Leaking coolant from where? One *crazy* mystery!
 

Now here's a mystery for the really brave...

My vehicle is a 1985 300CD-T (Calif.). Mileage is 195,602.
Issue: Dried residue from coolant appears on the transmission pan and the support panel for the driveshaft's intermediate bearing. I have attached three pictures.

Why do I think that this stuff is dried coolant residue? Because it has the whitish-bluish color. The new M-B coolant that I use is blue colored.

When does this residue appear?
Only after driving the vehicle on the freeway for 10-15 minutes and around 60-65 MPH.

How much coolant disappears from the coolant reservoir?
Negligible. The fluid level barely appears to drop.

All of the original cooling system hoses were replaced in July of 2011 at 178,394 miles.

I have checked the tightness of all of the cooling system's hose clamps and tightened a bit more if it seemed okay to do so.

I have used a dentist's mirror to look completely around and underneath all cooling system hoses, metal lines, and connections and have found no sign of coolant leakage anywhere.

My current guess:
Coolant is leaking from around an exterior and rear (near the firewall) portion of the head gasket from a very small spot that only leaks when there is high pressure in the cooling system (read: highway driving at highway speed). The coolant then leaks downward, blows onto the underside components (i.e., transmission pan, support panel for the driveshaft's intermediate bearing) and dries. :(

Other thoughts?

I am no expert on coolant splatter, but it doesn't seem to me that you would have such a widely dispersed pattern from a drip at the rear of the engine.

M best luck finding coolant leaks has been with a pressure tester. With the car sitting still, you won't have the splatter issue. High on my list of suspects would be the radiator tank seams.

Perhaps Iis condensed water from running the air conditioner. I agree the splatter looks suspect. Keep checking the coolant level I would worry if it drops significantly after a week.

I'd also suspect the lower seam on the radiator.

The upper seam on mine has intermittent small leaks and creates a similar spray patter on the upper components under the hood, sprayed around by the airflow there.

Hey, thanks for all these great replies! :)

Now then, a little feedback...LOL :P

I replaced the radiator with a new Behr unit and also replaced the auxiliary water pump at 178,394 miles (July, 2011 - the same time that I replaced all of the cooling system hoses). I have checked the auxiliary water pump for leaks and detected none. I have not inspected the radiator and its seams with a fine tooth comb, so to speak. I will do that.

Doing a pressure test had slipped my mind. That is a great idea! I will probably use a local M-B shop to have them do that.

I will say this: A few days ago, I scrupulously cleaned the area where the dried coolant was everywhere I found it, let the engine idle for about 20 minutes, and could not find a singular drip. This must of had to do with the fact that the cooling system was not fully pressurized. Correct?

Great observation about the pattern of the dried coolant being too dispersed for a drip to be originating at the rear of the engine. That is the benefit of other minds thinking about this! :D

As for running the air conditioning vis-a-vis condensation: I have not been using the air conditioning...

Where I live, just about any chain auto parts store will lend a pressure tester. It's just a matter of replacing the coolant bottle cap with the tester fitting and pumping it up. Then listen and look for leaks.

Ah...Good to know! I'll stop there tomorrow and check it out! :)

If and when I finally track this dang leak down, I swear, I'll be happier than a pig in...well, you know... :P

I would guess the water pump, the coolant is dripping out of the weeping hole on the pump....dropping down and then splattering back. It may not be to the point of no return and that is why it only shows up on a high speed freeway trip...

A few thoughts:

1) What color is your washer fluid? Mine is blue...

2) If it's really antifreeze, I'd suspect a leak inside the heater core. This could easily find it's way out the a/c condensate drains, in the right general area.

Good point. Well, I will go take another look at the water pump and weep hole.
However, I replaced the water pump and gasket at a very recent 193,360 miles (November of 2013).

The color of the washer fluid is clearish and slightly yellow. I pulled up the reservoir and found no evidence of liquid underneath or around the reservoir.

I hadn't considered the heater core.
Where might I check inside the car for possible leakage of this unit?

Should be smelling the hot coolant in the car if the heater core. That's with heat on of course.

Okay.
Well, I haven't run the heat in a while. I will do so and report back.
Thanks!

I borrowed a pressure tester and my results were hoses not tight enough at:
1. Metal line off of thermostat housing
2. Hose at top of radiator
3. Large hose at bottom of radiator
As I mentioned above, I replaced the radiator fairly recently and I think that I was overly cautious about not over tightening the hoses to the radiator.

Anyway, I tightened the three hoses. Then I left the system pressurized for about 30 minutes. The pressure dropped about 1-2 PSI over that time. Unfortunately, the problem still remains.
I took the car out onto the freeway twice and both times, there was plenty of dried residue on the transmission pan and intermediate bearing mounting bracket.

My guess at this point is a heater core leak.
I currently, have some other climate control vacuum problem so when I tried to blow heat, I just got cool air. Incidentally, I recently replaced the mono valve unit, not just the insert, for a cool $380 from M-B.

Is there any way to check for damp carpet around the heater core at the passenger's feet (that is where the heater core is, isn't it?) or elsewhere for a heater core leak?

Also, is the fact that the pressure dropped 1-2 PSI while I left the system pressurized with the tester for about 30 minutes indicative of the leak?

Not necessarily a big enough leak to be of any consequence. If you stopped the visible coolant leaks, you have likely solved any problem worth worrying about.

Also, the heater is valved on the downstream side, so it doesn't need to be "on" to leak.

So, do you mean that under all conditions, there is coolant flowing through the heater core on its way to the radiator?

The coolant won't be flowing if the monovalve is closed, but the heater core will be subject to the pressure present in the cooling system.

Unless one's sense of smell is significantly compromised, detecting a heater core leak while driving won't be difficult.

Ah, okay.
And is this right?: The monovalve is closed by default (no power is applied to it) but opens and allows coolant to flow to the heater core when power is applied (pushbutton unit is activated into Heat mode)?

-------

My sense of smell is working just fine.
Which then tells me: If I am not smelling coolant in the cabin, then there still must be a leak of coolant somewhere in the engine compartment.
Sheesh...what a puzzle and nuisance!

I still have the cooling system pressure testing tool so I may take another stab at the problem by re-pressurizing the system and looking for leaks again...

The monovalve is closed electrically and is spring-loaded to the open position when not powered.

Ah hah!
My understanding was only completely opposite of the facts. Facts duly noted. Now eating humble pie.

Thanks for steering me right!
:-)

and if measuring with a meter, the monovalve will have +12v all the time the key is on, but the -12 is modulated by the climate control to energize the valve closed...

If it really bothers you, switch to NPG and run the system at zero pressure. It would be a lot easier (and cheaper) than changing a heater core that is probably not leaking. If it was leaking enough to drip from the evaporator drain, you would definitely smell it and your windows would be fogged with the residue.

vstech:
Thank you for the info. about the electrical. Got it.

rscurtis:
I'll plead ignorance: What would you mean by "NPG"?
Also, yes, I agree, if there was a heather core leak, I would likely be smelling the coolant and there is no such smell in the cabin.

Update:
I was conversing with a long-time M-B tech. in my town today and this is what he told me:
1. The expansion tank cap may not be holding pressure and could be causing a little leak. Consider replacing it. It was last replaced 9 years and 44,000 miles ago.
2. Check the tightness of the clamps for the hose coming off of the expansion tank.

So, I have a little more detective work to do.

Thanks-

Just for kicks, stick the probe of a volt meter into the coolant and ground the other probe. Any voltage approaching .3V indicates electrolysis is going on and the coolant needs to be changed. So says the destructions in a heater core that I had. The core manufacturer was obviously trying to prevent premature warrantied failure.

NPG- non-aqueous propylene glycol, or waterless coolant. Home » Evans Cooling. Watch the video with Jay Leno.

Where is the evaporator drain on a w123? ThNk you.

The drain is inside the car, just above the transmission tunnel, behind the dash. There is typically a short foam piece that carries the draining condensate out of the car through a hole in the floor. It drips on top of the transmission.

Sometimes the pressure tester itself leaks. Test again and consider it fixed unless you find wet, lose fluid or smell coolant.

So, I have pressure tested the system three times:
The first time, I found several hose clamps (as I mentioned above) that were leaking due to not being tight enough. I tightened the clamps and eliminated the leaks.
The second time, I pressure tested the system about 14 hours after driving the vehicle. I kept the pressure on (none added) for about 30 minutes. No leaks found.
The third time, I pressure tested the system about 18 hours after driving the vehicle. I kept the pressure on (none added) for about 30 minutes. No leaks found.

As I also mentioned above, I plan to replace the expansion cap and check the tightness of the hose connected to the expansion tank.

At the present time, no signs of coolant found anywhere, no smells of coolant.

Good stuff from the sound of it.

I certainly will be relieved if the dried coolant residue turns out to be originating from the expansion tank cap or hose.
Eliminating this leak will mean *no* leaks from my car. I've taken care of all of the rest of 'em! Not bad for a 29 year old M-B with 195,000 miles! :P

Another thing you could try if you haven't already is go down to a 7# pressure cap. it's a lot easier on the radiator and heater core. I run NPG in my signature car at 0 psi and it works great.

What kind of coolant are you using that is white? The correct Mercedes coolant is pale yellow. Those splatters looks like like white paint. Are you sure it's coolant?

Take a few drops from the expansion tank and put it on the radiator neck, let it dry and take some pics and see it it looks the same.

Hmmm, I am not that familiar with this.
So, do you just go to a McParts auto supply store with the old cap and tell them that you want a cap that fits (or is that what the 7# designation means?) that will create 0 psi in the system?

This does raise this question in my thinking:
Why would M-B use a cap that keeps the system at 1 bar (approx. 14 psi) if it would be easier on the radiator and heater core to use 0 psi?

Thanks-

I am using the latest M-B "Anticorrosion/Antifreeze AGENT" (part number BQ 1 03 0004) acquired from M-B in May of 2014. It is colored blue.

Lowering cap pressure lowers the boiling point of the coolant. I have no experience with NPG coolant. I have read NPG has serious issues in very cold climates. Is that true? I assume you live in Florida?

Why did you change from the spec MB coolant or Zerex G-05? The newest MB coolant is for aluminum head engines and may not be the best your cast iron 617.xxx engines.

Are you 100% sure those white splatters is from the coolant? Will you do that test I suggested?

How flammable is it? Anything raised to a high enough temperature will burn. If you put a wick in the antifreeze mixture and can't light it with a match or a spark, don't worry about it.

Is that true? I assume you live in Florida?

I have not had any issues even at 2* F. I'm not worried about the boiling point, it boils at 370* F. The original racing formula did have some viscosity issues.

5cylinder, sorry for the confusion. The 7# cap reduces the pressure on the system. I run mine at 0 PSI by leaving the cap loose. The reason for the higher pressure from the factory is to guard against the most extreme temperature situations. There is no benefit to pressurizing a cooling system more than necessary. All that extra boiling point won't be doing you any good when you're stuck in traffic with a blown hose or a split radiator tank.

PG is NOT flammable. It was easy to do a bit of research. Per the NFPA, a liquid needs a flash point temperature below 100°F to be considered flammable. Anything above that is simply combustible.

Gasoline is, obviously, flammable. It has a flash point of -45°F.

Diesel is not considered flammable. The MSDS for D2 refined by BP says it has a flash point of 100.4°F. So it's considered combustible, just over the edge. How many of us have D2 leaking on our hot engines, without stressing?

The MSDS for PG shows a flash point of 228°F. And it doesn't auto-ignite until 700°F.

So, will PG burn? Of course. Anything that is combustible has the potential. But I'd suggest that having a leak of PG in to your car's cabin or even on to the engine is not a fire hazard. If your car caught on fire for other reasons then sure, the PG would ignite given enough energy/heat from said fire and perhaps make it worse but it's moot, then, as your car was already burning down!

Wouldn't leaving the cap loose allow air (oxygen) to enter? Oxygen promotes oxidation (rust). Isn't that not good for cast iron engines?

Got it. Thanks for clarifications. The coldest environment that my car experiences is, say, 30 degrees. Rather mild. And that is inside a garage.

Ok I have a 602 engine with an aluminum head am I to believe g-05 is not what I should be using?

Indeed, NPG+ from Evan's has a flash point of 233*F or 111*C. Again, WAY higher than diesel fuel. I don't know anyone here who stresses about a diesel leak on their engine. I've had a return hose that broke in half an was pissing diesel all over the side of the block. Nothing even remotely problematic.

I know that a diesel leak CAN be dangerous, I'm not saying it isn't.

However I would say it's much more dangerous than a waterless coolant leak.

You would need both that temperature, and an ignition source. That's usually lacking.

My point is that NPG is safer than many of the other fluids we use in our engines and my initial response was to point out that it's in no way "flammable" per the definition.

And by the way, I don't use NPG, don't plan to, not even really interested in it. But 10 minutes of research debunked the claim that it was in the flammable category, much less even as ignitable as diesel fuel. You're acting like it's the same as running gasoline in the cooling system.

I think the key here is being more clear and honest with the language used. You keep saying it's flammable. NOT true.

What it is, is IGNITABLE. I.E. when all factors come together it can ignite.

In case you're interested, PG is also ignitable at concentrations over 40% in a PG/water solution. Most people run higher concentrations, maybe 50/50. The NFPA banned this practice in fire sprinklers after a case where anti-freeze in the water from a fire sprinkler made a grease fire worse. That's a situation where all factors came together and the PG in the water burned.

Are you going to stop using antifreeze and just run plain water? Or take a calculated risk?

No on two counts. First, even with the cap tight, there is air in the system. Second, the "N" in NPG stands for "non aqueous", i.e., there is no water in it. No corrosion, cavitation, or electrolysis.