![]() |
|
|
|
Registered
Join Date: Mar 2004
Posts: 206
|
How to pass CA smog with a CIS Lambda
Hi,
I just did some experiments with a CIS Lambda controlled car (not a Porsche, but very applicable for CIS Lambda 911s). Got it to pass with flying colors using an LC-1, simulating a NBO2, but I did some more experiments nevertheless (smog shop owner was curious and cooperated). Turns out the CIS Lambda used is using a 1 wire O2 sensor. Same as the Porsche CIS. At the low loads/speeds where the smog check happens (basically lugging a 911), the 1 wire sensor is not getting hot enough to operate correctly (verified it by characterizing the impedance of the sensor, which is temp. dependent and measuring it in operation). 1 wire sensors are heated by exh. gas only. In the 911 application this problem should be even more severe because the gas has to go through the heater boxes and more piping from the right engine side. The (relatively) cold sensor outputs too little voltage (unless brand new) and is too slow and the Lambda computer regulates therefore to richer than it should. Replacing the stock 1 wire sensor with a 4-wire aftermarket sensor made the problem go away. A 4-wire sensor uses 2 wires to heat the sensor (one htr wire to switched 12V, the other to Gnd). The sensor ground wire goes to the engine ground, sensor output wire to the lambda computer. This way the sensor is hot enough during low load smog-check conditions. Also would be hot enough at idle. The CIS Lambda computer has an idle throttle switch so it goes open loop under idle conditions. Porsche knew the sensor is too cold at idle, but they did not anticipate the new CA smog laws when they designed the CIS Lambda system. Regards, Klaus |
||
![]() |
|
In the shop at Pelican
Join Date: Aug 2002
Location: Los Angeles, CA
Posts: 10,459
|
Good idea...
I wondering, do you actually need the 4 wire? Couldnt you do it with a three wire? |
||
![]() |
|
Registered
Join Date: Dec 2002
Location: Redondo Beach, CA
Posts: 542
|
So at idle it's open loop, but at the speeds used for the smog check, it's not?
I think that's what you're saying but just wanted to verify...
__________________
-Todd '89 930, '97 TLC (Toyota Land Cruiser), '96 T-100pick-em-up '95 BMW R100 GSPD (gone but not forgotten), '07 BMW R1200GSA |
||
![]() |
|
Hilbilly Deluxe
|
That sounds like what he is saying, but it is wrong.
CIS-Lamda 911s, meaning 1980-1983 911 SCs absolutely do not run open loop at idle. If your O2 sensor is good, it will work fine at idle, provided you have allow the engine time to get to normal operating temp. If the Lamda system was open loop at idle, unplugging the O2 sensor (forcing the system to open loop) would have no effect on how the engine runs at idle (which it does, try it and see). You also wouldn't be able to test the function of the Lamda system by connecting a dwell or duty cycle meter to the test leads by the fuse block if the system was open loop, because the duty cycle would be the fixed open loop number, (85% duty cycle for 1980, 65% for 81-83). Hell, you can even set your idle fuel mixture using output from the Labda system, by tweaking the mixture until the duty cycle fluctuations average ~50%. I don't have the manual in front of me, but I believe the only time a the system should run open loop is when the O2 sensor is not yet up to operating temp, the O2 sensor is unplugged, or the engine is at full throttle. Tom |
||
![]() |
|