Turbo sedan. Failed NOX. Here's the data.

1st column: Test; 2nd column: Reading, 3rd column: Limit


40 km/h

HC ppm 44 66
CO % 0.20 0.37
NO ppm 641 505

idle

HC ppm 24 200
CO % 0.06 1


Can one conclude anything in particular, based on the fact that HC and CO are well under the limit, but NO is above? For example, can I say one way or the other if I do or don't need a catalytic convertor? I'm surprised that the idle tests have laxer limits, yet my car performs better on them.

I know that high NOX relates to lean mixture / high combustion temperatures. Does high octane gas likely make for hotter or colder combustion temperatures? I put 1/4 tank of high octane in, just before the test.

Another thought. My tank pump isn't working. Is it likely that this would cause a lean mixture, hence NOX?

I usually do an oil change before the test. Didn't have time this time, but I guess with HC results passing, lack of an oil change is not a factor.

I believe that my peak turbo pressure is about normal. Given that this intake air system seems to sustain pressure, I guess that means that at least this portion doesn't have vacuum leaks. I suppose that any connections through check valves though could.

Anyway, unless someone thinks otherwise, my plan was to replace the catalytic convertor and hope for the best. Would hate to do this, however, and not have it cure the issue.

Thanks,
--
David Armstrong - '86 240(350k km?), '93 940T(270k km), '89 240(parts source for others) near Toronto

David, I am no rocket scientist, but my friend is a Volvo factory trained mechanic and tells me emissions test have gotten so stringent that most old Volvos, especially hot-burning turbos, won't pass. He told me to put a couple gallons of the highest octane I could find and then add two bottles fuel tank deicer (methal alcohol), let it warm up and do the test. I have had seven Volvos, three of them failed, and was able to pass them this way. Cools down the combustion chamber and lowers the NOX, he says. Not the most scientific technical advice, but beats parking the car. No guarantees, but good luck.

i think you are barking at the wrong tree.

simple logic would suggest if your cat is bad it would not be smart enough to discriminate between hi nox and low hc ... et al.

a bad cat is a bad cat and makes a mess of all the gases coming into it.

hi nox is traditionally as you rightly observe caused by air leaks some where upstream of the cat. it can also be aggravated by a failing o2 sensor sending back the wrong signals to the ecu.

how old is your o2 and when was it last checked for accuracy?
have you really checked for vacuum leaks?

arbitrarily changing cats is not the same as actually determining what is the cause. and they are not cheap to buy to boot.

This is the kind of confirmation I was looking for, and why I posted the full details. Sometimes simple logic doesn't hold true, but if your experience has been that these results don't point to the cat, I'll look elsewhere.

I've always had a hard time doing the "definitive test" to prove or disprove that there are vacuum leaks. Just re-read the FAQ, and I guess I'll try Bruce's propane method, but I'd much prefer to be able to isolate sections and use pressure leakdown (that's how I did the turbo/intercooler tests).

Any thoughts on whether the failed tank pump may affect this?

I'll do the FAQ O2 sensor tests too, but I would have thought that this might also affect the HC and CO readings adversely. Simple and quick enough confirmation, regardless.

Thanks for the input,

David

--
David Armstrong - '86 240(350k km?), '93 940T(270k km), '89 240(parts source for others) near Toronto

failed tank can be ruled out by simply at the schrader valve under the fuel rail at the front testing the pressure. if it is good that is not the issue. if the regulator is not leaking internally and is good you have to look elsewhere.

hi nox is an engine running too hot and too lean....unlike runway models too hot and too lean is not a good thing in a car.

air intake not desired is usually the cause unless the engine is not getting the fuel it needs

I keep forgetting that this car has a handy schraeder valve test point, unlike the 240, where you need a specific adapter. I'll check that.

It does look like it could use an new O2 sensor. It appears to only switch between .2 and .7 volts, and it doesn't react very fast - even stutters a bit when falling to lean indication. Pulling the FPR line causes it to switch high, but it doesn't hold there when left off; it goes back to oscillating. Similar thing when pulling one of the small vacuum lines off the throttle housing - switches low, but then goes back to oscillating. It doesn't seem obvious from the FAQ, however, that the O2 sensor will cure my high NOX: "Airtex Automotive reports replacing degraded oxygen sensors has the potential to reduce a vehicle's emissions of hydrocarbons (HC) by 23% and carbon monoxide (CO) by 33%". According to this, the O2 replacement will likely improve the 2 items I already passed on, but doesn't mention the NOX improvement I need.

I have to say, that I'm still not confident in the methods defined for spotting vacuum leaks. With one small hose pulled off the throttle body, I hit the spigot with carb cleaner, and got no reaction. If I shoot it right down the spigot, then yes, I do get an obvious change, but I hardly trust that I would ever notice a typical small leak using this method. I'd prefer to do either a vacuum pump or pressure holding test for a more definitive result, but there's too many other factors/conditions that allow air to pass, and I don't know enough to design the perfect vacuum/pressure test.

--
David Armstrong - '86 240(350k km?), '93 940T(270k km), '89 240(parts source for others) near Toronto

Well, I did get the in-tank pump fixed, and I did replace the O2 sensor. The retest had NOX around 495 - just under the limit (although some of the other items actually got worse, but not bad enough to fail).

Subsequent to all this, I noticed one day that the vents for the heating system didn't seem to be engaging predictably. Bingo, maybe? A leak in one of the servo diaphrams would allow unmetered air to be constantly pulled from this system and into the intake manifold.

When I tested for vacuum from the line under the hood that goes through the firewall and on to the heater control system, I did get a leak in all but the defrost-only position. I took apart the dash and tested each of the vacuum lines from the heater controls. All were airtight, with the exception of the yellow line. As per the FAQ, this is the line that activates full floor flow (and thus turns off defrost). By detaching that line from the servo and plugging it, I no longer get a leak when testing from under the hood. As pointed out in the FAQ, the downside is that defrost never gets fully turned off, hence there will never be as full a flow to the floor.

Is it reasonable to think that such a leak would cause enough airflow into the intake manifold to impact significantly on the NOX reading?
--
David Armstrong - '86 240(350k km?), '93 940T(270k km), '89 240(parts source for others) near Toronto

once i have a bad emission test and go about sorting things out i always detach the neg bat terminal once i am done for half an hour to let the ecu codes purge. then the ecu has to relearn based on my driving habits but more importantly on the new equipment installed.

Hi, David. One of my bricks failed the emissions test in February when I bought it, because of some bad gas. Specifically, the gas had been in the tank for some months before I purchased the car. After the initial tank (and after the initial test), the subsequent tank blew a completely clean and under the limit test. Perhaps, the pump is picking up random crud or there is something else amiss...?
--
1991 745; 1993 965 -- RIP: 86 764T; 79 262C