Engine temp Q - 1996 318ti

[Bills86e]

Is the fan cycling on & off now, when AC is off?

[bazar01]

The fan was still not cycling on and off. I got the new radiator yesterday and installed it. Ran the engine for a while and the gauge started creeping up so I shut it down.
I pulled the fan switch and jumpered the terminals on the plug. It's a 3 prong plug. The fan ran on the first pair and did not run on the second pair, so I pulled the fan switch off the radiator. I will test for continuity tonight when submerge in boiling water. I can probably just heat test it with a propane lighter and test for continuity.
Just in case it tests bad, I already bought a new fan switch from Autozone with a 91C and 99C temp rating.
Hoping for the best.

[tiFreak]

did you bleed the coolant? it's a little more complicated than other cars

http://www.318ti.org/forum/showthread.php?t=27983

[Bills86e]

I am going to replace the ECT sensor. I figger that is the trigger
to the ECM/DME sending a signal tru the Black with Green wire
at the Radiator Cooling Fan Switch.

[bazar01]

Originally Posted by tiFreak

did you bleed the coolant? it's a little more complicated than other cars

http://www.318ti.org/forum/showthread.php?t=27983

Yes, I bled it twice already. I was getting heat out of the vents. This morning, the reservoir is all the way down. I added distilled water up to the mark.

I tested the old and new fan switch in boiling water last night. Only one pair of contact closed at boiling point. The other pair is not closing. What the h%ll? There are two contacts. One is supposed to close at 91C and the second at 99C. Both old and new sensor behaved the same way. This is not right. Time to look at the electrical diagram. I will post my findings later.

[mohaughn]

Have you checked your oil? Make sure there is no coolant in the oil. Overheating problems combined with coolant going away could mean a HG leak as well. Otherwise, looks like you are running down the usual suspects.

[bazar01]

Originally Posted by Bills86e

I am going to replace the ECT sensor. I figger that is the trigger
to the ECM/DME sending a signal tru the Black with Green wire
at the Radiator Cooling Fan Switch.

Looking at the wiring diagram, that sensor has nothing to do with the radiator fan controls.
The slow speed or what they call the normal speed in the diagram will never come ON even when the low temp contacts (2 & 3) on the radiator temp switch is closed because the power wire to the fan is fed from the high speed relay K22 terminal 87a, and K22 will not be energized until the high temp contacts (3 & 2) on the radiator fan switch is made.

I wonder why it was wired that way. I may need to rewire the slow speed section of the radiator fan so it will cycle ON and OFF. It was just not right for the slow speed not to come ON when engine is idling in stop and go traffic.

correction: 9/27/09, 1PM. Maybe the ECT you are talking about is read by the ECM and closes a separate set of contacts between the relay and the radiator fan switch. Did you replace the ECTsensor located on the cylinder head and under the intake?

[bazar01]

I believe the slow speed power wire should be moved to terminal 87 and the high speed power wire to 87a of relay K22. This way, the slow speed will run when the radiator fan switch low temp contacts are made. The high speed will only run when K22 is energized and passes power to terminal 87a.

Now I just need to find relay K22 and switch the power wires between 87 and 87a of relay K22.

I will post results later.

[bazar01]

Ok. I did more tests this morning. Since my fan works in high speed, I need to concentrate on the slow speed section.

K21 is the slow speed relay. It is the yellow relay closest to the firewall. Terminals 30 and 87 are the 12V power feed to the fan. Jumped these terminals and the fan ran. Good! So terminal 30 is always powered even when ignition switch is OFF. The electrical diagram then has a typo error between 87 and 87a.

Now, I need to test the coil terminals of relay K21. K21 coil is triggered by low temp contacts on the radiator fan switch. If I jump the brown and white/green wire on the connector side, it should pull in the k21 relay and the fan should run. If it does not, I will connect the white/green wire to ground and see if the relay pulls in. If it does, I am missing a ground connection. If it does not pull in, the coil int he relay is bad. Then I will swap similar relays later. I will post results.

[vjmvjmvjm]

Originally Posted by bazar01

Ok. I did more tests this morning. Since my fan works in high speed, I need to concentrate on the slow speed section.

Have you checked the fan resistor? If the fan runs at high speed but not at low speed, chances are that the resistor mounted on the fan is bad. This is a very common problem on the M42 / M44. It has nothing to do with relays, switches, wiring, grounds, etc. The resistor is located on the fan assembly.

Unfortunately, the resistor is not available separately from BMW. But you can use an E34 resistor, or get a VW fan resistor. There should be more info in the archives on replacing the resistor.

Vic

[bazar01]

Originally Posted by vjmvjmvjm

Have you checked the fan resistor? If the fan runs at high speed but not at low speed, chances are that the resistor mounted on the fan is bad. This is a very common problem on the M42 / M44. It has nothing to do with relays, switches, wiring, grounds, etc. The resistor is located on the fan assembly.

Unfortunately, the resistor is not available separately from BMW. But you can use an E34 resistor, or get a VW fan resistor. There should be more info in the archives on replacing the resistor.

Vic

Thanks.

I already found out the resistor is open so I just spliced the wires together so the fan just runs in high speed for both low and high fan switch contacts. See my post # 15 on page 1.

I just found out I am looking at the wrong schematics. I posted the early model schematics.
Here is the late model schematics.
On the late model, the relay coil for K21, low speed relay, does not go straight to the fan temp switch. It passes the ECM contacts then to the fan switch. Something is telling the ECM to close the internal contacts to pass the ground connection to the fan switch low temp contacts.
I verified by attaching a wire jumper from terminal 85 to ground and the slow speed contacts closed and the fan ran.
For now, I will just install a by-pass switch to activate the fan in traffic by connecting a switch from terminal 85 and ground.

[bazar01]

Here is the by-pass switch to activate the radiator fan thru the radiator fan switch. This will by-pass the ECM contacts.

Another way is to wire the by-pass switch straight to ground instead of to the fan switch so the fan runs with the flick of the switch.

[bazar01]

SUCCESS!

With A/C OFF, the radiator fan now cycles ON and OFF.

[Bills86e]

Duh.......Is is cycling on the Hi or Lo?????

[bazar01]

Originally Posted by Bills86e

Duh.......Is is cycling on the Hi or Lo?????

It is cycling in high speed for now since I shorted the dropping resistor in front of the radiator fan so the fan runs both in high speed only.
I am getting a 1.0 ohm x 90 Watt resistor from Digikey to replace the stock resistor so I can get the slow speed back on the fan.