Re: Finding coolant leak and checking temp gauge - Page 2 - PeachParts Mercedes-Benz Forum

juribe2 · #16 12-21-2006, 10:32 PM

Hi Mike,

I don't think you have oil in your coolant. Belive me, when oil gets in the cooling system, you'll get a real mess and your expansion tank will look like filled with butter. I'm telling you from experience.

I think what you see in your coolant is just some minor contamination for water used to fill the system or maybe some corrosion from somewhere in the system. I think you can leve it like that and try to find the leak first. Later you could do a flush and replace coolant if you feel unconfortable with that.

Temp at idle should be close to your thermostat setting and could get hotter when driving up a hill or during hot summer days. My thermostat setting is 87C and my temp reading is usuarlly below 90C but sometimes it gets up to 100-105C when driving on a mountain road or something like that.

I just want to say that I like your plan a lot. I think you should bypass all the heater system and check if the car holds the coolant level for a couple of days. I think It will. That way you can be sure where the problem is located. I think that running with the heater core bypassed is ok and you can use your car that way but I also think you could expect a couple more degress in your coolant temperature than the thermostat setting because the heater core would probably help remove some heat from the coolant.

Go ahead. Bypass the heater core and let us know what you find. I hope you get a prompt answer !

michaeld · #17 12-22-2006, 06:54 AM

juribe2

Hey, thanks for your words of wisdom. I think that's where these forums shine best; guys who may not be mechanics, but they went through the same thing themselves and in that area they now now what they're talking about. And then they enlighten a forum member whose going thru it for the first time.

I've always heard to be aware of oil in the coolant, or coolant in the oil. But I've never actually SEEN a car or a pic of a car in that condition, and then when I see something a little out of the ordinary, I realize I'm not sure what I'm looking at. Thanks for helping clear me up.

The servo system relatively easy to bypass. Just take off the servo and insert copper pipe where the servo was to connect the (two?) hoses. The heater core would be more difficult (at least for me) because I can't see it. I've been told that I could bypass the heater, but would rather not do it for long because 1) I'd lose my defroster and 2) what you said about the car running a little hotter. The servo I would try w/o for a while and see what I thought...

I'd also rather do one at a time for the sake of that "process of elimination" thing. But yeah, thanks for your vote of confidence in my plan. It always makes me feel a little better when other car people like a plan than when they don't think it'll work.

I've been dealing w/ a gauge that has been reading hot since about August. I had a new radiator core installed about a year plus ago. I replaced the thermostat and the upper/lower hose. The water pump is less than 5 years and 10000 miles old from the PO. And the fan clutch is a veteran part but seems to be doing its thing as well as ever. At this point I either have a bad gauge system or I'm overheating somewhere. It could be the aux water pump, or the water pump. The car is clearly not overheating at idle, which is another reason to rule out the fan clutch.

Expansion tanks are great; the only down side to them is checking out a water pump. There, they really suck! I'm just not sure how to check one out in this car (and, if it turns out to be bad, neither do the mechanics I took the car to!).

So, yeah, bypassing what seems to be leaking and installing a different temp gauge. Those will be my two tasks. I'm hoping I had a bad gauge, and the servo just picked a coincidental moment to go. If I am still overheating after doing those things, it's time for that trip to the mechanic to find out once for all why I'm overheating. It just seems to be beyond my ability to diagnose.

I'll be sure to keep everyone informed for the record, so that the next guy who experiences a similar problem will be able to benefit from MY hard-earned wisdom.
Mike

juribe2 · #18 12-22-2006, 10:59 AM

Hi Mike,

About 2 months ago I had a bad heatter hose and It was not possible to get it in my country, so I had to order it online and wait a couple of weeks to get the part. My mechanic bypassed the heater core so I could use the car during those days. He just removed the hose (going from the engine block to the heater core) and intalled a couple of rubber stoppers with fasteners. I had no problem with that. My car usually runs under 90C but with the stoppers installed it ran between 90 and 95C which is safe to the engine.

I don't know how to tell if the water pump is doing it's job. I had to replace mine about 5 months ago but It showed a sever coolant leek. After removing the pump, I could see the impeller very damaged.

I'm not sure. Just an idea. But I think you could remove the thermostat and close the thermostat housing again. Then, fill the system. Start the engine cool and remove the upper radiator hose for a very short time to see if the pump is actually pumping. Turn the engine off before it gets hot. Install the thermostat and radiator hose again. Just an idea but I'm not sure if it will work. I was just thinking that the pump takes the coolant from the lower radiator hose and push it thru the upper radiator hose but when the thermostat is open. Maybe leving the thermostat and removing the bypass hose will show you the same thing. I'm just not sure.

meltedpanda · #19 12-22-2006, 01:16 PM

I think your temps are fine, go after the servo idea, I somehow, once again

missed the discussion that you actually saw hissing in this area. Th eAux water pump only serves to circulate water at idle, easy to check , apply 12V and listen for the spin
Good luck

michaeld · #20 12-22-2006, 10:57 PM

Well, I got a little update tonight.

I removed the air inlet grille (cussed it off might be more accurate) and took just enough of the panel under the glove box off to see back to what I'm pretty sure is the inside firewall. I saw absolutely no evidence of a coolant leak at either location. And - if there HAD been a leak in the heater core, I would think that I would have seen SOMETHING there. I pulled back the passenger floor carpet all the way to the rubber, and saw nothing.

If anyone has had a spectacular failure in their heater, could you tell me where you saw it leaking? I don't want to tear the whole dash apart to find something that probably isn't even there (i.e. a leak). I am 98% sure that my leak came from the ACC servo (even though I can't see any external evidence there, either - at least since I cleaned up the coolant). I don't want to break any more of those &@$*^# plastic retaining plugs than I have to.

If I CAN'T locate a leak in the heater system, I will be bypassing the servo, installing a different water temp gauge, and driving the car around until I see either a) a leak or b) a hot gauge. If I spring a leak, hopefully I'll at least be able to verify its location for certain this time; If I'm still running hot (on the new gauge), it's to the mechanic for a full diagnostic.

As for verifying the water pump, PLEASE GOD LET THERE BE AN EASIER WAY!!! I remember when I replaced the thermostat just lying under the car intermittently cussing and crying as I tried to get everything off and then everything back together.

If I even AM overheating, I am doing so only while driving, and not at idle. I guess I'll know whether or not I'm running hot when I stick in the new gauge.

Here's a question re: the aux pump. Where is it? I believe it's under the battery box, but I'm not sure.

Mike

michaeld · #21 12-23-2006, 12:14 AM

Great googley moogley!

I was just looking at the engine trying to figure out where juribe2's mechanic put his plug to bypass the heater core, and suddenly I saw something clearly pertinent to my problem.

I had stuck a UV dye in the system to better find the leak, and ran the car primarily looking at the servo area. The UV dye has a "glow" even w/o the UV light and glasses. Well, tonight I just glanced at the water pump, and BLAMMO! I spotted a leak coming from the joint between the pump itself and the upper housing (where the upper hose connects). I had NEVER seen it leak there before, and believe me I have looked a bunch! Maybe I needed the dye in the coolant to see? Anyway, I tightened the bolts up, and will clean up the leak and run the car to see if I get any more coolant there.

Pumps will leak before they fail, but would they leak there (at the joint where the pump and housing connect)?

Here's a pic:

The leak only seems to be occurring at the pump/upper housing (where the 212F switch bolts in) connection, and then dripped its way elsewhere. Is THIS the divine sign that my pump has failed?

meltedpanda · #22 12-23-2006, 05:17 PM

I have never seen a water pump fail and leak at the seal between the pump and the block. It could happen but it would be very unusual. They often just show dimise but leaking out the weep hole. Unless I am looking at this wrong it looks like the fluid is above the pump housing???

If so we may be back to head gasket, crack.

Heater core failure presents itself with fluid into the cabin floor and most definitely around the core

michaeld · #23 12-24-2006, 02:43 PM

Quote:

Originally Posted by meltedpanda

I have never seen a water pump fail and leak at the seal between the pump and the block. It could happen but it would be very unusual. They often just show dimise but leaking out the weep hole. Unless I am looking at this wrong it looks like the fluid is above the pump housing???

If so we may be back to head gasket, crack.

Don't say head gasket crack! DON'T SAY IT!!! Dang. Too late. You said it.

I was so filled w/ angst when I read your post that I went back out and took another good look at my water pump. What I saw made me feel a little better. It does not appear that the leak is occurring between the block and the pump, but rather on the pump itself, at the joint between the pump and the upper hose housing (where the 212F switch screws in). This joint - upon panicky inspection - is higher than the block/pump connection (in fact, higher in elevation than anything else that shows coolant), and it would be unlikely to have fluid if IT were not the thing leaking.

The water pump on my car consists of three parts. The pump itself bolts to the engine, and then there are two housing cases that bolt onto the pump which mount to the rubber hoses. It is the joint between the pump and the upper housing case which appears to have the leak.

When I ordered a thermostat, I additionally ordered a gasket. When I installed the thermostat, there was no gasket, and the gasket did not fit the pattern. I thought I'd just received the wrong part. Now I'm wondering if that gasket is for the UPPER housing (the thermostat installs in the lower).

I will run the car again today, and this time I will be watching that water pump like a hawk to see where the first dribble of a leak occurs.

But the discovery of the water pump leak has me wondering if my failure occurred there, or at the servo? When I opened the hood, it was the servo I saw hissing and spitting coolant (even though I can't find a leak in the unit now). When I drain the coolant to work on the pump housing, I suppose I will go ahead and bypass the servo, figuring that it is unlikely to have survived (functionally) after what I saw it doing (hissing and spitting coolant).

Or has someone seen a servo hiss and spit coolant (and apparently dump a bunch of it in the process) and then continue to work afterward? I'm thinking it's fried, and the water pump is not the true source of my significant leak.

Another thing I'm wondering is if the second pressure tester I used WAS actually pressurizing the system - but the gauge dial was bad and so showed no pressure increase (i.e. the dial never moved from zero). I pumped and pumped on the thing. When I looked at the tester unit, it appeared that someone had either dropped or pounded on the cap, so that it wouldn't seal. But maybe it did seal. And maybe I overpressurized the system and that water pump/housing connection was the thing to go?

meltedpanda · #24 12-24-2006, 10:56 PM

in the picture the brass sender is your engine temp interlock - correct? This send signal to ACC to turn on if you are calling for heat. Is the green line fluid?
Assuming that it is, is that the termination point or beginning point, It still looks to me that it is coming from behind the thermostat housing?
You did not crack the housing when you changed out the thermostat?
YOur sure you got a good seal with your gasket? Did you clean both surfaces before putting it in?
I dunno this is a puzzling one, keep looking

you will find it

milterino · #25 12-25-2006, 11:59 AM

On mine, 1976, There is a short rubber hose about 2 inches long 1 inch diameter which connects the top of motor to the top section which the top water hose connect to. It's hard to see and mine has a small leak which I have yet to get to. Check that. Also mine has a shut off valve (after market installation) which shuts off the heater circulation completely and it runs fine with the circulation cut off.
I would like to thank everyone who has helped me over the past year and wish you happy holidays. Milt.

michaeld · #26 12-25-2006, 09:32 PM

Hi, guys.

The short question is: would someone do me a big favor and squeeze your radiator hoses after the thermostat opens and tell me what you feel?

It may just be too hard to describe - or even show- exactly where the leak is occurring on the pump. I did my best to clean the pump and surrounding area, and ran the car for 15 min. I could not see any leaks at all even using my UV dye light. I know I have to reseal at least some part of the water pump. The pump-engine seal does not seem to be a problem (but now that you mention it, milterino, my short hose might have a tiny leak, also).

Here is my bigger issue:

I want to verify the operation of water pump. I found a leak in the upper housing, and need to at least go in and replace the gasket. But do I need to replace the pump itself?

How does one test a water pump in an "expansion tank-equipped car?"

According to the MBz Technical Companion (p 114), "Pump effectiveness can be felt with a hand on the outlet hose to the radiator. With the thermostat wide-open, pulsations should be clearly felt."

Okay. Well, which hose is the outlet hose? On every car I've ever had, the upper hose housing on the water pump contains the thermostat, and the flow goes from the engine through the thermostat to the radiator via the upper hose. Well, on the 450SEL, the thermostat is on the lower hose. A diagram found on 20-005 of the MBz CD shows hose 'A' (the NON-thermostat hose, i.e. the upper hose) being the outlet from the engine to the radiator. Okay. Weird, but okay. Maybe its a reverse flow radiator? Well, I look in the Haynes manual and read (step 10 of Water pump remove/install on page 84) the phrase, "the upper water inlet connection to the water pump." This would obviously go to my understanding that the thermostat-bearing hose flows to the radiator, and the other hose would be the inlet from the radiator to the water pump.

So, as usual, the more reading I do, the more confused I get. I'm trying to figure out which hose to feel. But I'm confused. So I feel BOTH hoses.

Results: the top hose gets hot, but I feel no pulse of any kind. The bottom hose does not get hot, but I only feel a faint hint of a pulse (more like a low vibration) about every 1 - 1 1/2 seconds. I can't say I "clearly" feel the pulses. I'd like to know how well this test works on a 4.5L w116 car w/ a healthy pump.

So I'm hoping someone will do me a huge solid on their healthy 4.5L w116. Could someone please start their car, let it warm up for several minutes until the thermostat opens, and then tell me: 1) if their lower hose is hot and 2) how easily/clearly they are able to feel the pulsations from the water pump? Boy, would I ever appreciate it!

This will tell me whether I need to just replace my water pump. It's a big job (for me, anyway) and I'd rather only do it if my relatively low-mileaged unit is actually bad. And someone can help me decide.

If there's another sure-fire test to verify a healthy water pump, I'd like to hear it.

Merry Christmas,
Mike

Mike D · #27 12-26-2006, 05:57 AM

Mike,

The lower hose is always the inlet hose no matter what make of automobile.

Fluid flows from the lower hose into the water pump, the pump "impells" the fluid into block passages, through the cylinder head, out through heater cores or auxilary devices and back to the top radiator hose where the fluid "drops" to the bottom to begin it's cycle again.

The heat collected from the engine is dissapated through the vanes located on each row of water passages in the radiator. The lower hose should feel much cooler than the upper if the radiator is doing it's job.

It's really doubtful you would ever feel a pulse at idle. Water pumps are impellor turbines and not a compression device "pump". The pressure in a coolant system is created by expansion of the fluid. The pressure is maintained only because coolant/water boils at a higher temp when pressurized.

Location of the thermostat has no bearing on it's effectiveness. It is just a heat operated restriction device. It is most commonly located on the upper hose simply for earlier opening and ease of replacement. I've seen electrically operated ones mounted inside the lower hose itself and not in the engine block.

9999 times out of 10000 when your pump is failing, coolant will leak out of the "weep" holes located in the pump housing. I have seen many instances of leaks between the aluminium housings and the cast pump housings.

Please, for the love of Gottlieb Daimler, do NOT try any "stop-leak" product! If the gasket is leaking, bite the bullet and replace it.

meltedpanda · #28 12-26-2006, 01:12 PM

i am with Mike - if the pump is bad you almost always always see fluid out of the weep hole. I once had a pump go bad on an old Lincoln, I though it was a hose, replaced em all, new clamps, kept leaking , finnaly gave up and the mechanic showed me the partially hidden weep hole when it was up on racks, pouring out fluid. My Old uncle who owned a garage use to move the fan blade back and forth to feel if the bearings were good. I doubt it was very relaible

michaeld · #29 12-26-2006, 03:10 PM

I used to think that the thermostat determined which hose was the outlet to the radiator (water remained in the engine until it reached the thermostat's opening temp, and then was released into the radiator to cool and recirculate). I wrote a post to my w116.org forum and got this reply from one of the members that I found insightful.

"The confusion about water flow probably comes about because M-B engines are a bit different.

In the usual arrangement the thermostat blocks the water flow completely until the water temperature reaches the opening temperature of the thermostat and then water can begin to flow through the radiator and pump.

In the MB setup water from the pump goes into the bottom of the block, rises up and comes out the top. With the thermostat closed the water is routed back into the pump and circulates through the engine again. With the thermostat open the water is routed to the radiator through the top hose, passes through the radiator, out the bottom hose and into the pump, and then into the bottom of the block as before. This way there is circulation in the engine when cold to minimize any hot spots.

Of course the thermostat varies it's opening depending on water temperature to control the flow and engine cooling.

My thermostat has an extra disk on the bottom to open and close the "cold" flow to the pump.

That's how I understand the cooling system to work, no idea if all M-B engines are like this."

I HEREBY PROMISE AND DULY SWEAR THAT I WILL NOT USE ANY STOP LEAK PRODUCT. I do not want the Mickey Mouse solution (Goofy, maybe, but not Mickey).

As for the bit about squeezing the hose and feeling water pump pulsations, I am not making this stuff up. It is in the Mercedes-Benz Technical Companion, page 114. It might be wrong, but there it is, in a trusted, up-to-now-considered reliable source of Mercedes-Benz repair information.

I would sure like to see a way to verify whether my pump IS WORKING!

While my pump is clearly leaking, the one place it does NOT seem to be leaking from are either of the weep holes (i.e. I'm the only one weeping in my garage!). The bearings are also tight and do not rock back and forth.

In any event, I would surely like to see positive test that can verify that a water pump is working, and not just a negative test that it is failing.

Based on what you guys are saying, I should assume that my pump is working (at least if I can fix the leaking housing joint). But I'd like that positive test for the sake of assurance.

Mike D · #30 12-26-2006, 04:35 PM

If you are getting heat from the heater your water pump is working. If you by-passed the heater feel the bypass line. If it's warm you have circulation.

If the engine glows red hot and peels paint off the garage wall after 10 minutes of idling then it's not working.

A common problem with older Benz's (Benzii?) is corrosion of the housing connecting the lower radiator hose to the block. My M110 was almost completely blocked. There was a pencil sized hole through the corrosion allowing a teeny bit of water flow.