High HC Readings on 1986 442

I'm working on a 307 V8, VIN Code 9 Olds engine. The issue is that recently it barely passed a CA smog check on the HC emissions. Since I'm going to replace the valve covers gaskets (leaking) and will have to remove most of the AIR system for access, I figure that it's a good time to fix the HC problem. The car has 33k miles on it, does not use any oil between changes, no trouble codes set, Check Engine light functions as it should with no problems, engine runs normally and fine. Timing is within allowable specs. AIR system checks out fine per the fault tree in the Olds factory shop manual.

Here are the emission test numbers:

Feb 2013:

Test Spd/RPM /%CO2 /%O2/HC(ppm) /CO(%) /NO(ppm)

15 mph /1636 /13.6 /2.1 /102 (fail limit is 103) /0.07 (limit is 0.54) /335 (limit is 762)

25 mph /1638 /13.3 /2.5 /65 (fail limit is 69) /0.02 (limit is 0.40) /451 (limit is 643)

Aug 2011:

Test Spd/RPM /%CO2 /%O2/HC(ppm) /CO(%) /NO(ppm)

15 mph /1635 /13.2 /2.3 /20 (fail limit is 103) /0.01 (limit is 0.54) /421 (limit is 762)

25 mph /1490 /11.4 /4.2 /21 (fail limit is 69) /0.00 (limit is 0.40) /476 (limit is 643)

I also have a '73 Impala with a 350 and AIR system, when it was under smog testing, it would have HC/CO results like the Aug. 2011 results above, without a catalytic converter. So, I am thinking that even with a weak or defective catalytic converter, I should not have the high HC numbers as shown in the Feb. 2013 results.

Any suggestions on where to look are appreciated!

Ok so if I'm reading this correctly on the Feb 2013 test the Co is 0.07 @ 15mph and 0.02 @ 25mph.

What type of induction system is it? What were the idle Co/Hc readings? Did the car have sufficient time to warm up and cruise at a steady state for some time before testing to burn off anything left over in the converter? Remember as the Co starts to lean out and go down Hc begins to rise. If you suspect the Cat, I suggest you disconnect it and measure with a Co/Hc meter what is coming out of the exhaust pipe to see if you have a mixture problem and compare to when it's connected.

FYI for your 73 Impala, because of the passing of Ca. Cap and Trade, which is now being implemented, if the state doesn't meet it's carbon offset numbers there is a very good chance the state will bring those cars 1966-1975 back into smog check to help get the numbers down. There will be no need to pass legislation to do it because it's already written into the current smog check law.

Having owned a 1984 Toronado for 28 years (sold 2/12), I became very familiar with the CCC system. When it came into widespread use by GM in '81, it was very advanced for it's day. However, compared with today's engine management systems, CCC is quite primitive.

I would suggest that you use an old fashioned analog dwell meter to 'see' what mixture command the ECM is sending to the QuadraJet. Set the dwell meter on the 6 cyl. range. Connect the lead to the spade terminal inside the green plastic housing at the QJet's mixture control solenoid mini-harness. While the CCC system is in open loop (before the O2 sensor heats up & starts responding), the dwell meter should read center scale. This is a 50% duty cycle for the solenoid (equal time on & off during the 10 pulses each second). Run the engine at high idle to 'get into' closed loop. The dwell should vary but stay fairly close to center scale. If the ECM is seeing a rich mixture via the O2 sensor, it will attempt to compensate by leaning it out. The dwell will then be higher (needle will rise to high end of scale). The MCS will be energized for a longer period during the pulse (longer duty cycle). This will hold the mixture needles down in the jets longer resulting in a leaner mixture. It the ECM sees a lean mixture, the reverse will happen.

This should give you a starting point as to how the ECM is attempting to control the mixture.

If the O2 sensor is more than a few years old, change it...! They get lazy over time and don't respond well to the oxygen in the exhaust stream. Bosch makes a good sensor for your system which costs about $25. Your E4MC QuadraJet may need a rebuild. Special tools are required to set it up so find a shop which can properly work on it. Unfortunately, it's quickly becoming a lost art. BTW, if you have it rebuilt, make sure the shop knocks out the hardened idle mixture screw plugs. It's SOP for an E4MC/E rebuild. You'll have to tweak the mixture screws while looking at the dwell meter. All of this is clearly explained in the chassis service manual.

Remember, the primitive CCC system will only set a trouble code for well defined issues (open or closed sensor circuits, mixture WAY out of range, etc.). It doesn't have the smarts to detect less dramatic situations...

When I was chasing a similar problem, I rugged up the dwell meter inside the car so that I could go for a drive & really see what was happening.

Let us know what you find.

Good Luck,

Paul

Helfen, you're reading the CO numbers correctly. It's an '86 with a stock E4MC carburetor. As for the warm-up time, let me digress and explain what's going on here in California. As of the first of the year, the state agency that regulates our biennial smog check program (Bureau of Automotive Repair), started a new program for inspection station licensing called STAR. As I understand it, stations can be STAR certified by meeting certain criteria, including conforming to established failure rates for the different vehicle types. In other words, if a station passes more vehicles than the norm for that class of vehicles, they can lose their certification. So, there is an incentive for inspection stations to fail more vehicles. BAR says that there is a "warm-up" or conditioning period built into the testing software, but I don't know if it's an adequate period of time for all vehicles, especially older ones that may have borderline emission control systems. The inspection station was busy when I brought the 442 in (fully warmed up, freeway miles, etc.), so it sat for about 40 minutes until the test. Knowing how fast the converter cools off on this car and goes out of closed loop, I wouldn't be surprised if everything was not fully warmed up before the test. But given the current scenario with discouraging stations from passing too many cars, it may be something that we'll have to deal with into the future. I just cited my Impala as an example of how inherently clean the AIR-equipped cars can be, I figured the 442 should have similar numbers even with a dead converter, if all is well upstream of the converter.

Paul, thank you for the tips, I completely forgot about a lazy O2 sensor and the problems that can cause without setting a code. I will check the sensor and MCS duty cycle. I have a set of very long test leads for my DMM, so on-the-road testing is another excellent idea. I had to do a similar setup with a vacuum gauge when troubleshooting the EGR system. If you don't have a chassis dyno handy, that's your only alternative.

I'm in the midst of a minor kitchen remodel, so I won't have time to check things out for a few weeks, but will be back with some results.

Helfen, you're reading the CO numbers correctly. It's an '86 with a stock E4MC carburetor. As for the warm-up time, let me digress and explain what's going on here in California. As of the first of the year, the state agency that regulates our biennial smog check program (Bureau of Automotive Repair), started a new program for inspection station licensing called STAR. As I understand it, stations can be STAR certified by meeting certain criteria, including conforming to established failure rates for the different vehicle types. In other words, if a station passes more vehicles than the norm for that class of vehicles, they can lose their certification. So, there is an incentive for inspection stations to fail more vehicles. BAR says that there is a "warm-up" or conditioning period built into the testing software, but I don't know if it's an adequate period of time for all vehicles, especially older ones that may have borderline emission control systems. The inspection station was busy when I brought the 442 in (fully warmed up, freeway miles, etc.), so it sat for about 40 minutes until the test. Knowing how fast the converter cools off on this car and goes out of closed loop, I wouldn't be surprised if everything was not fully warmed up before the test. But given the current scenario with discouraging stations from passing too many cars, it may be something that we'll have to deal with into the future. I just cited my Impala as an example of how inherently clean the AIR-equipped cars can be, I figured the 442 should have similar numbers even with a dead converter, if all is well upstream of the converter.

Paul, thank you for the tips, I completely forgot about a lazy O2 sensor and the problems that can cause without setting a code. I will check the sensor and MCS duty cycle. I have a set of very long test leads for my DMM, so on-the-road testing is another excellent idea. I had to do a similar setup with a vacuum gauge when troubleshooting the EGR system. If you don't have a chassis dyno handy, that's your only alternative.

I'm in the midst of a minor kitchen remodel, so I won't have time to check things out for a few weeks, but will be back with some results.

Jon, How can a station loose it's licence if the cars it test all pass? Does this mean the testing station mechanic does something to the car to fail the car? I'm in the process from moving from California and where I'm at we are still testing at Bar 90 procedures. One of my friends has a auto repair/smog station and I can tell you it is legal to do a pre-test without running bar codes on the car. Before I test I make sure the car is driven on the fwy first and brought right in to do the testing, this is legal too. you just need a shop owner who is willing to cooperate with you. a car sitting for forty minutes has gone way past it's hot soak period and should be either driven on the dyno for a good period of time before the test or be driven on the road so the cat is fully warmed up along with the engine. If that cat is fully warmed up you might find that the lazy O2 sensor is not a problem. For certain when you go to test be nice, but firm about your rights as to testing procedure.

I'm wondering if the higher HC readings might be due to some oil consumption, ever how small? In that case, it's been another highly-recommended thing to have a fresh oil change in the engine. The oil collects HCs from normal ring "leakage" on the combustion cycles, which if it gets back into the intake tract (possibly from older valve seals/guides with a little bit of wear on them), then it's just like a too-rich carb mixture.

Back when we already had the first RFG in TX, one of our Mopar club guys had his father's shop (which he grew up in) do an emissions test on his '70 Road Runner 440+6 4-speed car. It passed 1975 emissions with factory tune specs. As RFG and the later E10 tend to make an engine run a little leaner, it's no real surprise that the Impala is as clean as it is.

One other key thing is to have a minimum of carbon in the combustion chamber. HCs can "hide" in the somewhat porous structure of that substance to be cooked-out when the engine later runs. Locally, a product known as "RXP" is a successful carbon-removal/pass emissions test fuel additive . . . if that might be a consideration.

The statement (or implication) that the newer STAR test centers must fail a certain amount of vehicles (in their particular "class") is like telling police officers that they MUST write a certain amount of citations each shift. Certainly, it's known that a certain amount of vehicles will be out of compliance, but trying to make that percentage into reality by actions at the testing station tends to go against the "fair and just" orientation of things.

What might also be happening is that they might be automatically failing vehicles which would otherwise squeek by in the numbers. Arbitrarily tightening the numbers by 10% or so to ensure that vehicles that pass really pass (with enough margin that another retest at another testing center would still result in "Pass") would tend to fail more vehicles, but also ensure that the particular testing station's test was accurate. BUT . . . if that's the orientation, THEN the station testers would be re-writing the statutes themselves, effectively . . . which might come under another level of scrutiny . . . although the re-tests might be free. Not unlike headlights always needing to be adjusted each year in safety inspections.

Please keep us posted on your progress.

NTX5467

When we had centralized emission testing in TX, initially, it became a MESS. The state operatives told us that there was a clock in the front of each station, at the curb, which had the wait time for vehicles to be tested on it. It was in state statutes that a vehicle could wait no more than 15 minutes, lest the station could be fined by the state. When the state operatives were questioned about the number of stations vs. the number of vehicles to be tested, with the suspected longer lines and wait times, they kept referring to the fact the station would be fined if the wait times exceeded 15 minutes. They were doing the dyno-style tests according to a specific driving cycle (IM240?).

MUCH hoopla resulted. Elderly citizens who drove their car to the facility, a few miles usually, and then received a "Fail" test result as the converter was not fully up to operating heat, or they needed an oil change, or some other minor things (i.e., new air filter element) didn't set very well. There were also some 4 wheel drive vehicles which were allegedly damaged due to the way they were put on the chassis dyno, etc. So many complaints that the whole system was scrapped and new Gov GWBush had to get it all straightened out. That's when we learned that the two contractors in the state were really part of the same CA company . . . so much for "monoplies"! We also learned that EVERYWHERE that company put in centralized emission testing facilities (operated by independent franchisees, usually), it was a repeat of what we saw in TX . . . especially on the "excessive wait times" issue. Of course, that wait time issue would directly relate to the heat of the converter, or lack thereof, after the vehicle would not be running.

I would be suspicious about any "times" built into the test system software to ensure the converter was fully operational. This might be real or it could be "fuzzy real" . . . or a perceived selling point by the system's vendor, even if the "time" might be marginal at best, in reality. Certainly, older converters could well need more time than a fresh OEM-supplied replacement converter, I would suspect--something the software probably might not allow for in its "prep" section.

Just some additional thoughts . . .

NTX5467

I'm wondering if the higher HC readings might be due to some oil consumption, ever how small? In that case, it's been another highly-recommended thing to have a fresh oil change in the engine. The oil collects HCs from normal ring "leakage" on the combustion cycles, which if it gets back into the intake tract (possibly from older valve seals/guides with a little bit of wear on them), then it's just like a too-rich carb mixture.

Back when we already had the first RFG in TX, one of our Mopar club guys had his father's shop (which he grew up in) do an emissions test on his '70 Road Runner 440+6 4-speed car. It passed 1975 emissions with factory tune specs. As RFG and the later E10 tend to make an engine run a little leaner, it's no real surprise that the Impala is as clean as it is.

One other key thing is to have a minimum of carbon in the combustion chamber. HCs can "hide" in the somewhat porous structure of that substance to be cooked-out when the engine later runs. Locally, a product known as "RXP" is a successful carbon-removal/pass emissions test fuel additive . . . if that might be a consideration.

The statement (or implication) that the newer STAR test centers must fail a certain amount of vehicles (in their particular "class") is like telling police officers that they MUST write a certain amount of citations each shift. Certainly, it's known that a certain amount of vehicles will be out of compliance, but trying to make that percentage into reality by actions at the testing station tends to go against the "fair and just" orientation of things.

What might also be happening is that they might be automatically failing vehicles which would otherwise squeek by in the numbers. Arbitrarily tightening the numbers by 10% or so to ensure that vehicles that pass really pass (with enough margin that another retest at another testing center would still result in "Pass") would tend to fail more vehicles, but also ensure that the particular testing station's test was accurate. BUT . . . if that's the orientation, THEN the station testers would be re-writing the statutes themselves, effectively . . . which might come under another level of scrutiny . . . although the re-tests might be free. Not unlike headlights always needing to be adjusted each year in safety inspections.

Please keep us posted on your progress.

NTX5467

You know I have a 76 Olds that I bought new. I have every emission test it has been through. The state gives the maximum allowable CO-HC levels on the copy of the test. If you look at the first test which was 1978 and the latest test the standards have been tightened down through the years and now there is testing and a standard for NOX. When the car was built there was NO standard or test for NOX. You would think you would give a older car a little break as it ages instead of making it jump through hoops to pass. As far as your comparison with police citations goes we only have to look at what is happening in Washington to see how the rules can be distorted/disregarded to meet a unseen goal.

Hi All,

A belated update on the situation, house work and other projects took a lot longer than anticipated. Paul, you hit the nail on the head. The dwell meter showed about 38 degrees at idle, but more significant was what the meter needle didn't do, and that was move much when the engine speed was raised. I checked the O2 sensor and the voltage was lagging behind the engine speed changes more than it should have. Changed the 02 sensor and the dwell meter showed about 20 degrees at idle, but at any given engine speed in closed loop, the needle is now slightly fluctuating around a given dwell reading, something that was not happening earlier. I wasn't due for a smog test this year, but my curiosity and a spare $40 prompted me to take the 442 in for a test. Here are the numbers for the previous smog check and the one done today:

Feb 2013:

Test Spd/RPM /%CO2 /%O2/HC(ppm) /CO(%) /NO(ppm)

15 mph /1636 /13.6 /2.1 /102 (fail limit is 103) /0.07 (limit is 0.54) /335 (limit is 762)

25 mph /1638 /13.3 /2.5 /65 (fail limit is 69) /0.02 (limit is 0.40) /451 (limit is 643)

Jan. 2014:

Test Spd/RPM /%CO2 /%O2/HC(ppm) /CO(%) /NO(ppm)

15 mph /1611 /11.6 /5.4 /47 (fail limit is 103) /0.05 (limit is 0.54) /281 (limit is 762)

25 mph /1594 /12.3 /3.8 /26 (fail limit is 69) /0.04 (limit is 0.40) /278 (limit is 643)

The HCs are down by over 50%, which was my big concern, as the Feb. 2013 numbers put me a little too close to failing. A pleasant surprise was the reduction in Nox, which was my last emission-related issue before this. Pre-test, I also did change the oil and filter as NTX5467 suggested.

And my observation about CA Bureau of Automotive Repair and "encouraging" smog check failures is based on their eligibility criteria for a STAR station, at: http://www.bar.ca.gov/80_BARResources/03_Standards&Training/Star/pdf/STAR_application_requirements.pdf If the link doesn't work, here's the requirement that I based my observation upon: "The station’s Similar Vehicle Failure Rate (SVFR) over the preceding calendar quarter (e.g., January 1 – March 31) must meet or exceed 75% of theindustry-wide failure rate for similar vehicles, as calculated quarterly by BAR." My interpretation is that an inspection station must pass vehicles at a rate no more than 25% less than whatever that vehicle group's failure rate is. So smog check "pass friendly" measures like driving a car onto the test dyno after a brisk freeway run, etc., are discouraged. This also means that if a station should be unfortunate enough to test enough well-maintained vehicles and establish a high-than-normal pass rate, a station could be in jeopardy of losing its STAR certification, since exceeding 75% of the failure rate is a no-no. This also helps to keep failure rates from decreasing.

Thanks again for all of the help. It worked!

Thanks for that update! I'm glad the numbers got better with the minor maintenance item installed!

One thing in the statute is just HOW can a station determine what the acceptable "pass" or "fail" rate might be for a particular vehicle's class? WHO has that data and how can it be accessed? I can understand the desire to keep everybody honest, BUT the language CAN indicate some sort of dis-honesty on the part of the operator/shop owner is necessary for them to keep emissions testing vehicles.

What the E4MC carb has in it is a solenoid which runs the metering rods up and down in the metering jets, on the primary side. It replaces the earlier "power piston" which operated the metering rods against a calibrated spring, varying with manifold vacuum levels. The "dwell" reading is related to "on" time, I believe, of the solenoid, which also relates to richness and leanness of the mixture. When the solenoid fails, you'll find some small, circular slivers of what looks like orange silicone rubber. That debris is from the solenoid itself, which means the solenoid "kit" has to be replaced. I believe it's a pretty painless operation, though. Just need to remove the upper part of the carb from the throttle body.

ONE thing to look at on the AIR system is the integrity of the valves on the air manifolds on the exhaust manifolds. Those one-way valves are just simple flapper valves, with rubber flappers in them. They keep hot exhaust gases from moving backward in the system, toward the main switching valve (which is usually plastic, of some kind). When the flapper valves wear and age, they let too much heat back upstream in the air manifold system. This tends to cook the metal manifold, from the inside out, many times with the outer chrome coating holding it together. When the heat gets to the switching valve, it melts it and sometimes it gets hot enough for "smoke".

The flapper valves are somewhat inexpensive and are also easy to change. Given the age of the vehicle, it might be a good idea to change them when you do your other repairs.

The Air Injection Reactor system basically adds air to the existing exhaust flow, for that last little bit of "burn" to help further decrease emissions. It ALSO dilutes the exhaust gas output, too.

I'm glad things worked out as good as they did!

NTX5467

NTX5467,

Thanks for the additional comments.

Apparently the CA Bureau of Automotive Repair has that info, I don't know if it's for public consumption. It bothers me that the shop I used for the Feb. 2013 test let the 442 sit in the parking lot and cool off, then brought it into the dyno bay and tested it. BAR says that there's a preconditioning portion of the test (warmup) built into the testing software, but as you noted, older converters may need a bit more time to reach their peak efficiency. You do have to admit that testing cars on the cold side of running will help keep the pass rate on the low side, ensuring that the shop doesn't jeopardize their standing with BAR. Doesn't do much for the customers, though.

When I first started to troubleshoot the high HC readings, I first suspected the AIR system, since that would be one of the first areas to check if it was a non-feedback system, which is what I erroneously had in mind. I was thinking about my '73 Impala, forgot I was working on the 442. I followed the 442's FSM troubleshooting fault tree and the AIR system checked out fine. I will replace the check valves when I replace my leaky valve cover gaskets, that's next on the to-do list.

I'll second the replacement of the check valves in the A.I.R. system, and also the diverter and selector valves. I had exactly this problem on my 86 307 Olds motor. The check valve in one of the A.I.R. manifolds had gone bad, allowing hot exhaust gas to reach the selector valve, damaging it. The selector valve directs the A.I.R. pump output to the exhaust ports when the engine is in open loop mode, but switches to the converter when in closed loop mode. With the valve failed, fresh air was sent to the exhaust ports all the time. The O2 sensor read this as an overly lean condition and commanded the carb to run full rich. Naturally, HC levels were off the charts. Since the selector valve has no computer feedback, failure won't directly set a code. While this may not have been your problem, replacing the inexpensive check valves is cheap insurance to protect the far more expensive diverter/selector valves (they're on the order of $120 or so each).