I said that I'd look into changing the temperature point at which the Auxillary Fans (AF) cut-in. Well, I now know WHAT to do but I would appreciate some help in HOW to make it a "pro" mod!

Here's the theory. These comments are for my W140 S500 w/119.970 engine. They may NOT be applicable to a different class of car (E series) and/or different years (95 and on).

S500 has two (2) fans that draw 35 amps(!) at full tilt. Assume this means 17.5 amps/fan.
The S500 has a three stage fan, some models only have two stages.

According to MB data, here's the as-delivered temp vs stage activation:
Aux Fans:
Switching Points------------------ 1st ------------ 2nd ------------- 3rd
Coolant temperature (C)--------100 on--------- 107 on -----------115 on
------------------------------------------- 95 off----------100 off------------ 107 off
Refrig pressure (bar) ------------- 14 ------------- 17 ---------------- 20 on
Actuation of AF (V)---------------- 7 ---------------- 9 ---------------- 12
Eq resistance of CTS----------- 310 ------------ 250 ------------- 200 ohms

The fans use the [CTS] Coolant Temperature Sensor [CTS] (NOT the ECT sensor (B11/2)) and turn on when the resistance of the CTS is as shown above. Note that to TEST the fans cut-in, 310 and 250 ohms are substituted for the actual CTS. This means that at 100 deg C, the resistance of the CTS is 310 ohms; at 107 deg C, it's 250 ohms (per MB documentation).

Thus to 'fool' the system, you need to add a resistor in parallel to the CTS to make it look as though the temperature is higher than it really is! Resistors in parallel LOWER the resistance value.

What temperature to use? I think it should be between 90 and 95 deg C. That's simple enough but HOW to do this mechanically (mount the resistor) is not clear to me yet. You'll see later how I did it.

Resistor Value:
Use a 1%, 1/4 watt FILM type resistor to ensure stability. For example, Radio Shack PN RSU 11345741 (1.1K). This is NOT a stocked value, so don't try to get one unless you special order it. There are numerous other sources for this type/value resistor, typically any electronics house. A CARBON type should be avoided since they will drift badly with heat.

So what is the value to use?
95 deg C ====> 1200 ohms
90 deg c ====> 1100 ohms

Note that the higher you choose to have the AF kick-in, the larger the value of resistor that's needed. That's because the sensor value changes exponentially near 100 deg C. The resistor should be "electrically" across the two (2) leads to the CTS sensor (B10/8). A value that is in between the stated values will move the point accordingly although it will be hardly noticeable.

Now how to mount this permanently? Waiting for ideas. Would like to see a mating connector with the resistor mounted so it can be plugged in!

My pic shows how I did it. Take the connector apart and solder the resistor into the pin leads where the wires go. Then snap the top back on, and plug it in. That's all there is to it! Link:

Testing shows that it turns on at 91-92 degs C, as best as I can read the temp gauge in the instrument panel. And takes the temp down to about 87 degs and then shuts off. With A/C on full, it's less than 95 deg C.

So far haven't seen any downside to this mod, ie 'fooling' the A/C system by 5 degs or so.

Remember that this change involves a system with TWO (2) coolant temperature sensors where the power train management is done via the ECT sensor and the A/C is done via the CTS sensor (ie a separate sensor). If NOT, this mod will not work. For vehicles with only ONE temperature sensor, this mod will also not work!

This modification is also FAIL SAFE. If the resistor were to fail, which is OPEN, then the fan ckt will revert to 'stock' operation.

Don't forget WATER WETTER (WW): You can now (for summer driving) add WW to a 80/20 % mixture of water/antifreeze plus a can of WW. This with the CTS mod will keep your car's temperature gauge well below where it used to be! Hope that you find it useful.

[Edited by JimF on 05-12-2001 at 02:52 PM]

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

I have thought of doing the same for some time, but have been hesitant because the temp sensing unit is also going to "Fool" the module. I think that has other funtions related to temp that you don't want fooled.
The other problem is the parallel resistance may be
right at that temp, but way off at others because the thermistor is changing with temp, but the installed one is not changing at all. So at lower /higher thermistor ohms with a constant value parallel will be way off.
Sooooooo..... are we affecting other ECU paremeters here ?????
Me no know-- me ....

Other ECUs do use this sensor...Beware.

__________________
Benzmac:
Donnie Drummonds
ASE CERTIFIED MASTER AUTO TECHNICIAN
MERCEDES SPECIALIST 11 YRS

The chart for the CTS has now been 'constricted' to a shorter ranges of values but it has the SAME shape as the original. At the hot end of things it impossible to tell the difference from the original 'curve' vs the modified curve. I charted both. It's at that end where the 'goodies' happen.

Note that this is A/C's sensor not the ECT. So we're not fooling anything associated with the engine power train management.

At cold, 20 deg C, my normal CTS looked like 4.7K; with the mod it looks like 960 ohms. This is well outside any normal function for the A/C. Even if it thinks it a little warmer than it is, so what? No harm, no foul!

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

Benzmac,

The other ECUs (LH-SFI and E-Gas) use the ECT (B11/2) NOT the B10/8!!

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

My setup is different in that it is a 2 speed fan and the low speed, [ a/c aux], is energized by high side pressure, not eng. temp sensor.

Arthur,

So does mine. Just added it to the chart. So it responds to high side pressure (in bar) and/or temperature.

So what comes first, high pressure or high temperature? Since high temperature also RAISES pressure, most likely, temperature leads pressure. Also, this changes works just as well with the A/C OFF.

Do you have two sensors or just one?

[Edited by JimF on 04-19-2001 at 08:42 PM]

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

Jim,
I have 2 sensors, but your system seems to be a little more entailed than mine. I'm sure it should be with the V8. I have an E320-124ch.
My other sens is a single pole to ground unit.
I remember seeing a post a few months ago that listed
a sensor mod. from Benz for a lower cut in with part # and all, but for the life of me, I can't find it via Search. I thought it was buy Benzmac, but not sure.

My fan set up [ stock] runs high speed aux. off eng. temp, but a/c is high pres. only. Of course, high temps results in high pressure, but the trigger is still pres vs. temp.
[ A good freon level indicator with that set-up. No low fan with a/c with high ambiant= possible low freon charge.
Lower than optimium high side.]
So your temp indicator gage runs off the eng. management ecu ??
Arthur

Arthur,

Yes, the temp gauge runs off of the ECT sensor (B11/2) but the fans are triggered from the CTS sensor. This is ONLY way this mod can work. That's why I did it!

The LH-SFI and E-Gas ECUs use the ECT sensor as their input and that's what you see on your instrument cluster. I thoroughly checked all of the ECU's input/outputs before electing to do this change. I would think that you car is very similar but don't know for sure.

Actually, the downside to the change is the heavier use of the two fans. If they go bad, they're not the nicest things to r&r.

There's over 42 years of electronic design engineering and investigative physicist training to "pay attention to details". Otherwise, you're a just screwdriver mechanic!

Re the different part number: if the varistor in question is 'changed' to have slightly different characteristics, the fans would come on at a lower temperature. So redesigning or using a different varistor would do the trick.

I had played with another idea, which I think you alluded to in your first msg: use of another varistor. I thought about that but trying to get one that, when taken in concert with the present varistor, would match the EXPECTED response in the receiving ECU makes the problem much more complex. You go from a linear to a quadratic equation. But, once done, it would keep the response identical to the original part.

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

Jim,
I can see you have checked it out . I will do the same .
The ohm/temp chart you posted a few weeks back is different than my sensor , so I will start from scratch.
I saw another post where the guy used a seperate thermistor
and a transistor switch.
As of now , I have the cabin switch bridging the a/c high
pressure sw. I'm the ECU .
Thanks
Arthur

Perhaps you refer to the post by ROLFE. His mod was quite ingenious!

Used a transistor driven from a "deceleration" switch (don't know what that is?). The transistor then turns on a relay which in turn, puts the fans on. So the fans come on during decel and I assume when the car is idling.

That's good but on a real hot day when say driving in the desert, the fans may need to be on more than just during decel in order to cool it enough!

Also the relay better not be a Radio Shack special! Need to handle lots of current. In my case, 35 amps! For one fan, the current is halved, so still needs to handle high currents otherwise the contacts will weld!

The chart that I posted was for the ECT sensor NOT the CTS sensor. There is no data on that sensor so had to get it by testing. The CTS is completely different from the ECT sensor. For example, at 20 deg C, ECT is 2500 ohms, CTS is 4700 ohms. If you get a part number for yours, let me know.

Please let me know the details of your car's system when you find out.

[Edited by JimF on 04-21-2001 at 10:24 AM]

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

Jim,
I have checked my schematics and now I remember why I did not do the resistor .
On the E320 the ETC sens.[B11/3] feeds the HFM-SFI mod-N3/4. That runs the aux fans and other systems.
The other ETC , [B/13, single wire to gnd. ], is gage sens. only. It is on the neg. side of the gage and serves no other purpose that I can see.
Different set-up than yours in the sense that there is no CTS. So that is where you are catching the trigger for the fans without worry about N3/4 management.
I can't do that . My data list no CTS, only pressure at drier for low fan with a/c.
I rarely use the cabin sw. I installed , but I have done this to most of my previous models and there are times
it has been handy.
Would be nice to go the auto route though.
I'm Thinkin' !!
Arthur

Arthur,

when you said the magic words, "... on the E320 the ETC sensor [B11/3] feeds the HFM-SFI module (N3/4) . . ", you said it all. The two system ARE completely different and my mod would not work!

And further, my mod is probably ONLY good for '94 S500 cars. In '95, they combined a "bunch" of modules into one, and went to HFM instead of LH. I'll bet that it won't work on the 95, 96 and on S500s because of this.

Where you temperature gauge is driven from a separate sensor, mine is driven through the LH-SFI module to the instrument cluster panel. The input for this is from the four terminal ECT sensor (B11/2). A lot of differences!!

__________________
Regards . . . . JimF
-------------------
'94 S500 Cpe

Visit my Mercedes Web Page

Jim,
How right you are !!
I just went back to the schematics and on a last page,
here it is.
A 10/8 AC temp sensor. And a footnote:
up to 8/94
from 9/94 > eliminated and rewired B 11/4 for FI.
I think that is when they went to Hot Film over Hot wire.
"Seek and Ye shall find"
Good one.
Arthur

This is too thechnical for me. But I understand this 1.1k resistor mod is not possible in my 1995 E320 W124 M104, right?

__________________
-----------------
Juan
1995 E320 Sedan
2008 Ford Escape
RC Helicopter