Aux Fans - My solution
 

Forum members:

I have received a tremendous amount of valuable information reading posts on the forum. In light of that, I thought I would post this in hopes it would help other members.

I bought my MB about 3 months ago - right at the beginning of summer. I soon discovered MANY differences in design philosophies between MB and japanese car makers. Some I like, and some I don't like. One of the things I particularly don't like is the way the cooling system functions ... in particular - the fact that the temps fluctuate widely between 85 and 105 C.

My first post was a question about aux fans (after reading several hours of prior aux fans posts) and got a very in depth description of the way the circuitry is designed. This was very helpful in acquainting me with the design, but left me somewhat un-impressed with the philoshophy behind the design ... that is that it is OK for the car to run at temps of 100C+.

The responses from that first post started me on my quest...Modify the circuitry so that the engine temps stay within a narrower range. The biggest contributor to temp swings seemed to be the AC. If you have read any of the Aux Fans posts you already know that the aux fans are enabled in the low speed by the closing of a pressure sensitive switch on the receiver dryer. This switch has a threshold level that appears to be reached only when the AC is working "above normal" and producing highside pressures at or above 280psi. However, when the AC is not working quite that hard, the switch is not closed, and the fans do not come on, and the engine heats up, etc....

My desire was to have the fans come on in low speed when the compressor was running. I found a schematic on this site showing the circuitry of the Automatic Climate Control and proceeded with my design.

The prerequisites of the design were.
1. I did not compromise the integrity of the existing circuitry.
2. The mods could be installed by anyone.
3. The mods could be "undone" very easily.

One of the mitigating factors in the ensuing design was that I did not have access to the factory maintenance CD. Therefore, I will readily admit that there is a better way to do this which I will address at the end of the post.

Conceptually what I designed was this.
Since the pressure switch on the receiver/dryer engages the fans, I would simulate that event when the compressor clutch has power to it. thereby tuning on the fans in low speed.

1. The plug that goes to the compresses has three wires. The wires provide power to the clutch (1 wire), and feedback from the compressor to the AC control that the compressor is indeed rotating (2 wires). The plug can be "opened" very easily. I unsoldered the wire from the pin that feeds power to the clutch (blue or green) and removed the pin. I bought a new pin from MB and soldered this wire and a "new wire" to it and put it back into the plug, closed the plug and re-connected the plug to the compressor. I now have my new "signal" wire in place.

2. The next step provides access to the pressure switch so it can be sumulated to close. There are two wires on the pressure switch and on my car (92 300E) they run about 6 inches before connecting to a wire harness. I unplugged the wires and attached adapters to the male ends of the connectors on these two wires. The adapters are 1 female to 2 male and provide the ability to attach back the original wiring harness, and the two new wires that will make up part of the new circuit. MAKE SURE THEY ARE INSULATED!!!!! There is 12 volts on one of the wires and on both wires when the pressure switch is engaged.

3. I then ran 3 - 14ga wires to my new circuit. The new wire that was added in step 1 (the compress clutch), and two new wires that will be attached to the adapters added in step two. The circuit merely takes the presence of 12V on the clutch to turn on a relay that simulates the closing of the pressure switch on the receiver dryer.

4. I went to Radio Shack and bought a 12V relay capable of handling 30A and mounted it near the battery. There is a nice open area there and is somewhat isolated from the heat of the engine. After mounting the relay, I ran the wire from step 1 (compressor clutch) to the relays activation terminal, and ran the two wires from the pressure switch to the relays contacts. That leaves one terminal left on the relay, and that is the ground. I found a small hole already drilled by the strut mounting and attached a lug to the hole with a machine screw and ran a wire to the ground terminal on the relay. MAKE SURE ALL THE CONNECTORS ARE FULLY INSULATED!!!!!!!!

5. The "turn on" voltage and current for the relay are 6V and 160ma respectively. In order to minimize the current drain on the circuit feeding the compressor clutch, I put a 50ohm 1 watt resistor in series with the ground wire mentioned in step 4. The value of 50 ohms was derived from the specs of the relay. The coil portion of the relay has a resistance of 66 ohms. Adding a 50 ohm resistor in series will leave just about 7 volts across the coil and will therefore engage the relay.

Now when the compressor is on, the fans run in low speed. After 3 weeks of testing, I have had no problems.

It is quite easy to disable the circuit. Simply remove the wire from the compressor clutch going into the relay.

It seems a better way to do this whole thing would be to locate the wire feeding the compressor clutch and add another feed to the low speed aux fan relay in the fuse box. Not knowing where a lot of this stuff is kept me from doing this.

This mod coupled with the mod that adds a parallel resistor to the temp sensing unit that turns on the aux fans in high speed (located somewhere elsewhere on this site) should keep the engine temps confined to a narrower fluctuation range.

I realize that doing this will conflict with philophies of a lot of members.

If you would like more details, I can provide as well as some pics.

Hope this is helpful.

If anyone would be so kind as to tell me where the O2 sensor is on my car I would appreciate that info.

Regards,

David Lee

need your extended instructions, please
 

I have a 560 sl that does go up in stop and go traffic. Here in Texas, (95 degree spring and fall days- Summer? Don't even ask!! ) we have above than normal 90 degree days. I would really appreciate it if you can e-mail me a detailed intruction, and the exact part number for the relay that I have to purchase from RadioShack.

Regards

I would like to see more details. Sound very interesting. Thanks.

Your O2 sensor is most easily located from under the car. You can't miss it.

a2aa, haasman, tkamiya:

First let me apologize for the slow response. I have been out of town for a week and without access to a computer.

Allow me to address all 3 posts.

1) I do not have a problem with temps getting to 100 on days I do not need the AC. I get to about 95 and that's about it. I have designed a circuit that I haven't yet implemented that will utilitize both the high and low speed fan relays and work in conjunction with the temp sensor much like the resistor mod located on another thread. In fact, I got the inspiration for the design from that very thread. In short, it will turn on the fans in low speed at a determined temp (adjustable), and then turn on the fans in high speed at a determined temp (also adjustable). I believe the newer MB models have this type of two speed control built into them.

One thing I did notice after changing the thermostat was that my temps run cooler than they did prior to changing the thermostat (82-85). It could be that the thermostat was marginal, but I rather think the lower running temps are due to the fact I replaced the coolant with a 40/60 mixture instead of 50/50. This of course should cause the engine to run slightly cooler with the reduced amount of antifreeze in the system. I'm sure you've read several discussions on this subject so I won't comment any further.

2) I would be happy to supply all the part numbers, as well as a picture (providing I can figure out how to do that), and will do that first chance I get next week (just got back in town and am going to the winston cup race tomorrow here in Charlotte).

3) Thanks for letting me know where the O2 sensor is. BTW, I bought a MB service CD last week.

To date I have not had any problems with the mod, and it has helped keep the car considerably cooler with the AC on. I live in Charlotte, NC and it gets pretty hot and humid in the summer. There's more that can be done, but for me, this is a start.

David Lee

OK, somebody enlighten me : " I was not happy with temps exceeding 100C." Well, if the car is designed to run up to 105 or 110C (like my diesel), let it!!!! Why second guess the engineers??? No offense, but I think a single engineer (or a team) from the MB R&D knows more about the car he/they designed then all of the members of this site put together. I think your "prejudice" is rearing it's ugly (or at least expensive) head here. "Other cars do run like this, so MB should too." Why???
Somehow, I trust 50 years of German engineering (although I do not like Krauts - they killed over 6 million of my countrymen and that is hard to forget. By the way, I'm Polish, not Jewish).

If MBs suppose to run like that-let them. They were meant to do so by people WAY smarter than us. Otherwise, just buy a rice-rocket.

By the way, this statement is in now way inteded to offend anyone, and if it did - I appologize.

Interesting comment. If I used that same logic, then I should assume it would be OK for the thermostat to have a wide fluctuation threshhold temperature as well. Yet the thermostat appears to open at 82deg by design.

Using your logic it would be good to put a 100deg thermostat in the car.

I'm not going to argue the point. I have presented my reasoning and I believe it is sound. What you appear to be saying is that MB engineers always engineer components and systems better than anyone in the world. One would just have to look at their new car satisfaction surverys to see that is not the case.

There are components in the car that are engineered magnificently. There are components that are engineered very poorly and leave consideralbe room for improvement. That's not just my opinion, but car manufactureres as a whole as evidenced by systems designed by high-end car makers.

A good example of a pathetically designed component is the OVP relay. Not even a first year EE major would come up with a design like the one in the OVP. What makes it stand out is how important this component is to the proper running of the entire car's system. To compound the frustration is the fact that MB has taken 2.00 dollars worth of parts (I am being generous) and put it in a package they mark up several thousand percent.

Would you have criticized my post if I had suggested that you can buy a set of struts that would give a better ride? After all, it's just another engineer trying to improve on something. Or has MB already designed the perfect ride and there is no room for improvement?

As for expensive, the parts were 10 dollars.

You have stated your opinion not based on data, but your own "prejudice" as evidenced by your "rice rocket" comment. Why do I have wire insulation that is cracking? Because either the wire harness is poorly made (possible room for improvement) or because the engine temps are getting hot enough causing it to break down and crack.

When someone can give me some data showing it is "bad" for a M103 to run between 82C and 95C then I will acknowledge I was wrong and remove my mod. Until then it stays and continue to be improved.

Sincerely,

David Lee

<<One thing I did notice after changing the thermostat was that my temps run cooler than they did prior to changing the thermostat (82-85). It could be that the thermostat was marginal, but I rather think the lower running temps are due to the fact I replaced the coolant with a 40/60 mixture instead of 50/50. This of course should cause the engine to run slightly cooler with the reduced amount of antifreeze in the system.>>

You mentioned an 82C thermo.
The original spec. for you car is an 87C.
That is start to open spec.

This may attribute to your temp change more than the a/f ratio difference.

Authur:

I did buy the thermostat at the MB dealer, so I assume it is what it's supposed to be. I said 82C based not on knowledge, but on the observation that my gauge registers around 82C in this fall driving weather (around 70F to 85F)

I believe the gauge to be correct because the high speed fans would come on at the proper time prior to putting in my mod.

Anyway, thanks for the correction.

David Lee

I see..
You may want to check both the temp gauge and thermostat.
The temp gauge/sensor should read accordingly:
60C-110 ohm
80C-67 "
100C- 38 "

It is surprizing to have that motor run at those temps with an
accurately funtioning 87C thermostat installed.

a2aa:

The radio shack part number is 275-226 and it is called a "30 AMP AUTO RELAY".

I took some pictures, but they were too big to post to the site. I'll have to retake them with reduced resolution. I will do that later (I'm off to my granddaughters birthday party).


Aurthur:

Thanks for the info on resistances. I'll check that out. It could be that I'm interpolating incorrectly when reading my dash gauge.


David Lee

Anyone noticed that you can actually run the engine with a more advanced ignition timing when the coolant temp is lower, especially below 100C. With a more advanced ignition timing, I'm getting better performance in terms of power and fuel economy. I'm don't know about MBs designing their engines to run up to 105C/110C but I'm certainly getting better performance when the temp is around 85C-90C in a tropical climate. If MBs were designed to run in such high temps, why the need to have different parts (temp sensors, fan clutch, switches, etc.) for cars in tropical climates? I was also experiencing wildly fluctuating temps (85C right up to 120C) before using the appropriate parts for the tropical climate. After the upgrade, temp remains around 85C to 90C in ANY conditions. IMHO, with a proper cooling system, temp fluctuations should be kept to a minimum.

what upgrades were they? nowadays, temps in some parts of the US are tropical specially in the summer where we can get 100F+. that is hotter than outside temps in many tropical locations. i believe that these cars were engineered for places like Stuttgart where summer highs are in the 70's. i am aware that there are components that can be replaced with tropical versions - the viscous fan clutch for one. was that one of the components?

David, nice idea but a
 

lot of extra work just to turn on the aux fan(s) a little bit earlier.

For my car, for example, the following holds:
Aux Fan Switch Points: ===== 1st ====== 2nd ======3rd
Refrig Press ON (bar)[lbs] ==14 [206]===17 [250]=== 20 [294]

So 206 lbs is very little pressure, so the Aux fan(s) start pretty much right after the A/C is started. I'm sure that your car is somewhat close to those numbers, so you went to a lot of extra work, $ cost and complication to turn the Aux fan(s) on a little early.

Very few people can sucessfully complete your results. I, as you know, make the Cool Harness, and what could be more simpe than that??? Then why do I have so many customers? And I have how-to-do pictures!

One of negative: Consider changing the relay to a more robust industrial grade relay for reliability. The 'weak' link in your design is that RS relay! If it fails, then the fans won't start! If you wan't my 'famous' RS story, let me know.

Anyway, it will do exactly what you wanted it to do, so enjoy!

Here is my $0.02:

Northern Europe, and Germany in particular, is a cold, generally wet place to live and drive. Lots of snow in the long, dark winters. High speed highways, and winding, narrow mountain roads, as well as tight Medieval City roads. All these features bred a number of special automobiles that we all seem to like because they handle well, brake well, have superior lighting systems, great turning radii, and superior stability at high speed.

High ambient temperatures are absent from the development equation. Stop and go traffic in high ambient temps are also not a part of the design's heritage. Neither was airconditioning until it became a "standard" option on Japanese cars, when MB, BMW, Porsche, Audi, VW and other well thought of European premium cost cars were forced to figure out how to make an effective HVAC system. I am not sure the present day system is really that successful yet.

So, I contest the statement that all those German engineers really designed the MB to run at 100 plus degrees C. They are much more focussed on making the cars work well at 100 plus mph.

To run cool at high speed you need a certain flow rate of coolant and air to remove heat from the radiator and the hot coolant needs a certain dwell time to get rid of its heat load. These features are juggled to try to get as small a radiator and fluid volume as is practical to reduce the frontal area (lower drag) and avoid carrying any extra inventory of coolant (weight). The air flow is a combination of the belt driven fan operation, and the high velocity of the air entering the radiator due to the car's speed. The coolant flow rate at speed is achieved with a pump of a particular size and fluid passages set to give fluid velocities that avoid damaging cavitation and erosion while remaining sensitive to weight concerns. These conditions typically set one or more of the design conditions for the system (high speed is not actually a typical high load point - high loads are all typically transient events unless you are racing).

Other operating points kind of have to suffer the consequences as long as they are deemed to be tolerable by the engineers. Unfortunately, all of this does not balance out to make the situation at idle or very low speed easier on the coolant system, especially when the ambient temperature is high (lowest air flow, lowest coolant flow, and lowest "delta T" between the coolant and the air).

To address the shortcomings of the system capacity at these "off design" conditions of high ambient temps and sitting in traffic (and having an a/c system installed), a bunch of Rube Goldberg junk is added. The things that are added are, an oversize belt driven fan with a complicated clutching system (electric or viscous, and either is a big dollar item to replace) to prevent the fan from eating too much power at high rpm (where it is not needed) or flinging itself apart. Then there are these electric fans with sensors and resistors and multiple speeds and so on, that will probably last for a few months if you run them continuously because they are not rated for continuous duty.

So, I believe the cars are intended to last longest and perform best at 80 degrees to 87 degrees C, and intermittent operation at higher temps is considered tolerable for the engine. It is definitely not a design feature to run there or the thermostat would be adjusted to run there all the time. Jim

jimF
 

jimF

I was able to view jim's design and it focuses on a 2 sensor design, correct? What about the cars with a one sensor design such as the earlier model 107's will this work as well??

David L,
Thanks for the help.