AAARRRRGGHHHH! Won't pass smog!

[Bishop]

Allright, I've finished up my 85 hatch with the Hitatchi carb and it runs great. Way more power than before the engine rebuild, even compression on all cylinders, smooth idle, and even better feul economy. The new clutch grabs great, and shifts smooth too.

Here's the problem, IT WONT PASS SMOG! It passes every test here in CA, except the CO (Carbon Monoxide) which is off the chart. To give you a perspective, a passing score is less than 124 ppm, a gross poluter is 220ppm, and my Sube is scoring upwards of 300 ppm! I even got a score of 0ver 350 once. Its insane and the smog guys dont seem to know what the problem is. They just tell me that it's, "running really rich."

The Y-pipe is in pretty rough shape, but I think its Cat is OK and the O2 sensor seems to be working correctly. It has brand new exhaust from the Y-pipe back, including a new Cat and Muffler. It SEEMS like its running clean, but the tests show otherwise.

Any Idea what's going on here? Any fixes that anyone esle has had luck with? Really, I'm desperate and looking for anything here!

[edrach]

I'm certainly not the expert here, but "running rich" is not your problem. A rich mixture means unburned fuel and results in high Hydrocarbons and not high CO. It might help someone to help you if you just list the emissions results (all of them including what' s passing). For example, WA state testing is done at both idle and cruise (2500 rpm). Is your CO high at both speeds? Are the HC levels close to failing or just very good? An '85 Hatch with a Hitachi shouldn't have any trouble but three areas I would look at since I've been down this road a few times before is check all the vacuum lines for leaks; spraying a can of starter fluid around fittings and lines and look for an increase in rpm and you might find a vacuum leak. Nextly, check the vacuum advance on your distributor; if that doesn't work your emissions at cruise go to heck. Lastly, check the PCV valve and the large hoses going to your valve cover gaskets. If you replace the PCV, use an OEM version from the dealer; some of the aftermarket ones don't work/last. Good luck with it.

[GeneralDisorder]

It could be that you are running rich, but you have a cat that is cleaning up the extra HC and not the CO.

 

It's hard with those feedback Hitachi's, but there are only really two sensors in there that matter - the O2 sensor, and the coolant temp sensor. You should check both of them for proper signal.

 

Next you should check the feedback solenoids mounted to the manifold. They regulate the fuel in the carb, and need to be signaling properly.

 

There is a HUGE amount of information in the FSM's about this system, and how to properly tune it. Have you changed the carb? If so, you need to completely retune the feedback system. Unfortunately it requires an exhaust sniffer to perform the proper tuning by the book. :-\

 

GD

[GeneralDisorder]

Download the service manual for the EA81 here:

 

http://www.finleyweb.net/default.asp?id=142

 

Page 42 has some info that may be usefull to you.

 

Find someone with an 83 or 84 FSM for all the gory details of testing the feedback system..... if you are really nice to a local dealer they may photo-copy the appropriate pages for you. My friend has an 83 manual as well. He's out of town till Moday.

 

GD

[GLCraig]

Are you sure that it's the Carbon Monoxide that too high because CO is usually measure by a percentage of the exhaust vapor not in parts per million. Where as Hydrocarbons are usually measured in parts per million and a high HC number usually means a rich condition.

[edrach]

Download the service manual for the EA81 here:

 

http://www.finleyweb.net/default.asp?id=142

 

Page 42 has some info that may be usefull to you.

 

Find someone with an 83 or 84 FSM for all the gory details of testing the feedback system..... if you are really nice to a local dealer they may photo-copy the appropriate pages for you. My friend has an 83 manual as well. He's out of town till Moday.

 

GD

I have an '84 FSM as well as a Bentley Manual which will cover your car. I don't have a scanner. If someone local to me can scan the appropriate pages, I'd be glad to lend them the manual for that.

[Bishop]

OK, here are the results of the last smog test and you guys were right, the HC is PPM and the CO is a %, but the chart I have shows it as follows:

 

 

 

 

I assume they list the Measurement like a batting average just to simplify things. Anyway, it confused the heck out of me

 

Oh, GD, I had the carb rebuilt but its not a new one. I think that is where i am going next, back to the carb shop to see if they can set it up for me on the car. Maybe they know a few tricks I dont and I have a feeling that is where the problem may be.

 

As for the sensors, I think they are all working OK. The O2 sensor would probably give a Check Engine lite if it wasnt signaling correctly, and the car would run like crap. The temp sensor in the manifold may be bad, but I dont really know how to check it.

 

Good lord, why do they have to make this smog thing such a pain?!

[GeneralDisorder]

Actually, the O2 sensor being bad wouldn't neccesarily cause the engine to run badly, or set the CEL. It depends on how bad the O2 is. The ECU can only tell that the O2 is bad if it does not signal at all, or if it's signal is out of the allowed specs. If cannot tell if the signal is just plain wrong. The O2's get "dumb" after a couple years and their effectiveness is reduced considerably.

 

Same with the coolant temp sensor.

 

These are relatively cheap parts.... about $25 for the O2, and about $12 for the coolant temp. Checking the O2 output can be done with a simple mixture monitoring device - you can find plans all over the net. The coolant temp is easy. You just remove it, and put it in boiling water while watching the resistance value. Or even just check the resistance while the engine is running.

 

If the carb runs good, then it's probably not the carb. There are only two adjustments on the carb itself - idle mixture and idle speed. If your idle quality is good, then likely the carb is fine. The signal it's receiving from the ECU however, makes all the difference. Without any signal, the feedback carbs natuarally run extremely rich. If the ECU thought the engine was cold, or that the mixture was wrong, your signal to the carb would be to run richer.

 

GD

[edrach]

I'm with GD as far as his last sentence about it running richer. Judging from the numbers the car is running too rich. HC might be within limits, but just barely. Generally when the car is running as required, the HCs are well within limits, not at the hairy edge. Check out the parts that GD mentions since the default condition when the part fails is for the carb to run richer. Another thing no one has mentioned and I haven't since it's so basic....make sure the air cleaner is replaced and/or clean. A clogged air filter will make your car run too rich.

[GeneralDisorder]

Those numbers are way rich. Last time I had a feedback Hitachi tested it blew like 30 PPM in HC....

 

GD

[rallyruss]

NOOOOOO:banghead:

 

CO- is an indicator of a rich mixture

HC- is UNburned fuel wich can be caused by really REALLY rich mixture but is normaly because of low compression or misfire type failures.

 

if the HC is cool then you really are not that far out. do you have air injection on that thing? if that does not work properly you will come out a little high on the CO side. what is the timing set at? sometimes a little retard(no jokes please) can help a bit. not too much just a couple of deg. the carb itself can do this too. posible float level problem?

 

I hate smoging cars with carbs. but maybey a little "accidental" vac. leak might help:rolleyes:

 

if its a feed back system check the o2 signal and the dwell on the mc solinoid. I am more familiar with this on a GM car so forgive me if I use the wrong terms.

oh and yes the temp sensor could be the problem as well check the voltage when cold and hot what do you get?

[GeneralDisorder]

He should have a single ASV on the drivers side. Pretty much a fool-proof system. The reed valva can fail.... but that's pretty obvious when it does. Big increase in noise - smell of exhaust gasses. Not pretty.

 

I do now recall that the ASV on the feedback's is vacuum operated.... check the vac signal to the top of the ASV. Or just replace with one of the non-vac style that work all the time.

 

The Hitachi feedbacks run REALLY rich when they have a malfunction in the control system. The gas mileage will go in the toilet. As in you will run 20 MPG in a car that should be getting 35.

 

GD

[Bishop]

Hey GD, not sure what you mean by ASV. Is that in referance to the return line from the carb back to the Fuel filter or something else?

 

Anyway, have you got part numbers for the O2 sensor and the Temp sensor, or a recomended brand? I figure they are old enough to afford replacemnt anywyay, and its a good place to start.

 

Thanks for the info guys, and I'll keep you posted as I pluck away at this dam thing

[edrach]

ASV--Air Suction Valve. Original part is about $70 from the dealer, but a newer version is available for under $20; all you need to do is remove the reed valve from the newer version and install it inside your older version. Search the archives for the write up on this.

 

Hey GD, not sure what you mean by ASV. Is that in referance to the return line from the carb back to the Fuel filter or somehting else?

 

Anyway, have you got part numbers for the O2 sensor and the Temp sensor, or a recomended brand? I figure they are old enough to afford replacemnt anywyay, and its a good place to start.

 

Thanks for the info guys, and I'll keep you posted as I pluck away at this dam thing

[GeneralDisorder]

Rather than spend any real money on the ASV, I make a trip the to junk yard and grab two or three of the reed valves themselves. Dismantle the ASV and remove it. They probably won't charge you more than $0.50 or $1 for several of them.... they won't have any clue what they are if you do it this way anway.

 

Now the cool part - normally I don't use used parts for something like this, but in this case it's alright because you will only have the ASV enabled for the emmissions test. After that, you just put a quarter in the valve body where the pipe from the head threads in. This will disable the valve, and it only takes about 15 minutes to do it. When you need to pass emmissions again, simply remove the quarter. This way, you don't need a new reed vavle because you are only using it for very short periods of time, and under those conditions it will never wear out.

 

Get an aftermarket O2 sensor. The dealer wants a fortune for them. Last one I got was a Bosch I think. Worked fine. You can check with the dealer on the temp sensor, but those are availible aftermarket as well. I replaced both on my feedback Hitachi and the arftermarket ones worked just fine.

 

GD

[Bishop]

OK, Im still not totally clear on the ASV thing. Here is a picture of my basic setup: http://www.iehardcore.com/Hitatchi-Emissions2.gif

Its kind of huge, so I just posted the link.

 

Anyway the nipple on the Solonoid Valve that has a circle around it is broken off and it is just capped off, as is the vacuum pipe that it is supposed to connect to. It didn't seem like a big deal, but it could be part of the problem. The EGR Pipe that is shown "x"ed out here is not on my car. Instead I have the other type of EGR that goes to the air box and to a large Air Thermactor Valve that is attached to the exhaust manifold.

 

Everything else in the picture is correct, and all the vacume lines (except the one that is broken) are connected according to the diagram under the hood. I wish I had a picture of that too, but I couldn't find one.

 

I'll try to take some pictures tonight and post them

 

GD, any tips on removing the O2 sensor? The repair manual makes it sound kind of tricky.

 

Thanks for all the posts, you guys will help me get this thing going eventualy

[Bishop]

OK, I'm retarded.

I was looking around again today and realized that the Air Solenoid that is broken controls the Air Themactor Valve that allows fresh air into the exhaust.

This valve is Broken like this:

 

So I hooked it up to constant vacuum, bypassing the control solonoid valve, and it started pumping. Sweet.

Now for the smog test. Wish me luck

[Bishop]

Allright, it's that time of year again. I figured rather than start a new thread, I'd just build on this old one.

 

I passed last time, but this time I had a sneaking suspision that I might fail. The car has been running rich (I know we disputed this before) I can tell by the fuel smell coming out of the exhaust and the semi-crappy gas milage (less than 20mpg).

So I had them run a Pre-test, here are the results:

 

 

As you can see, I passed the pre-test. . . and I did OK. BUT!!! I failed the EGR functionality test. Supposedly when you apply vacume to the EGR the engine should stumble or stall completely, but mine did nothing. They tell me that I need special tools to clean out the carbon deposits from the EGR passages in the manifold, but I think they're full of crap.

 

Anyone ever run into this issue before? If so, what the heck did you do to fix it?

[hatchsub]

When they're applying vacuum to the EGR valve all its doing is opening up and letting exhaust gas into the engine. If it was good and they did that at idle then the engine would stumble. My bet is the EGR valve itself is shot. They applied vacuum to it and it didnt open up cause there is a hole in the diaphram.

[Virrdog]

I think someone on here showed how you can pull the the actual valve off and soak it over night in some sort of cleaner to get it cleaned up.

[Bishop]

I'll have to take a look at it this weekend. I'm pretty sure the smog shop guys told me that they unhooked the EGR and then tested the manifold passages directly, but now I'm not exactly sure what they did. I need to look at the manifold to see what they are calling the EGR again, because this subaru isn't set up like any one I've seen in a manual. I've just got to look at it again, then figure out what I need to do.

[Virrdog]

You can always do the seafoam through the intake manifold. Whether it helps or not, its fun.

[daeron]

You can always do the seafoam through the intake manifold. Whether it helps or not, its fun.

 

Yarrr!!!

[GeneralDisorder]

I'll have to take a look at it this weekend. I'm pretty sure the smog shop guys told me that they unhooked the EGR and then tested the manifold passages directly, but now I'm not exactly sure what they did. I need to look at the manifold to see what they are calling the EGR again, because this subaru isn't set up like any one I've seen in a manual. I've just got to look at it again, then figure out what I need to do.

 

The only way to test the EGR is to apply vacuum to it's little nipple and that will pull open the port allowing exhaust gas into the intake manifold. The engine will stumble as the EGR isn't supposed to be opened at idle.

 

The EGR's get exhaust carbon build-up on the actuator pin, and get stuck in the closed position. They are NEVER to be replaced unless they leak or have a bad diaphram (very rare). Remove the valve from the back of the manifold, and clean it with carb cleaner and a scraping device (screwdriver, etc) till you can make it move freely and re-install. Should be fine - you can test it's operation by attaching a length of vac tube to the nipple and sucking on it - should cause the engine to stumble at idle.

 

There's nothing in the "passages" of the manifold that can be tested so I don't know what they are talking about. Just go ahead and clean the EGR and put it back on. Should be fine. Looks like your numbers are plenty clean so once you have the EGR working you should be golden.

 

GD

[Bishop]

Excellent, Thanks GD

You always come through when I'm in a pinch.

I'll check it out this weekend and I've got plenty of carb cleaner so I should be good

 

I'll let you know how it goes.