When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
For the first time in 9.5 years, the 951 has failed the emissions test.
I just finished changing AOS seals (removing/replacing the exhaust crossover pipe in the process), bunch of new vacuum lines, installing new ICV, deleting the CV and replacing with a Lindsey MBC, new plugs, air filter, oil, & cleaned fuel injectors.
I installed a new idle throttle screw & adjusted the idle as per workshop manual directions (Edit: minus the CO procedure; I only performed the "jumper-out the ICV and set the idle" procedure). FYI the car runs great. While working, I removed the CO test pipe; I pulled it out and replaced it with the factory cap.
Here are the failure results along with a previous good result. Something isn't right. What is it and how can I fix it?
Previous pass test result:
Edit (additonal info): 1988 Porsche 951. The previous result is from a few years ago (the test is every 2 years and I couldn't find the very last one) at a different facility. Seems each facility has their own description specs re: vehicle weight lol
Can anyone explain the "Dilution" score? The old one appears to be air/fuel ratio; the new one is a %. I sure hope I'm not running 8:1 air/fuel!!
Last edited by Dan Martinic; 05-29-2018 at 06:27 PM.
Reason: Tried to alter main thread title
looking at it your idle is rich
i dont think the car would even run at 8.x:1 but if it did, youd smell it for sure.
what is the car idling at (rpm)? i think a 951 should be around 850?
how old is your oxygen sensor? a failing one can make the mixture wonky, and a dead one will make the DME default to a rich mixture.
The car is running better than ever! No smell, hesitation, etc. A little bit of off-throttle downshift shudder has returned (it's been like that forever, disappeared after idle adjustment, now slowly creeping back) but other than that, I can't tell anything wrong.
But yes, 8.x:1 is obviously a problem.
The car idles nice and steady showing midway between 800-1000 on the dash (I set it with a tachometer hooked up to the coil showing 800, ICV jumpered out, and basically the same when ICV jumpered back in).
I do recall dropping the CO sensor--I had removed it from the crossover, unhooked the electrical connection up top, then proceeded to lower it down, which led to it dropping about a foot hard unto plywood sensor-end first.
Maybe I busted it. I'll have to look up how to test it.
Either that, or the shop's equipment is way off. Notice it's the same ratio at idle AND 2500rpm. Can it really be running so darn rich yet behave so normal??
I had a similar situation a couple of years ago, the car had always passed fine and then failed on CO.
Turned out to be a vacuum leak. In my case, the boot that connects the turbo to the intercooler pipe had split on the back side (out of sight of course). My guess is that with all of the work you did, you probably have a leak somewhere.
Easiest way to find a leak is to either fabricate a tool to pressurize your intake tract or buy a Boost Test Adapter from Arnnworx
Attach the tool to the intercooler coupler, apply 10-15psi of pressure, listen for leaks and spray some dilute dishwasher soap where you think the hissing is coming from to verify the leak.
Good luck and fingers crossed it's not under the manifold!
Easiest way to find a leak is to either fabricate a tool to pressurize your intake tract or buy a Boost Test Adapter from Arnnworx
Attach the tool to the intercooler coupler, apply 10-15psi of pressure, listen for leaks and spray some dilute dishwasher soap where you think the hissing is coming from to verify the leak.
Good luck and fingers crossed it's not under the manifold!
Greg- maybe I'm missing something, but how does air not leak out the other side of the intercooler pipe when you hook up the tool?
I've used a similar contraption--homemade that fits into the large J-Boot opening. I didn't plug anything else, but the air kept going through what I assume is exhaust valves.
Still... built enough pressure to clearly blow bubbles at the ICV and CV, among other places.
I guess the intercooler method needs a cap on the open side
This morning, I drove to work with a disconnected O2 sensor plug. Absolutely zero difference in engine operation.
I'm guessing I would have noticed something different?
Next, I will test by blowing propane into the intake manifold (via a vacuum port on the Laust manifold--yet another great use of this cool mod!) and see if the voltage changes!
Make sure to not pressurize the crankcase through the AOS when you do the test. Pull and cap the AOS port at the j boot.
Oh! I have to ask.. why not? I haven't heard this before, and I did indeed test both before & after AOS seals replacement. Does the compressed air test damage the seals?
I use 8-12psi air (my cheap gauge isn't too accuratej
After some driving, I disconnected the O2 sensor and connected a multimeter to check the sensor's voltage:
Idle: .5xx (slight variations in the smaller numbers)
Steady aprox. 2500rpm: no change from idle
When releasing the throttle from the steady test, it quickly drops to about .1xx then back up to .5xx
Revving it higher or quick throttle bursts cause no higher reading--it holds steady at .5xx until I close the throttle (see above behaviour)
Next, I pulled a random vacuum line (near the fuel EVAP stuff) and noted hardly any change in the readings (.4xx at most drop)
Then, using an empty port on the Laust vacuum manifold, I first pulled the plug which led to a bigger drop in reading (.3xx). I noted that the vacuum seemed to be "sucking" much stronger at the empty Laust port and it was certainly loud!
Finally, I opened up some propane and placed it under the open vacuum port. The reading remained steady, if anything it dropped a bit to .4xx
I'm no mechanic, but from what I understand, these tests indicate the O2 sensor is likely bad
I've ordered a new one. $134CAN plus taxes (Bosch OE not universal)
Here is a graph from Bosch's "Automotive Electric/Electronic Systems" book, which shows emissions vs. AFR for a typical (well functioning) engine with and without catalytic converters.
From that it looks like you are idling too rich.
Also, when disconnecting the (narrow-band) O2 sensor, you will be running on the base-map (open loop mode), which the DME does by injecting a signal slightly less than 0.5V on the missing pin.
I believe the weak point in the feedback loop is the ground connections (3 wire O2 sensor), where the ground of the 0 to 1.0V signal has to go through many dubious connections between the sensor and the DME.
10-23-2017, 08:30 PM
Decode my Emissions Test Failure
