Reducing NOx emissions- '92 Golf

I have a '92 Golf (1.8L gas engine, Digifant II control, no EGR) with
less than 60,000 miles. The car has passed many emissions tests over
the years. The car runs well, with lots of power and good fuel economy.
It does not ping under any normal load. On the last ASM2525 test I
got the following readings under load:

HC 0
CO 0.01
NOx 764

The NOx is 20% over the allowed limit here. The HC and CO readings seem
suspiciously low. I'm wondering if anyone has suggestions for why the
NOx might be so high, and how to get it under the limit. The following
have occurred to me:

1. I was using 10% ethanol blend fuel- probably not a good idea if NOx
is a problem.

2. It's been suggested air leaking into the catalytic converter could
give these results. There don't appear to be leaks, but the exhaust
system is using cheap clamps, not the proper VW pipe connectors. I plan
to tape the joints before a retest.

3. I'm using 60%glycol/40%water coolant (winters here get pretty cold).
Switching to 40%water/60%glycol for the test should lower the head
temperature a bit.

4. Yes, I should check the timing. I'm still looking for a 27mm hex head
to get the plug out of the inspection port.

5. I'm running some commercial fuel system cleaner (Redline) through the
system to try and clean the injectors.

Any other ideas would be greatly appreciated.

Thanks in advance,

Garry

Garry Tarr wrote:
>
> I have a '92 Golf (1.8L gas engine, Digifant II control, no EGR) with
> less than 60,000 miles. The car has passed many emissions tests over
> the years. The car runs well, with lots of power and good fuel economy.
> It does not ping under any normal load. On the last ASM2525 test I
> got the following readings under load:
>
> HC 0
> CO 0.01
> NOx 764
>
> The NOx is 20% over the allowed limit here. The HC and CO readings seem
> suspiciously low. I'm wondering if anyone has suggestions for why the
> NOx might be so high, and how to get it under the limit. The following
> have occurred to me:
>
> 1. I was using 10% ethanol blend fuel- probably not a good idea if NOx
> is a problem.

Steady state it should not matter, as the fuel system is supposed to
close the loop on O2 in the exhaust gasses. (BTW, did you get the test
results for O2 content?)

> 2. It's been suggested air leaking into the catalytic converter could
> give these results. There don't appear to be leaks, but the exhaust
> system is using cheap clamps, not the proper VW pipe connectors. I plan
> to tape the joints before a retest.

Doubtful, the exhaust system is under pressure, so any leak would go the
other way.

> 3. I'm using 60%glycol/40%water coolant (winters here get pretty cold).
> Switching to 40%water/60%glycol for the test should lower the head
> temperature a bit.

You may want to re-read what you wrote there!

> 4. Yes, I should check the timing. I'm still looking for a 27mm hex head
> to get the plug out of the inspection port.
>
> 5. I'm running some commercial fuel system cleaner (Redline) through the
> system to try and clean the injectors.
>
> Any other ideas would be greatly appreciated.

Low CO and HC with High NOx is typically a sign of the engine running
too lean. Perhaps an O2 sensor problem, although from what I read this
usually causes a rich condition. Could also be that you get some
un-metered air entering the intake manifold through a leak somewhere.
There are some good Digifant fuel system pages on the web, see what you
can find with Google.

>From: Garry Tarr
>
>I have a '92 Golf (1.8L gas engine, Digifant II control, no EGR) with
>less than 60,000 miles. The car has passed many emissions tests over
>the years. The car runs well, with lots of power and good fuel economy.
> It does not ping under any normal load. On the last ASM2525 test I
>got the following readings under load:
>
>HC 0
>CO 0.01
>NOx 764
>

Maybe engine is running too hot or
maybe there are some deposits on the valve stems sucking up the fuel
(due to possibly poor fuel and not enough long distance driving) or
maybe fuel pressure low or
maybe a good tune up or engine service
(test the CO at the plug, richen the CO and that should help)

let us know!
later,
dave
Reminder........
Before you criticize someone, you should walk a mile in their shoes. That way,
when you criticize them, you are a mile away from them, and you have their
shoes. Frieda Norris

Randolph wrote:
> Garry Tarr wrote:
>
>>I have a '92 Golf (1.8L gas engine, Digifant II control, no EGR) with
>>less than 60,000 miles. The car has passed many emissions tests over
>>the years. The car runs well, with lots of power and good fuel economy.
>> It does not ping under any normal load. On the last ASM2525 test I
>>got the following readings under load:
>>
>>HC 0
>>CO 0.01
>>NOx 764
>>
>>The NOx is 20% over the allowed limit here. The HC and CO readings seem
>>suspiciously low. I'm wondering if anyone has suggestions for why the
>>NOx might be so high, and how to get it under the limit. The following
>>have occurred to me:
>>
>>1. I was using 10% ethanol blend fuel- probably not a good idea if NOx
>>is a problem.
>
>
> Steady state it should not matter, as the fuel system is supposed to
> close the loop on O2 in the exhaust gasses. (BTW, did you get the test
> results for O2 content?)
The test print out doesn't give O2. There is something called
"dilution" reading 15.3. This sounds a lot like air/fuel ratio to me.
>
>
>>2. It's been suggested air leaking into the catalytic converter could
>>give these results. There don't appear to be leaks, but the exhaust
>>system is using cheap clamps, not the proper VW pipe connectors. I plan
>>to tape the joints before a retest.
>
>
> Doubtful, the exhaust system is under pressure, so any leak would go the
> other way.
>
>
>>3. I'm using 60%glycol/40%water coolant (winters here get pretty cold).
>> Switching to 40%water/60%glycol for the test should lower the head
>>temperature a bit.
>
Oops, switching from 60%glycol to 40% glycol is what I meant!

> You may want to re-read what you wrote there!
>
>
>>4. Yes, I should check the timing. I'm still looking for a 27mm hex head
>>to get the plug out of the inspection port.
>>
>>5. I'm running some commercial fuel system cleaner (Redline) through the
>>system to try and clean the injectors.
>>
>>Any other ideas would be greatly appreciated.
>
>
> Low CO and HC with High NOx is typically a sign of the engine running
> too lean. Perhaps an O2 sensor problem, although from what I read this
> usually causes a rich condition.

Yes, failed O2 sensors are supposed to force the micture to the rich end.

Could also be that you get some
> un-metered air entering the intake manifold through a leak somewhere.

Possible- when I tapped the manifold vacuum at the hose leading to the
air cleaner temperature sensing valve, the reading was only 16 inches of
Hg. Is there a better place to measure the vacuum?


> There are some good Digifant fuel system pages on the web, see what you
> can find with Google.

Thanks!

Does it have a EGR valve on that engine? CO and timing has an effect on NOx
so I would check the timing and CO% before the cat converter for the correct
settings. I think CO% before the cat should be 1% from memory. You may still
have the emissions tag somewhere under the hood which should give the data.

"Garry Tarr" > wrote in message
...
>I have a '92 Golf (1.8L gas engine, Digifant II control, no EGR) with less
>than 60,000 miles. The car has passed many emissions tests over the years.
>The car runs well, with lots of power and good fuel economy. It does not
>ping under any normal load. On the last ASM2525 test I got the following
>readings under load:
>
> HC 0
> CO 0.01
> NOx 764
>
> The NOx is 20% over the allowed limit here. The HC and CO readings seem
> suspiciously low. I'm wondering if anyone has suggestions for why the NOx
> might be so high, and how to get it under the limit. The following have
> occurred to me:
>
> 1. I was using 10% ethanol blend fuel- probably not a good idea if NOx is
> a problem.
>
> 2. It's been suggested air leaking into the catalytic converter could give
> these results. There don't appear to be leaks, but the exhaust system is
> using cheap clamps, not the proper VW pipe connectors. I plan to tape the
> joints before a retest.
>
> 3. I'm using 60%glycol/40%water coolant (winters here get pretty cold).
> Switching to 40%water/60%glycol for the test should lower the head
> temperature a bit.
>
> 4. Yes, I should check the timing. I'm still looking for a 27mm hex head
> to get the plug out of the inspection port.
>
> 5. I'm running some commercial fuel system cleaner (Redline) through the
> system to try and clean the injectors.
>
> Any other ideas would be greatly appreciated.
>
> Thanks in advance,
>
> Garry
>

In article >,
Garry Tarr > wrote:

> I have a '92 Golf (1.8L gas engine, Digifant II control, no EGR) with
> less than 60,000 miles. The car has passed many emissions tests over
> the years. The car runs well, with lots of power and good fuel economy.
> It does not ping under any normal load. On the last ASM2525 test I
> got the following readings under load:
>
> HC 0
> CO 0.01
> NOx 764
>
> The NOx is 20% over the allowed limit here. The HC and CO readings seem
> suspiciously low. I'm wondering if anyone has suggestions for why the
> NOx might be so high, and how to get it under the limit. The following
> have occurred to me:

It seems to me that the system is running lean. The extra
air will reduce the HC and CO and increase the burn temperature
causing the excessive NOx. The good fuel economy is consistent with
this.

> Any other ideas would be greatly appreciated.

I doubt that an air leak into the exhaust is the problem because this
would cause the system to run rich. Because they're relatively easy checks,
I would confirm the fuel pressure is correct and that the inlet manifold
isn't leaking and then point a finger at the lambda sensor. Rich shift is
not a very common failure mode so I'd want to test the sensor (switching
time and response to a propane flame on a bench) before condemning. If the
sensor had been replaced recently I'd wonder what was killing it, perhaps
a coolant leak.

Charles