Chrysler 300M Enthusiasts Club
  • 3.5 overheating/running hot problems

  • Repair Questions and Answers.
Membership Banner

Repair Questions and Answers.

Moderator: Moderators

 #258850  by user97
 
Just wanted to post about my experience dealing with 3.5 overheating issue (the same should apply to 3.2 engines).

Problem: Engine runs hot, fans cannot bring the temperature down due to little flow ( coolant flow) across the radiator. Radiator along with water pump, timing belt, etc are brand new, the system has been properly bled. For more info see my original thread at:

http://www.300mclub.org/forums/viewtopi ... 10&t=25432

My first question in dealing with this problem was how the wax pallet of the thermostat is exposed to hot coolant. In more tradition upper outlet thermostat the answer is obvious -- coolant in the outlet is at engines operating temperature. However, in case of an inlet thermostat, the answer is not as obvious. In the case of outlet as well as inlet thermostat the engine must have a bypass that allows the block temperature to be equalized when the thermostat is closed. My initial assumption (based on illustrations in FSM) that the cabin heater acts as the bypass loop. However, in my opinion, this is not the only bypass loop that exists.

The reason manufacturers switched from more traditional outlet to inlet thermostats is to avoid "shocking" the block with cold coolant under low ambient temperature conditions when the thermostat first opens. Additionally, inlet thermostat reaches the equilibrium temperature faster. The reason the inlet thermostat is able to achieve these feats is because its wax pallet is exposed to BOTH the hot coolant stream of the bypass loop as well as the cooler coolant (when the thermostat opens) coming from the radiator. As a consequence it does not suffer from tendency to over-regulate to the same extent as an outlet thermostat that reacts ONLY to the hot coolant exiting the engine.

As you can see from the first image (all images credit http://www.dodgeintrepid.net) there are actually two bypass loops that are in play. The first one is the one is the documented in the FSM heater loop. As you see, the heater inlet into the block is thermally isolated from the wax pallet of the thermostat -- the flow from the heater return merges with the radiator return a few inches beyond the top of the thermostat's spring. The reason for this is that the heater return flow should have no effect on the operation of the thermostat. Imagine the following scenario: car at operating temperature, very low ambient temperature, heater on high, engine at max load. If the heater return flow was the "regulator" for the thermostat, it would never open as long as the heater return flow is under 180F. To circumvent this issue there is a second bypass loop which serves two purposes:

Act as "regulating" or feedback loop for the thermostat. This is how thermostat is exposed to engine's coolant temperature. This loop consists of a small recessed hole (approx 7/16" in diameter). The recess is slotted, which allows the bypass to "shoot" hot coolant along the length of thermostat. Note: on the first image the block is upside down. The bypass loop also mixes the hot engine coolant with cooled coolant returning from the radiator thus avoiding "shocking" the block when thermostat first opens under cold ambient temp conditions.

In this setup thermostat function very much depends on the bypass flow (internal to the block) being within the original spec. Any obstruction reducing this flow will result in thermostat closing way to soon (due to being overwhelmed with cooler coolant returning from the radiator). As you can see from the second image, the machined cavity in the LEFT (i.e. driver's side) head that allows a relatively small stream of coolant to flow from cooling passage into the block bypass gets obstructed by heavy scaling in neglected cooling systems. As a result the bypass flow becomes weaker and weaker. As the bypass flow weakens the thermostat is no longer exposed to the engine's hot coolant in sufficient extent and "thinks" that the engine is cooler than it is. Hence the bizzare behavior I described in my first post (link above). Assuming that the car is stationary, as the engine first warms up the coolant in the engine and radiator is at reasonably close temperature. Since fans are off, the radiator return flow is eventually hot enough to keep the thermostat open. Once fans kick (with minimum running time of 4 mins) they eventually bring the radiator return flow closer and closer to 180F. Even though the engine is much hotter than 180F (in my case closer to 200F), due to insufficient bypass flow, the thermostat is overwhelmed by the return flow and begins to close. As the flow across the radiator slows down, the fans bring radiator return flow lower and lower in temperature, resulting in thermostat closing further and further. As this flow is insufficient to cool the engine, engine temperature begins to rise, eventually going above fan activation threshold even as the fans are running. Hence you end up with the issue of always running fans. Equilibrium state eventually is reached by at much higher temperature (in my case at around 236F).

As an aside, blocking off this internal block bypass flow is the reason why these cars overheat when the left and right head gaskets are swapped during engine rebuild. L and R head gaskets are identical, except for the the cutout to accommodate the cooling passage to internal bypass loop, which terminates in at the recessed hole mentioned above (see second and third pictures). Thus swapping the gaskets results in cutting off the internal bypass flow and the thermostat NEVER opens.

Also, the problem gets aggravated when radiator is replaced -- because originally radiator is as scaled up as the rest of the engine and looses it's cooling ability roughly at a proportional rate to the loss in the bypass flow. When the radiator is replaced the bypass flow gets overwhelmed with the new and improved cooled water flow resulting in higher operating temps as described above.

As a second aside, the dashboard gauge is not linear (at least in F units). Typical 3/8th mark corresponds to roughly 195F (based on OBD II read out). I had car get as hot as 236F idling and the gauge just barely reached the 1/2 mark. My guess is that the 1/2 way mark corresponds to roughly 236-240F. Boiling point of coolant (50/50 ratio) at 16PSI is 267F (which should correspond to the red zone). Hence you have only 27-31 degrees F to go from 1/2 to red zone. By contrast 1/4 mark correspond too approximately 175F, hence it takes about 20F to travel from 1/4 to 3/8. Just something to keep in mind. My gut that this was probably done on purpose as to make the typical operating temp fluctuations in the range of 180-220F appear to be nearly imperceptible to the driver.

As a third aside, having removed and re-assembled thermostat 4 times (trying various t-stats as well as during flushing attempts), I found an alternative to staking (as Bill kindly pointed out) thermostat in place. What I did is to use large industrial staple (from a cardboard box) that was long enough to sit across inlet housing large hose nipple (where you would normally place the hose. Then I used two thin copper wire strands to attach the thermostat (properly aligned in the housing, spring facing towards the block), to the staple by twisting them on top of the staple until the thermostat was snugly held in place. This allowed me not to worry about the positioning of the thermostat while installing the housing and the gasket. Once everything was in place, I simply untwisted the strands and removed them along with the staple.

In an attempt to remedy the problem I tried shooting water into the bypass whole via a small clear hose -- water went through but the flow was clearly obstructed. After blasting with compressed air/shooting water I was not able to achieve a significant improvement.

I think it is a common problem with these cars as I read posts trying to deal with mysterious overheating issues, eventually assuming that the culprit was the head gasket (which might even be blown by that point due to running hot all the time). Some of the pictures are taken from this post (http://www.dodgeintrepid.net/showthread.php?t=171387) about engine rebuild due to what was *THOUGHT* to be a bad head gasket. Notice that during re-assembly they cleaned out the aforementioned bypass machined "pocket" in the left head and the problem was gone.

Personally, I did not want to rebuild the engine to address this problem when nothing else is wrong. Since the generation I 3.5 engine (which is very similar in design to gen II 3.2/3.5) had an outlet rather than inlet thermostat, I decided to try to modify the cooling system to use outlet thermostat.

First step was simple: remove thermostat from water inlet and re-install the water inlet (thermostat housing) without the thermostat. Then I bought Meziere inline thermostat housing (http://www.meziere.com/ps-571-0-WN0071.aspx) as well as an old Chevy V8 180F thermostat that fits into this housing. I drilled the thermostat with 3/32" drill (needed only one whole on top to make sure that system bleeds correctly, but for safety drilled one on top, one on the bottom). Original t-stat had a 7/64" hole with a jiggle valve. I did not bother with the valve. Marked the location of the top hole on the outside of the inline thermostat housing so I knew which way it should be facing. I also noted the location of the wax capsule within the housing as it should be facing the block. Then I spliced the top hose to install the housing with the thermostat as close to the block as possible. I personally picked the location where the hose comes out of the first bend and runs parallel to the timing belt cover. I bought a second hose to use as a model and measured everything ahead of time to make sure that the modified hose aligned the same way as the original one.

NOTE: to tighten the Meziere housing you will need 2X 2-1/2" wrenches. I bought sink wrenches from home depot and they worked great and cost only $7 or so a piece.

In conclusion: I've been running the car with a modified cooling system for 3 days now and the temps are normal (just below or just above 3/8th mark or 187F - 192F on the freeway or 187-214F idling with AC off or 187-198F with AC on) The difference in temp with AC on vs. off is due to different fan cycling patterns as per FSM.

CREDIT: None of the photos are mine -- found them on dodgeintrepid.net forum. The original pointer to the root cause of the problem is also not my invention -- in desperation I started dropping by various dealers and talking to mechanics. It was one of the mechanics who mentioned that they fix this mysterious overheat on 3.2/3.5 by removing the left head and clearing out a machined pocket in the front but did not elaborate how it helps. This got me rolling to try to find more info. The idea that the bypass is internal to the engine came from this article (http://www.underhoodservice.com/Article ... 5l_v6.aspx) about Gen I engine that mentions "water pump bypass internal to the block".

DISCLAIMER: All information above is provided as is and is not represented as suitable for any purpose. The information represents my ideas about what might be the cause of the problem and is not represented as correct, accurate or actionable. I do not know if my modification is safe for the engine, especially as it relates to "shocking" the block with cold water when t-stat first opens up or due to any other known or unknown reasons.

If it helps somewhat -- great -- I spent a lot of time trying to figure this out.
You do not have the required permissions to view the files attached to this post.
 #260558  by user97
 
As a follow up -- the car has been running perfectly after the conversion -- it is now my daily driver and 1.5K later (mixture of city and freeway driving) it behaves perfectly (~189-195F freeway temps depending on the ambient temp).
If I did the job again, I would drill only one whole in the thermostat and probably fabricate a jiggle pin valve.

Off the topic but related, there have been quite a few posts lately about cars overheating to the point of boiling (under pressure). Typical description provided is that of the car being fine until all of a sudden shooting up into the red zone. I think this is due to the non-linear behavior of the temp. gauge that I mentioned in my earlier post. For a while, when my car was running pretty hot I did not even realize this until I paid a close attention to the needle (1/2 on the "problem car" vs 3/8th on my other LHs). In fact, what prompted me to look further was the fact that once I saw the needle a tiny bit above 1/2 mark (after the car was shut off hot and restarted shortly thereafter). On restart the needle dropped to 1/2 mark very quickly, but since I have NEVER seen this before with any of my other LHs, I started looking. Had I not had other LH vehicles, I would have never thought about it twice. Hence, I think that the non-linear behavior of the temp gauge can mask mild overheat problems. OBD II scanner with live data capture capability came in very handy. Alternatively, infrared thermometer (~$20 at autozone with a coupon) pointed at metal water neck on top of the engine (coolant outlet) at close range is also accurate within 1F relative to OBD II readout.

Use of wrong coolants seems to be the culprit -- the "problem car" had a mix of dexcool, generic green, as well as Chrysler's long life added over the years based on service receipts. Since it is a 2000 it presumably originally came with green coolant. At some point in '03 (based on service receipts I got with the car), the water pump was replaced due to "noise" under warranty. Receipt notes 1 Gallons of Chrysler long life coolant being used in the repair (presumably mixed with tap water). Then I have service receipt from a quick oil change place noting that coolant was flushed and replaced with Prestone's Dexcool. RESULT: Jello type mess in lower parts of the engine, HEAVY scaling in the upper parts of the engine, badly clogged heater core.
Chemical flushes got rid of Jello and some of the scaling (had some really large chunks of scaling come out along with fair amount of sandy substance), replaced the rad (along with water pump, etc) but the internal engine block bypass passage was still fairly clogged -- hence the thermostat mod.
User avatar
 #260569  by Bill Putney
 
There's a lot of good information in that post, Alex! Thanks.

What does a jiggle pin do that a plain old hole wouldn't do?
 #260592  by user97
 
Jiggle pin is supposed to restrict coolant flow through the hole under normal operation, while allowing for draining, air purging/refilling. In the stock location (i.e. inlet thermostat at the bottom), the primary purpose of the hole is to allow draining of the coolant from the engine block as well as to facilitate refilling of the radiator (though the latter would most likely still happen through the upper hose, when the engine block is full). When the car is running, the jiggle pin is supposed to obstruct the flow of coolant through the hole in the thermostat. I am saying obstruct rather than block because it is far from a perfect fit.

As an aside, the Duralast/Motorad t-stat I bought from Autozone had the jiggle pin installed in reverse. You can see it in the catalog image as well (http://www.motoradusa.com/ProductImages/419.jpg) as compared to Stant t-stat that has it in the stock direction (http://www.stant.com/Resources/images/p ... /14358.jpg). The result was interesting -- system was impossible to drain (I let it drain for over 4 hours and I still drained less coolant that normally comes out in 10-15 mins) -- the flow was a mere drip once the coolant in the radiator drained. I had to blow air into the overflow to get any meaningful flow -- that's how I discovered that jiggle pin was in reverse. This also explained why the system took a seemed to take a tiny bit longer to come up to 180F temp with this t-stat than the other t-stats I tried.

In my arrangement, I needed the hole in the t-stat to allow the air from the radiator to escape when the system was being filled. A jiggle pin installed in such a way as to obstruct normal flow (i.e. from the block into the rad) would still allow the air to escape during fill/purge operations, but would obstruct the coolant flow through the hole during normal operation thus negating hole's effect on the thermal equilibrium. But truthfully, the effect of my tiny holes is negligible -- may be it takes a tiny bit longer to come up to operating, but that's pretty much all -- it's evidenced by the fact that there are thousands t-stats out there in our cars that have jiggle pin installed in reverse, which is similar in effect to my just having a hole there under normal operating conditions. T-stat's have such a dramatic variation in operation (even amongst t-stats of the same brand) that I really doubt that a tiny hole in t-stat would have a meaningful impact. I guess I would fabricate jiggle pin mostly for the sake of "doing things right" rather than its actual affect.
User avatar
 #295372  by SpeedBallDEVIL
 
Thanks for posting. This should find its way to the knowledge base if it hasn't done so already...
User avatar
 #295433  by paintballdude05
 
This is a very good read indeed. I bet this is part of the issue with my '02 Intrepid. I've been through a handful of rads and my car is always operating at slightly higher than 3/8 on the gauge(my special stays at 3/8 no mater the temperature or any other factory for that mater) and fluctuates slightly.
 #319741  by 300mguy
 
user97 wrote:Then I bought Meziere inline thermostat housing (http://www.meziere.com/ps-571-0-WN0071.aspx) as well as an old Chevy V8 180F thermostat that fits into this housing.
The Thermostat housing from Meziere is still available, including on Amazon(http://www.amazon.com/gp/product/B003ND ... PDKIKX0DER). However, currently Meziere does not sell 180F thermostats that would fit in. They only have 170F and 185F. Could you recommend a 180F thermostat they would fit in that housing? For example, would this one http://www.amazon.com/Stant-45358-Super ... auto_img_y fit?
User avatar
 #319751  by Mottman
 
The EVIL Sickness is also having some wierd engine temp issues. Basically, the cooler the weather= the hotter the engine. I truly believe in you'r theory, and you'r solution. I am going to try it myself as well before winter. I already started buying the items needed to do this. Just wanted to know, do you have a part # for the thermostate you used that fits the MEZIERE inline housing ?
User avatar
 #319981  by paintballdude05
 
I started a thread some time ago and document my search for the right thermostat for the Meziere unit. You don't need (I'd recommend against it too) to buy the meziere thermostat, there are many other ones that will fit.
 #319988  by user97
 
X2 - no need to buy their t-stat -- any old chevy "standard" t-stats fit (and you can get 180 one at nominal cost).
 #319990  by 300mguy
 
user97 wrote:X2 - no need to buy their t-stat -- any old chevy "standard" t-stats fit (and you can get 180 one at nominal cost).
I just bought this one: http://www.amazon.com/gp/product/B000C2 ... UTF8&psc=1 and it seems to fit. Will report the outcome after the conversion is completed. Also, I decided not take the head off and not to create an external bypass for now.
 #319991  by 300mguy
 
user97 wrote:X2 - no need to buy their t-stat -- any old chevy "standard" t-stats fit (and you can get 180 one at nominal cost).
Do I have to buy the sink tool you used or can I tighten the Meziere thermostat housing by hand? If hand tightening would not be enough, can you post a link to the sink tool you bought?
 #319994  by user97
 
I am pretty sure that it was a 2 1/2 inch wrench, but not 100% positive.

It looked awefully similar to this:

http://www.homedepot.com/p/BrassCraft-3 ... ly3KdvD_ug

If not identical...

Just measure the housing -- I am pretty sure that it takes 2-1/2" wrench.

I do not think you can tighten enough by hand -- I am simply not strong enough -- so I used two of the wrenches.
 #320359  by 300mguy
 
I completed my thermostat relocation. The parts used were http://www.amazon.com/gp/product/B003ND ... UTF8&psc=1 and http://www.amazon.com/gp/product/B000C2 ... UTF8&psc=1. The new thermostat is, of course, in the upper radiator hose. The old one has been removed.
thermostat.jpg
The good news is that the overheating and the fluctuations are gone. Before the conversion the temperature fluctuated between the green and the red line. When it would reach the red line the fans would go in "high" and the temperature would drop to the green line. After the conversion the temperature stays very close to the blue line. However, this is lower that what it was before the bypass got plugged (I am guessing it got plugged, the head has not been removed).
CoolantGauge.jpg
Here is my question. Where is the temperature sensor on this engine? Is it is close to the place the stock thermostat was? If this is the case, could it be now measuring the temperature of the radiator instead of the engine? In other words, is it possible that the displayed temperature is lower than the actual? If the fans are triggered by this inaccurate temperature, could it be the case that they kick in at higher actual temperatures, which would mean when the car is stuck in traffic it would run hotter until the fans kick in?
You do not have the required permissions to view the files attached to this post.
User avatar
 #320373  by Bill Putney
 
300mguy wrote:...Here is my question. Where is the temperature sensor on this engine?...
If the metal outlet pipe in your photo were transparent, you could see it. In fact I think you're seeing its connector in the shadows just under the MTV actuator. With the exception of hot spots around the cylinders, it's reading the point of hottest coolant temperature (just before the coolant goes to the radiator.
 #320382  by user97
 
T-stat relocation does cause it to run a bit cooler -- in stock setup 180 degree t-stat is controlled by a mixture of hot engine coolant from the bypass as well as by the cooler coolant returning from the rad. Now you have a t-stat that is controlled by hot coolant exclusively. In my experience the difference is minor -- the car runs at about 190F in winter with cabin heater off. The other question is the size of the hole you made in the t-stat. It really should be pretty small.

In the end, personally do not think that the slightly cooler operating temp makes a noticeable difference (did not notice any drop off in mileage relative to other LHs that I have with stock setup and check engine light stayed off).

How big was the whole you drilled in the t-stat? In cooler weather, the car runs at just a tiny bit below 3/8th and in hot weather it runs a tiny bit above the 3/8th mark. You might want to check gauge calibration by running cluster self test -- in some cases the gauge (mechanical portion) is a bit off and does not stop on the correct marks. One of the stops during cluster self test is 1/2 point -- you will see where it actually stops... As Bill said, the sensor measures the hottest point in the engine (aside from cylinder jackets), so you do not run the risk of silent overheat.
 #320383  by user97
 
Here is the pic of the temp gauge from my car with relocated t-stat in the summer (typical):

Image

and here is typical cool/cold weather temp (fall/winter/early spring):

Image

The T-stat has been relocated for the past 25K miles by now.
 #320431  by 300mguy
 
user97 wrote:T-stat relocation does cause it to run a bit cooler -- in stock setup 180 degree t-stat is controlled by a mixture of hot engine coolant from the bypass as well as by the cooler coolant returning from the rad. Now you have a t-stat that is controlled by hot coolant exclusively. In my experience the difference is minor -- the car runs at about 190F in winter with cabin heater off. The other question is the size of the hole you made in the t-stat. It really should be pretty small.

In the end, personally do not think that the slightly cooler operating temp makes a noticeable difference (did not notice any drop off in mileage relative to other LHs that I have with stock setup and check engine light stayed off).

How big was the whole you drilled in the t-stat? In cooler weather, the car runs at just a tiny bit below 3/8th and in hot weather it runs a tiny bit above the 3/8th mark. You might want to check gauge calibration by running cluster self test -- in some cases the gauge (mechanical portion) is a bit off and does not stop on the correct marks. One of the stops during cluster self test is 1/2 point -- you will see where it actually stops... As Bill said, the sensor measures the hottest point in the engine (aside from cylinder jackets), so you do not run the risk of silent overheat.
I didn't actually drill the thermostat. Instead, I used a heavy duty punch tool. So I used the 2mm bit I let it punch only half of the circle it would punch had it gone all the way through. So the hole is pretty small. It is probably smaller than the smallest drill bit that could be bought in the home improvement stores (may be 1/16").

After driving it for a few more days I am not sure I have completely solved the problem, though. The most prominent affect accomplished was to lower the normal operating temperature. By normal I mean while driving with 60mph or more. Then there is not fluctuation and the temperature stays close to the blue line. Before the relocation the normal temp (60mph) was at the green line.
However, when the car idles for a few minutes the temperature start climbing up again. After idling 15 minutes the it still goes all the way to the red line. It takes longer then before the relocation because the starting temp is lower. So my problem is postponed by a few minutes but not resolved.

What would your temperature be if you would let it idle for 15 minutes when the air temp is about 50F? Would it go all the way up to the red line? In general, would such wide fluctuations be normal, either before or after the relocation?
 #320432  by 300mguy
 
Bill Putney wrote:
300mguy wrote:...Here is my question. Where is the temperature sensor on this engine?...
If the metal outlet pipe in your photo were transparent, you could see it. In fact I think you're seeing its connector in the shadows just under the MTV actuator. With the exception of hot spots around the cylinders, it's reading the point of hottest coolant temperature (just before the coolant goes to the radiator.
Thanks. It is good to know that the thermostat relocation did not likely change the relevance of temperature gauge reading, i.e. there isn't much risk of silent overheating.