worse. For the first drive cycle you'll only get a temp DTC with no MIL, you can disable 2 trip monitoring by shorting the SCS aka jumping the 2 pin blue diag connector.

backinblue92

iTrader: (1)

Well it turns out that the rich condition is gone. After patching the pre-o2 downpipe hole it had remained, however I hadn't reset the ecu thinking o2 values are o2 values, regardless of the "memory" obd2. So after resetting the ecu and putting some miles on it, my mix is confirmed good. CO is hanging around .22%, over average but not enough to cause high HC. It makes sense that the leak was only causing the o2 to run rich at lower speed as the exhaust would pulse more at lower rpm, thus sucking in more air to throw off the a/f mix.

So I'm still misfiring somewhere. No DTCs on that either, but the HCs show its there. All 4 plugs are perfect, so there's no oil fouling or any signs of anything other than a healthy engine.

I'm starting to wonder if a bad coil pack or my 9 year old battery is causing the misses. My understanding is that a coil pack would either be good or bad, not partially failing?

minilogoguy18

There's a lot of things you need to check, fuel pressure, ignition timing because even though the distributor is not adjustable timing can be off due to a faulty sensor inside. EGR, blocking the EGR valve at idle is an easy way to find if the valve is leaking, the valve only operates in a certain RPM range and should not be open at idle. Valve lash, this can be checked without opening the hood which is why I asked how advanced of a scan tool you have, knowing how to look at MAP data you can find a tight valve which will cause a miss, this is actually very common on later Honda V6 engines.

thirsk66

iTrader: (2)

The only things backinblue92 hasn't checked is inside the distributor and his external coil. He has already adjusted the valves and changed just about everything emissions related besides the EVAP system items.

backinblue92

iTrader: (1)

Fuel pressure is 32 at idle and 42 with vac hose disconnected, and holds pressure steady. So the fuel return line, injectors, fpr all work and don't leak. When I start up, car turns over twice before starting, so that seems all good. Battery is 12.2 off, 14.2 running. Egr is checked and working, when applied vacuum at idle car stalls or comes close to it. Valve lash has been adjusted and is .006 intake .007 exhaust (I think. I remember setting it to the FSM low end, might have been .007/.008) Plugs, wires, cap, rotor, fuel filter, oil, oil filter done. Tps sensor reads .5 idle, 4.5 WOT iirc but it was fine, map sensor reads 5.0 constant 2.9 decreases with vacuum, um, I haven't checked the o2 circuit but did replace the o2 sensor. New cat too. And of course no dtc anywhere.

Thirsk is right the only place I haven't checked is inside the dizzy and the evap but I'm honestly not sure what to check. A more advanced obd2 scan tool would be nice but I'm once again honestly not sure how to read the more advanced settings. I'm all ears and willing to learn, I'm no car idiot lol. I can do any work I need to do. I'm just not sure where to apply it.

backinblue92

iTrader: (1)

Although I'm pretty sure the evap has nothing to do with misfires from what I've read.

Oh, and possibly the most important thing. The car runs better than ever. There's no hesitation. No stalling, no wandering, surging or jumpy idle. For all intents and purposes, you'd never know there's something wrong. I don't feel any symptoms.

There is black smoke that smells like gas though. So those HCs are def still there

holmesnmanny

I'm not really sure why you think blowing air into your exhaust is going to let you find a tiny exhaust leak unless you're using a smoke machine.

There's really only one reason a car runs rich, while there would be many reasons it would be running lean.

I suppose if you had a misfiring coil it would be possible for a non firing cylinder to display as a rich condition but normally you would feel the misfire in this situation and I know this from experience.

You really need to understand the basics of what exactly an oxygen sensor does.

backinblue92

iTrader: (1)

I believ if f I had an exhaust leak, I wouldn't be passing the 25mph test just fine. It'd kill me over the whole spectrum of varying speeds. There's two potential places for it to leak, exhaust manifold gasket and the header/downpipe connection. Then it hits the o2 sensor. I've checked it all over several times.

The o2 reads the amount of oxygen in the exhaust and adjusts the mix accordingly. Obviously. The o2 can retard the timing as far as it will go to counteract the condition but if something is causing a lot of unburned fuel to still pass through the exhaust, there's only so much the o2 can do.

Running rich and unburned fuels are not mutually exclusive. If I've got that wrong, please educate me. I've read about all I can take of a million articles on the internet

minilogoguy18

Well there is only so much a standard O2 sensor can do, it's very limited. It doesn't adjust timing btw, only injector duty. The distributor controls timing along with the knock sensor.

The post above you is wrong, MANY things can cause a car to run rich, check all the things I've posted, all of them can cause the car to run rich. Haven't read anywhere that you blocked off the EGR valve (loosen it, slide a business card under it and tighten it, leaving it hooked up) or actually pulled the cylinder head cover and check valve lash, a compression test wont pick up a single slightly too tight exhaust valve. Don't drive the car with the valve blocked, just observe fuel trim at idle, exhaust gas is inert, it cannot be burned twice, it cools combustion temps by reducing the oxygen in the intake charge hence richening the mixture, same thing happens with a tight exhaust valve.

backinblue92

iTrader: (1)

The o2 senses the mix and in addition to injector duty sends the signal to the dizzy to adjust timing though right? Or is that the knock sensor that sends the timing adjustment to the dizzy? Could a knock sensor be my issue in any way? Something causing ignition timing to be set off? I should check it anyways I suppose. I have a timing light. Never used it tho lol. Cause I've never had obd1.

I thought I mentioned I have set valve lash. All my valves were a combo of loose and tight, and they were all set to the low side of the FSM spec.

I had already removed the egr and cleaned it, along with the ports, but I didn't try blocking it off completely and checking my fuel trim. I assumed since the egr system is functioning as it should that wasn't my issue. I do see what you're saying about exhaust gas being inert and richening the mix.

You may have missed my post #12 above. I detailed all I've checked.

Might take the dizzy off tonight after I check it with a timing light if my timing isn't where it should be. Gonna have to YouTube how to use the light lol.

minilogoguy18

Yeah, I skimmed and must've missed it.

Although, if you can't shoot for the middle of the range they give for valve lash it's better to be loose, tight is what will keep a valve from closing completely and cause a misfire.

The EGR system can be working no problem but all it takes is either wear on the valve seat or carbon buildup on the backside of the valve or seat for it to hang open, you almost wont even be able to tell it's open by just looking at it. Blocking it off with a business card is the easiest method, EGR operation generally only takes place between 3-4k RPM under load, at idle the valve is always shut.

backinblue92

iTrader: (1)

What I did for the valve lash, was say if it said the range was .006-.007, I set it so the feeler would be scraping at .006 and .007 wouldn't fit. So I suppose that'd be considered middle range? After I did the lash my numbers improved by 35 HCs and another compression test brought it from 230-235-205-225 to 255-245-232-255 (diff guage 2nd test).

The smog test runs at 15mph, 1750rpm and 50% load, then the 25 test is running at 2000ish rpm and no load. So if the EGR isn't supposed to be opening at that low rpm/load, that could be the culprit. So yes, now I see what you're saying about the business card block off. If there's a chance the valve is staying open even though I've cleaned it all up, then it'd be recirculating exhaust gas and causing misses due to oversaturation relative to fuel mix? Or not enough air? Misses that also wouldn't be specific to any one cylinder, and may not be enough to throw trouble codes either. Which in turn may not even set off dtc because to the computer everything technically is working fine. Hmm that actually sounds quite plausible in that it could explain a ton of my symptoms. Wouldn't the knock sensor see that and adjust, though? Or is it to a point where if this is happening, no adjustment can overcome it?

minilogoguy18

Knock sensor only adjusts ignition timing if the mixture ignites before the spark occurs. EGR makes the mixture rich because of the very low amount of oxygen in the exhaust gas, it cannot be burned twice.

backinblue92

iTrader: (1)

Thank you that def makes sense. I tried to take the egr valve off again and the nut is jammed and started stripping. I need to find that 12mm crows foot I used last time. I'll figure it out.

Say I use this business card trick and it clears the problem is my only alternative to buy a new egr valve or further cleaning can take care of it?

minilogoguy18

Replace it. Although I'd still like to see if you ever actually get a DTC from any of this, I already mentioned how you can force a DTC from a single drive cycle.

backinblue92

iTrader: (1)

You mean by jumping the blue 2 pin? I'll have time after work tomorrow about a half hour before the shop closes to try this so if I jump the blue SCS with the obd2 scanner plugged in, it'll force a DTC if that is indeed the problem?

holmesnmanny

If you want to test the egr system you simply pull the connector and wait for the light to turn on, when it turns on the egr system is turned off. You don't block the valve unless you have a chipped ecu and can turn off the egr system manually.

holmesnmanny

Odds are your catalytic converter is bad and isn't able to burn off the hydrocarbons during the 15mph test, assuming you don't have an exhaust leak.