McDave

It's hard to tell from the photo but if that green corrosion in the lower left corner is on the fins too, then it's time to replace it. My radiator guys says his braising won't last long when they come in like that, even after the hot tank. He may have to block off a couple of the worst ones too. Since these car's cooling systems are already marginal and you're doing a cooling related repair job, why take the chance of blowing the gaskets again? I would replace it.

Yep, you can lead 'em to water but you can't make 'em drink. Apparently it's a thankless job too.

Well it's about time! I'm glad you figured it out! BTW If you had bypassed that solenoid with a longer vacuum hose as suggested earlier, you might have gotten to this point sooner. But who cares, it's fixed!!!

Sounds like you have a bad coil wire or a plug wire if you were near the dist.

The reason I suggested you blip the throttle instead of checking it at certain rpms is some egr valves need exhaust backpressure to operate. You don't get much backpressure unless you either quickly ramp up the rpm, or put a load on the engine (or rev it to 3k rpm). If it's an automatic, have your SO put it in drive and give it a little throttle while you feel the diaphragm. If it's a 5 speed, tee in a vacuum gauge on a long hose between the solenoid and valve (you should have a vacuum gauge in your toolbox anyway) and drive it. PS I'm assuming you're not running headers and a low restriction muffler and no cats. On some cars just gutting the cat is enough to cause an egr valve to stop opening due to lack of backpressure.

I'm convinced you can repeat that over and over until you're blue in the face and some people will just not get it until they try it for themselves. And even then...

I just use a wrench to turn the alternator pulley since it's easier to get to on my Loyale. If it starts slipping as it nears compression, push down on the alt belt as you turn the pulley.

I'm having a hard time visualizing this (need coffee). What do you drill the hole into?

Since the vacuum going into the solenoid seems stronger than what is coming out of it, you might try getting a longer vacuum hose and bypassing the solenoid temporarily to see if the egr valve acts better. If it does replace the solenoid. PS I hope it idles lower than 1500 once it warms up.

The '90 FSM covers '90-'94 so that's the one you'll have to go with. The bottom port on the canister is supposed to be open. I can post a photo of my '93 canister showing the top 4 with the one plugged from the factory if that will help you convince the ref. I would go to a junkyard and find any old instrument cluster that uses a Service Engine Soon light, take the plastic film strip out of the cluster, cut out just the SES part, then attach it to your cluster where your SES light bulb will shine through it. I feel for ya man. I have a very low tolerance level for bureaucracy and could never live there. Leaving ca would be a good thing, but be careful where you move to. Your suby probably wouldn't pass Austin's smog test either.

Any rpm above an idle. For your testing purposes just blip the throttle. For the smog test, if you really want to make sure it's opening at 15mph/1849 rpms, you could tee in a vacuum gauge just before the valve and drive the car 15 mph/1849 rpm and watch for the vacuum signal. I really wouldn't bother though. If it opens when you blip the throttle, it's working. I listed all the checks to be done in a certain order. If it passes the first checks you really don't need to go through all the rest (I should have made that clear so I went back and edited it). Here is the edited version... Do the following checks in order. If/when it passes the initial quick test, none of the other checks need to be done. After making repairs in the other steps, rerun the quick test. Make sure the car is warmed up first since you shouldn't be getting egr on a cold motor (causes hesitation). Start the car, put your finger under the egr valve and open(blip) the throttle a bit. You should feel the diaphragm raise up. Now bring it back to an idle and manually lift the egr diaphragm with your finger. The engine should idle rough or stall. If it did stall and the diaphragm did open at part throttle on a warm engine, then the egr is good. You passed! End of quick check. If it didn't stall at an idle when you opened the valve manually, then the passages are clogged up. Clean out passages, rerun quick test. If vacuum didn't open the valve at part throttle, then either the egr valve is bad, the solenoid is bad, the solenoid isn't getting vacuum, or the ecu isn't commanding the solenoid to open. So, check to make sure that vacuum passes through the solenoid to the egr valve at part throttle on a warmed up engine. If it's passing through the solenoid and not opening the valve, then the valve is bad. Replace the valve, rerun quick test. If the vacuum isn't passing through the solenoid, check to make sure the solenoid is getting vacuum from the throttle body at part throttle. If you have vacuum at the solenoid and the solenoid doesn't pass the vacuum through to the egr valve then either the solenoid is bad, or the ecu isn't commanding the solenoid to open. Time to break out the volt meter to check for the command. Note: If the coolant sensor to the ecu is bad telling the ecu the engine is cold all the time, then the ecu never will command the egr solenoid to open. Final note: Since this is a SPFI upgrade, make sure your wiring from the ecu to the egr solenoid and coolant sensor are right. Of course like GD said earlier, the ecu should be throwing codes if they aren't, but it's something that needs to be checked if you're not getting the command. These systems seem to run so lean it would probably run better! But yeah, you only need to check that if the ecu isn't commanding egr flow.

Hc and No. The high No was the main clue to the egr not working. There are other things that can cause high Hc, like a misfire. Ok lets take it from the top. Edit: Do the following checks in order. If/when it passes the initial quick test, none of the other checks need to be done. After making repairs in the other steps, rerun the quick test. Make sure the car is warmed up first since you shouldn't be getting egr on a cold motor (causes hesitation). Start the car, put your finger under the egr valve and open(blip) the throttle a bit. You should feel the diaphragm raise up. Now bring it back to an idle and manually lift the egr diaphragm with your finger. The engine should idle rough or stall. If it did stall and the diaphragm did open at part throttle on a warm engine, then the egr is good. You passed! End of quick check. If it didn't stall at an idle when you opened the valve manually, then the passages are clogged up. Clean out passages, rerun quick test. If vacuum didn't open the valve at part throttle, then either the egr valve is bad, the solenoid is bad, the solenoid isn't getting vacuum, or the ecu isn't commanding the solenoid to open. So, check to make sure that vacuum passes through the solenoid to the egr valve at part throttle on a warmed up engine. If it's passing through the solenoid and not opening the valve, then the valve is bad. Replace the valve, rerun quick test. If the vacuum isn't passing through the solenoid, check to make sure the solenoid is getting vacuum from the throttle body at part throttle. If you have vacuum at the solenoid and the solenoid doesn't pass the vacuum through to the egr valve then either the solenoid is bad, or the ecu isn't commanding the solenoid to open. Time to break out the volt meter to check for the command. Note: If the coolant sensor to the ecu is bad telling the ecu the engine is cold all the time, then the ecu never will command the egr solenoid to open. Final note: Since this is a SPFI upgrade, make sure your wiring from the ecu to the egr solenoid and coolant sensor are right. Of course like GD said earlier, the ecu should be throwing codes if they aren't, but it's something that needs to be checked if you're not getting the command.

Normal. That particular vacuum hose is hooked up to "ported" vacuum. The vacuum port inside the carb is just above the throttle plate and doesn't start drawing vacuum until you open the throttle. Some cars use ported vacuum, others use direct manifold vacuum but the soleniod doesn't allow vacuum to flow to the valve at idle otherwise it would stall, as you saw for yourself. So yes, your egr system likely is not the cause of your test failure. (sorry about the goose chase ). On to the miss!

These type threads always seem to fall apart right about now. Here's one of the better explanations I've seen on why... http://aardvark.co.nz/hho.shtml A couple weeks ago I got into it with a water2gas affiliate. After thoroughly debunking his claims, I asked him, while looking at my Silverado, if he wanted to know how I doubled my gas mileage. How? I waved the title to the Subaru I just bought from him.

Ok good, it's getting vacuum and opening only at part throttle like it's supposed to. Now all you need to do is verify that it's flowing and actually having an effect. Temporarily apply vacuum to the valve at an idle. The engine should run rough and maybe even stall. If it does your emissions problem is likely not egr related. Put the original vacuum hose back on and look elsewhere for the cause of your missing. Note: You can reach under the valve and manually lift the diaphragm up to accomplish nearly the same thing, but that won't tell you if the diaphragm is moving enough under vacuum power to do any good.

Yeah, I didn't realize carbureted Subarus had electric pumps when I wrote that. My mistake. Cold Chuck, have you tried looking down the carb throat for gas boiling over after a hot shutoff? Float level good? Two bad carbs (not likely)? Too much fuel pressure? Reminds me of the time an older Chevy Monte Carlo came in flooding. A coworker held the pressure gauge on the mechanical fuel pump while I cranked it over. He said look at this! It read 6 lbs? He said no, the needle spun all the way around once and stopped at 6 lbs! New pump fixed that one.

Ah, right you are.

I hear ya. It's idle and 2500 rpm here. I wonder if his is actually idling at 1849?

I think you misunderstood what I was saying. You can test the egr passages by applying vacuum at an idle. If the engine stalls or tries to, then the passages are clear, and the valve is opening. Egr comes on at part throttle, right where he is failing: 1849 rpm. Egr is solely responsible for high NO (NOX). Also, if there is no egr at part throttle, chances are it's getting pre-ignition which will cause high HC.

Since your NO(X) is also very high I would say your EGR valve is not opening at part throttle (1849 rpm) and as a result you are getting some pre-ignition (you may not be able to hear it) and thus your HC goes up too. Here's a real good article detailing the relationship between the 5 measured gasses: http://www.automotiveu.com/zack5gas.htm Bottom line: Get those vacuum lines routed correctly. Make sure you have vacuum to the egr solenoid so when the ecm commands it, it will pass the vacuum to the egr valve. Also make sure the egr valve will open and when it does, that it will try to stall an idling engine. To test it, start the car and apply manifold vacuum directly to the egr valve. You can do that with a hand vacuum pump or by unplugging one of the other vacuum lines temporarily and plugging it straight into the egr valve. If it doesn't stumble, reach under the valve to see if the vacuum is lifting the plunger up. If it's not then the valve is bad. If it does lift it up and doesn't idle rough or stall, then the egr passages are plugged up. Remove the egr valve to clean them up. In conclusion, if the egr soleniod is getting vacuum, and the soleniod passes the vacuum to the egr valve when the ecm commands it (while driving down the road), and the egr valve gets that vacuum and opens up to clean passages, then your emissions test should pass. Good luck! PS You could still fail on visual due to the missing air injection, so you might want to get an estimate to replace those parts to show the inspector. Assuming they are still available they may cost too much to replace and he can grant you a wavier... if your state is anything like mine.

It sounds like you're getting vapor lock. Check your fuel lines and hoses routing to make sure they are not near anything that gets hot, like the exhaust or heater/radiator hoses.

Yeah I see what you and GD are saying now. If it's like mine where the a/c (and water pump & crank) only belt is inboard, you can take off the inboard belt, and the compressor, but you would have to leave the a/c bracket since the alt bolts to it. The outboard belt to the alt just skims over the a/c outer pulley (and around the water pump & crank), so the compressor really doesn't need to be there. The ideal solution is like GD says and to find the right bracket from a non a/c car so you can relocate the alt inboard (at least in my case with factory air).

Double post - n/m.

Is his '90 Loyale different from the '93s? See my post above. My '93 has a dual pulley clutch and the car can be run just fine without the rear belt.

There are two belts on the a/c compressor. The one towards the radiator just skims over the top of the pulley and drives the alternator. Don't remove that one. The compressor and rear belt can be removed, but if there's any possibility you (or the next owner) would ever want a/c again I would leave it in place to keep the system clean and dry. It's really not that heavy.