560SL Temperature Sensor Test
 

In pursuit of eliminating potential contributors to my SL's high idle syndrome, I have been performing various tests on the components related to idle management, namely the Idle Speed Control Unit (ISCU) and its peripherals. In the course of testing the ISCU a second/thorough time around, I again found a fault with the "TF Signal" as input to the ISCU. This signal relates engine temperature information as it is relayed by the N3 CIS-E Control Unit. Since my ability to directly query the CIS-E is currently limited, I opted to attempt confirmation of the functionality of the sensor itself, which is known formally known as "B11/2 Coolant Temperature Sensor (EZL/CIS-E)". Here's what I found...

Where is it?
 

Just finding B11/2 on my car was a bit of a challenge. The Electrical Troubleshooting Manual (ETM) contradicted itself by stating its location in two different places. On page 207, the text claimed that it was located at "left rear of engine". While the caption for Figure 2 stated "top right side of engine". Forum member Richard Wooldridge recalled that the left-rear sender is actually tasked as the "S71 Cold Engine Lock-out Switch" and this is confirmed in the 1986 Service Manual Supplement. The ETM was correct in placing B11/2 at the top right side and this is confirmed by other drawings in document 07.3-121 on pages 2 and 35.

This photo shows B11/2 on my car with its two single-contact connectors detached:

http://home.windstream.net/eurekan/c...0307.b11-2.jpg

What does it do?
 

The electrical schematics indicate that B11/2 is equipped with two similar thermistors whose electrical resistance drops as the engine's coolant temperature increases. These values are digested independently by the EZL ignition and the CIS-E control units in their management of the engine's performance. I found are no identification on either B11/2 or the two separate connector jacks to suggest that they are not interchangeable. I found no indication in the electrical schematics to suggest otherwise. Note that later model years utilize a single 4-pole connector for B11/2.

http://home.windstream.net/eurekan/c....b11-2.etm.jpg

http://home.windstream.net/eurekan/c...-2.073-128.jpg

For my high-idle situation, I am concerned with the B11/2 connection to the CIS-E control unit, which is made by a Green with Red striped wire.

Within service manual document 07.3-121 titled Testing electronic components of CIS-E injection system at page 31 is a table declaring the desired resistance levels for B11/2 at various temperatures...

http://home.windstream.net/eurekan/c...-2.073-121.jpg

My Observations
 

As my car has been retrofitted with a digital temperature sender & gauge, I have a reasonably accurate means to measure the coolant temperature.

http://home.windstream.net/eurekan/c...temp.gauge.jpg

Using this and a standard multimeter, I recorded the resistance values of B11/2 while the car was warming up from a cold state. Initial (before starting the engine) resistance to ground values from the two sockets on B11/2 were found to be 2.43 and 2.48 kOhms. The ambient air temperature was about 80 deg F.

http://home.windstream.net/eurekan/c...07.vom.tdv.jpg

The black Tachometer/Dwell/Voltage meter was set to display engine RPMs and connected to port 1 of the diagnostic socket identified as "X11".

The Green with Red striped wire (leading to the CIS-E) was left disconnected for the meter's probe, while the Green with Black striped wire leading to the EZL was reconnected.

Before startup, the car's digital temperature gauge indicated 79 deg F (26 deg C). The engine was started and the following values were noted:

Ohms / deg F / deg C
1230 / 105 / 41
890 / 122 / 50
630 / 142 / 61
478 / 158 / 70
368 / 176 / 80

At approximately 179 deg F, the Viscofan clutch apparently engaged, as the level of noise and air movement generated by the cooling fan increased substantially. After this, the temperature gauge indicated decreasing temperatures while the resistance values began to climb.

435 / 170 / 77
440 / 167 / 75

At this point, the coolant temperature and B11/2's resistance values stabilized.

My Conclusion
 

Based on the minor differences between the declared and observed resistance values, I consider the B11/2 sensor to be functional. When the engine is warmed up and the Green with Red striped wire leading to the CIS-E is disconnected from B11/2, no idle change is descernable. However, when the Green with Black striped wire leading to the EZL is disconnected from B11/2, a significant (~300 rpm) increase is observed. Switching the wires connecting to B11/2 has had no effect.

I currently feel that my high idle problem lies somewhere between the sensor and the ISCU, which is currently not receiving an adequate amount of signal voltage from the CIS-E control unit. I will be continuing tests to isolate the cause. Confirmation or condemnation of this test methodology is appreciated.

BBB

My 87 560 developed a high idle problem and became increasingly hard to start when warm. It progessively got worse to the point the car would only start when cold. I spent several hundred dollars at a Mercedes indy garage; accumlator, relay, fuel pump, and so on, but the problem persisted. In disgust I brought the car home and began to research the issue. I replaced the sensor you describe and the car now starts when warm and the idle speed problem disapeared. Your technical skills are far above mine, but based on my (limited) understanding of the function of the sensor, the high idle problem would most likely be accompained by the hard start when warm issue if in fact the sensor is malfunctioning. I hope this is some help, and I really enjoyed your post.

You meant 2.43 K and 2.48 K ohms, before starting the engine, No? I think your methology for testing the sensor is fine. But it would be helpful to know the voltage measured to ground while the sensor is connected. It doesn't matter that the resistance is correct if the voltage supplied by the EZL or CIS are wrong. Comparing voltage readings with the chart would be more indicative of a problem and might suggest a problem with either the EZL or CIS controller.

Thanks for sharing your experience, it helps shed light on a the symptom as it relates to the part I'm questioning. Although I don't have a noticeable warm-start problem (yet), I am considering replacing my B11/2 since the Resistance values I measured yesterday and the Voltages I measured today are slightly off the manual's specifications.

Yes, you are right. I edited the post.

Thanks for the suggestion. I was able to take a few voltage measurements at Pin 21 (input signal from B11/2) of the CIS-E Control Unit today. The engine was not fully warmed up, but the values were close to those charted in the manual. I want to make a graphical plot and figure out my error rate so far. Look for an update later this evening. I will probably do a full survey of the B11/2-to-Pin 21 voltage progression from cold engine to hot (80 deg C) tomorrow. Today's tests have me doubting my B11/2 sensor's accuracy and the manual's spec for the TF Signal value from the CIS-E's pin 9 that feeds into the ISCU's Pin 2. I'll be posting that question separately and later as well.