Topic: Fluctuating Coolant Temperature

[Carbon Sky]

I typically leave my DIC on coolant temp, and for the most part it is very consistent.  Typically about 91* C +/- 2*, and only ever hitting a max of about 96*, and a low of 89*.

However, lately, I've been noticing the coolant fluctuating between 86*-101*.  The only thing I can think of is a partially stuck thermostat.

Has anyone experienced this?

I was planning a coolant flush this summer anyways.  I can always throw on a t-stat if needed.  Anyone have a part number on it?

Thanks for any help!

[elff]

You could have some air in it too. 
Might not be a bad idea to go through the purge process.

DDM has a nice video on how to do it.

[miller11386]

put in the check valves as well. they are cheap

http://www.skyroadster.com/forums/f5/coolant-237-yikes-40267/#post557485

[Carbon Sky]

Yah, forgot to mention, I checked the coolant level in the expansion tank, and it's been at the same level as it's been for the last 3-4 years since I got her.  And I did a pretty thorough general inspection recently when I swapped on the summer rubber, and there were no signs of leaking coolant.  But yes, definitely this does NOT eliminate an air-bubble in the system as a possible problem.

I've already bought the 2 check valves, and was waiting for the weather to get better before I flushed the coolant, and install the valves.  I guess since I'll be doing the coolant and check valves, I might as well at the very least inspect the t-stat while I'm in there, and replace if necessary since I'll have opened up the coolant system.  Plus I don't really trust Dex-cool past 3 years, and since I can't guarantee I'll get it all out, I'm going to stick with Dex-cool (Dex-cool doesn't play well with other coolants).

Guess I might have to get to this sooner rather than later.  Guess it's not horrible.  I was looking to run AMSOIL P.I. fuel system cleaner through the gas tank, clean out my intake with GM Cleenz, and swap out my spark plugs with a new NGK Iridium set in April anyways (picked up the plugs last weekend)

And elff, I guess I could always throw on your GMPP coilovers, and get a proper alignment at the same time   Waiting to purchase the INA-W910 is serving to be beneficial, the W910 and the W900BT have both been dropping in price (about $100-200) since I started compiling components for the big audio swap around the new year.

Thanks guys!

[diamoney]

The fluctuating temps may also be an indicator of the coolant temperature sensor about to go out.  On the 2.4L it is fairly inexpensive and easy enough to replace (especially if you already havr the t-stat exposed).  I haven't checked to see if it is in the same location on the 2.0.

[Carbon Sky]

The fluctuating temps may also be an indicator of the coolant temperature sensor about to go out.  On the 2.4L it is fairly inexpensive and easy enough to replace (especially if you already havr the t-stat exposed).  I haven't checked to see if it is in the same location on the 2.0.

True, definitely gonna keep that in mind.  If I inspect the t-stat and it passes and open/close test with boiling water poured over it, the sensor will be my next area of focus.  Thanks!  But if it is the sensor, I would think it would cause issues like error read outs on the DIC, as opposed to just slight fluctuation. 

[tazz]

Mine was both the temp sensor and the T-Stat.
The slight fluctations caused by the temp sensor the huge 40* swings later by the T-Stat.  Would drop from 190 to 150 and gp to turn the heater on and cold air even though the car was driven for 20 min then it would open again and it would heat up.

[Carbon Sky]

Mine was both the temp sensor and the T-Stat.
The slight fluctations caused by the temp sensor the huge 40* swings later by the T-Stat.  Would drop from 190 to 150 and gp to turn the heater on and cold air even though the car was driven for 20 min then it would open again and it would heat up.

Interesting.  No code issues from the temp sensor failing?

Hmm, well guess I'll find out soon.  Think I might take a look at this next long weekend if the weather is nice.

Edit: Took a peak in the area, looks like the ECT is screwed into the same cast aluminum housing that the t-stat sits in.  Holy 5hit is that sensor close to the downpipe/turbo.  There's like 2" between them!!!  At least there's this metallic heat reflective shield around the sensor, but still, that area must get crazy hot.  Anyways, haven't driven it since Friday, but we'll see if that problem crops up again, doesn't happen very often . . . at least not yet.

[diamoney]

True, definitely gonna keep that in mind.  If I inspect the t-stat and it passes and open/close test with boiling water poured over it, the sensor will be my next area of focus.  Thanks!  But if it is the sensor, I would think it would cause issues like error read outs on the DIC, as opposed to just slight fluctuation. 

You are right and so is Tazz.  The T stat cannot physically jump or drop 10, 20, or 40 degrees in an instant, but that is how it appears on the DIC- without codes!- until the dreaded  - - -  triple dash readout.  And that may only be temporary.   The sensor being electronic causes whacked numbers.  Good news is, it really is a simple fix (when you figure out how to remove the wipers)!

[Carbon Sky]

You are right and so is Tazz.  The T stat cannot physically jump or drop 10, 20, or 40 degrees in an instant, but that is how it appears on the DIC- without codes!- until the dreaded  - - -  triple dash readout.  And that may only be temporary.   The sensor being electronic causes whacked numbers.  Good news is, it really is a simple fix (when you figure out how to remove the wipers)!

It doesn't drop in an instant (takes a minute or two do say go from 88* C to 96* C).  And if you think about it, if the t-stat is slightly stuck, and goes to unstuck or vice versa, it could allow the temperature change to happen relatively quickly, but not instantaneously.  For example, if it's slightly stuck from start up, there's cold coolant in the radiator, and after the car reaches normal operating temperature, if the t-stat is stuck mostly closed, if it does open up, that cool coolant in the radiator will rush in, and lower the ECT reading significantly.

Drive in to work this morning was uneventful.  Temperatures stayed normal, around 91* C.

Anyone with the part number for the coolant temp sensor, and the thermostat?

[Carbon Sky]

Well, I've been keeping a close eye on the coolant temps, and they've been really stable ever since that episode that lasted about a week or so.

It's again holding steady between 89-96* at the extremes, and typically 91-93* again like it did for most of its life.

Like I said, I'm changing out the coolant as part of regular maintenance, and since I'll have the coolant out, I'll inspect the t-stat.  But right now it's looking good.  I wonder if I'll find anything upon visual inspection of the t-stat.

[miller11386]

if you pull the t-stat, just replace it. If you think its still OK, keep the old one around, but I know with the work that is required to pull the t-stat, i would definitely replace it. Its too cheap not to.

the "easy" way to get to it requires to pull wipers and cowl off the front of the windshield. its a mega pain.

[Carbon Sky]

if you pull the t-stat, just replace it. If you think its still OK, keep the old one around, but I know with the work that is required to pull the t-stat, i would definitely replace it. Its too cheap not to.

the "easy" way to get to it requires to pull wipers and cowl off the front of the windshield. its a mega pain.

I've already ordered it.  Tried to give my business to CED since they gave me great service in the past, but I just couldn't beat the price on Amazon for the AC Delco, identical part.  Should get to me Monday according to tracking.  My Prosport gauges should arrive tomorrow.

I'm going to do the coolant regardless, so might as well open up the t-stat and take a look, but DAMNIT!!!  She's been super stable again since that 1 week or so episode.  The only thing I can think of is I had put in the first new tune the car has had in almost a year (10-11 months).  Maybe the tune cause the fluctuating?  But anyways she's back to being 90-93* C like usual, instead of 88-99* like it was doing when it was fluctuating.

EDIT:
Actually, both tracking systems were being conservative.  Both items arrived today.  Powered up the prosport gauges to make sure they are functional, and everything checks out. I"ll have them blue daytime and red night time.  And the Escort Redline above them, all above the center vent pod.

[GXP Jim]

Sounds like my old thermostat that your describing, yes they do stick. Mine was covered by warranty and my dealer had to remove the cowl to get at it.
Put the check valves in too. I also had a bottle of water wetter added to cooling system.
Problem solved.

[Gentleman Jack]

Sounds like my old thermostat that your describing, yes they do stick. Mine was covered by warranty and my dealer had to remove the cowl to get at it.
Put the check valves in too. I also had a bottle of water wetter added to cooling system.
Problem solved.

I've used water wetter I other cars.  How many folks are using it here?

[Carbon Sky]

Sounds like my old thermostat that your describing, yes they do stick. Mine was covered by warranty and my dealer had to remove the cowl to get at it.
Put the check valves in too. I also had a bottle of water wetter added to cooling system.
Problem solved.

I bought the coolant valves a while back knowing I'd want to change them out when I do the first coolant flush.  I know the Dexcool has a 5 year recommended service life, but at 3.5, that's about as long as I want to go.  It's cheap insurance in my opinion.  Anyways, since I'm doing the coolant and the valves, and all that stereo work, taking off the cowl will probably help with other work anyways.

As for water wetter, it does what it says.  My only concern is as coolant ages, its pH will change, causing silicone compounds in the water wetter to come out of solution, and turn into blobs.  So as long as you monitor your coolant regularly, and change it at a reduced service interval, water wetter does what it says.  The important part is not to rely on it.  If you "need" it to stay at an appropriate temp range, your radiator is undersized.

[TomatoSoup]

After seeing the Water Wetter comments yesterday, I did some reading on the WWW and am not convinced it's such a great idea.  Seems to be designed as an additive to distilled water for racing where it can drop temps by 10's of degrees.  But the claims for it when added to street car coolants drop to a degree or two.  With the well-known issues of mixing stuff with Dexcool, I don't think it's worth the risk for me. 

Maybe as CS says, if you monitor and change coolant frequently, it may be OK.  But them why bother with all that for a degree or two delta?

[Carbon Sky]

After seeing the Water Wetter comments yesterday, I did some reading on the WWW and am not convinced it's such a great idea.  Seems to be designed as an additive to distilled water for racing where it can drop temps by 10's of degrees.  But the claims for it when added to street car coolants drop to a degree or two.  With the well-known issues of mixing stuff with Dexcool, I don't think it's worth the risk for me. 

Maybe as CS says, if you monitor and change coolant frequently, it may be OK.  But them why bother with all that for a degree or two delta?

Exactly, Water Wetter is primarily designed to be run with distilled water alone, and without anti-freeze.  They say it can be run with a 50/50 mix of antifreeze, but that tends to put the pH, particularly if it has aged, to be too low/acidic, and that causes the silicone to come out of solution, and when it does, it comes out in the form of blobs/goo.

And like others have mentioned Dexcool is primarily a long-service life coolant, that supposedly has additives to help buffer the pH over its usage.  However, Dexcool does NOT like to be mixed with other coolants.  While the "any car any year" coolants readily mix with other coolants without issue, Dexcool does not.

So if you REALLY want to run Water Wetter, just my suggestion but either run it with distilled only, or with less anti-freeze, and more water, and keep an eye on it.

Edit: did a quick search for alternatives.  Found this interesting pdf.  Apparently, Redline isn't the only maker of a product like this, and apparently Redline's product is not the best according to this independent test.

http://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&ved=0CF0QFjAD&url=http%3A%2F%2Fwww.amsoil.com%2Fnews%2F2011_April_testing_shows_amsoil_dominator.pdf&ei=4-CST8uCCISt6AHXj5jABA&usg=AFQjCNE1v52wwH58Ww5-FKj_DGYDE4LKGw&sig2=scN_Xx9ltZr7fltDngq2BA

[Gentleman Jack]

Ok, no water wetter for me.  I recall using it in my old truck that only ever got hose water in the radiator.  It had a small leak that I fixed with some gum or something, then added WW to it.  Ran great!  I miss old shitheaps that never die.  I think that's why I keep stopping to look at all the square headlight wranglers for sale around here.  So cheap, so easy to fix if they do break.

Anyone want to buy a Solstice?

[Carbon Sky]

Ok, no water wetter for me.  I recall using it in my old truck that only ever got hose water in the radiator.  It had a small leak that I fixed with some gum or something, then added WW to it.  Ran great!  I miss old shitheaps that never die.  I think that's why I keep stopping to look at all the square headlight wranglers for sale around here.  So cheap, so easy to fix if they do break.

Anyone want to buy a Solstice?

Water Wetter isn't a "bad" product.  It just has to be used properly.  It was originally intended to be used by weekend racers that would be willing to get a little more efficiency out of their coolant, in return for less service life, and non-compatibility with coolants.

Like I said, WW works best if run with distilled run.  It can be run with typical "all-applications" coolant.  But it's best to run it with more distilled, and less coolant than the typical 50/50.

And personally, I'd never ever ever run it in any car that even a little Dexcool in it.  Simply put, Dexcool doesn't play well with others.

And for many of us that will never face sub-freezing temperatures, they can get away with a more diluted mix, or even just distilled.

The typical "all-applications" coolants are green, and they tend to work better with WW than the orange Dexcool with the additives that supposedly make it longer lasting.

[Carbon Sky]

Well, got an alarming coolant temp reading on the way home from work yesterday, while on the highway cruising under light load.  So I'm 99.9999% sure it's the t-stat.

Bought it a while ago off amazon (great price, was even cheaper than the usual suspects like crateenginedepot, and gmpartsdirect), and picked up some Prestone dexcool, and distilled water today.  Will get to it Sunday.

Best price in town for the Prestone dexcool was walmart, about $3-5 bucks cheaper than Can Tire ($15 vs 19), and the distilled water was about half the price, just over a dollar at Walmart vs over two dollars at CT for 4L.

[DeepBlueGXP]

Burping the car would be my first maintenance response.  Changing the t-stat would be the second.  It's on the rear of the engine on the 2.0 and isn't easy to get to.

[Carbon Sky]

Burping the car would be my first maintenance response.  Changing the t-stat would be the second.  It's on the rear of the engine on the 2.0 and isn't easy to get to.

Since the day I got the car, the coolant level in the expansion tank has not moved.  And previous attempts at the "burping" method yielded no gurgling noises, so I'm pretty sure this is not air in the system.  Plus, I was going to change the coolant in a month or so anyways (I don't trust Dexcool to go 5 years, and the car is more than 3, less than 4 years old right now).

[Carbon Sky]

Well since I was doing the t-stat and requisite coolant today, I figured it'd be a good time to clean the intake, and change out the spark plugs.  My oil life indicator was around 25%.

Started with doing the intake cleaning.  I disconnected one of the hoses attached to the intake manifold that runs to the solenoid on the intake manifold.  I sprayed GM Cleenz into the hose as fast as I could spray it out of the can.  I would then reattach the hose, rev the engine a bit to make sure the engine sucked it all through, and repeated a few times.  I let it sit for a while to let it sit and dissolve a bit.

Then started up the car, and sprayed some GM Cleenz through out one more time, then let it idle for a bit, and rev it a little to make sure I cleared it all through.  I figured at this point, I've sprayed away any dirty stuff, and sucked through any remaining solvent.

I then started draining the oil since it was hot, and I wanted the block to cool more before dumping the coolant.  And the oil could drain while I completed the intake cleaning.  I Put on a Fumoto oil valve a while back so it's a breeze now.  I put a clear hose on it, and into an old oil jug from the last oil change.  I used some compressed air to clean off the plastic cover and sound dampening cover just to keep from getting dust/debris into the oil filler hole.  And I was doing the spark plugs anyways, so the entire top of the engine got cleaned off.

I then disconnected the rubber coupler between the cold side intercooler pipe, and the throttle body, and cleaned it out.  Surprisingly there was what looked like condensation had been in there at some point, and that there was a brownish powder residue left over from the evaporation of the condensation.  Not rust though as it was not magnetic at all.  No oil to be found in the tube.  I wiped out the inside of the throttle body area with a rag sprayed with some GM Cleenz, some light carbon deposits were seen.

Then reassembled the intake, but then started the spark plugs.  Disassembled the top . . . pretty straight forward.  Blew compressed air into the tops of the heads before removing the spark plugs.  Did one spark plug at a time.  Torqued to 20 ft lbs with a dab of 5w-30 on the threads.

I then disassembled the wiper blades.  Just a tip, I used the smallest 3 jaw puller set that I had.  Worked like a charm compared to the horror stories I've read about how hard the are to get off.  Mine had some grease between the threaded stud from the motor and the wiper arm.  But during reassembly, I put some extra anti-seize.

I then drained the radiator from the petcock.  Then disconnected all the electrical connections at the back of the engine block near the t-stat.  The hoses were far more difficult.  There are 2 hoses coming off the t-stat housing.  I was only able to remove the smaller one due to the orientation of the spring clamp on the larger tube.  Luckily I had a stubby 10mm wrench that I could get onto the back side t-stat housing bolt.  Trying to get a socket onto it pretty chewed up the back of my right hand.  The front bolt is no issue.

The old t-stat had no visible problems with the gaskets, but the centre brass? portion was severely discolored, and almost appeared to be corroded.  The coolant was definitely not in its prime either.  It was reddish/pink/brown and had deposits in it.  The fresh dexcool was bright neon orange.

Put the heater core valves into the 2 tubes.  The one above the engine block, and the one coming off the top of the radiator.

Reassembled everything.  Filled up the expansion tank with distilled water.  Ran it through the car with the expansion tank elevated, and with the hose in front of the block alternately clamped and not clamped.  I found this was very effective at getting air out of pockets.  For example before doing a procedure like this, the coolant temp would easily climb over 100* C.  How high?  I didn't care to find out.  But after a few repetitions, the temps dropped to about 88-90* C.

I filled and dumped with distilled water twice.  Then started mixing in some Prestone Dexcool.  I'd pour about a quart in first, then a quart of distilled water.  Run the car a bit, the level would drop, and repeat.

Haven't driven it much since.  But over about a 30 minute drive, the coolant temps were a stable 88-91* C.

[DeepBlueGXP]

Did you happen to take photos?