Overheating - Please help; before I junk the car!!!

[Porschamaniac]

Today, I experienced an emotion, which I can honestly say that I have never associated with my 944 before; I honestly considered leaving parked, with the keys in the ignition, in the hope that someone would steal / wreck / burn it.

I guess it's just been a long slog and just when I think I'm seeing light at the end of the tunnel, it turns out to be some b'std with a torch, bringing me yet more work to do!!!

Having spent months doing a thorough belt job (both belts, all tensioners, all seals, water pump, thermostat) I finally got the car running again on Christmas Eve. After a short road test, it was clear that it was running fine - just overheating.

Initially, I attributed this to an air lock - so a week of adding coolant and bleeding followed - to no avail!!

Having noticed all the signs of a failed thermostat (hot engine, cold radiator) I resolved to replace the aftermarket part that I fitted with the new pump, using a genuine Porsche part.

So, last week, I got to it. After a struggle getting the snap ring out (why didn't I fit it with the tabs at 12 o-clock!!) the new Porsche part slotted in beautifully.

With the system filled and bled, all seemed good. At idle, the fan cut in with the tep guage at the 3/4 mark and cooled it quickly to the 1/4 mark.

All was good until the test drive - fine for half an hour or so, but then gradually crept up to the red area.

On returning home, I noticed that the fan was not running, so, today, I slotted in a brand new OE Thermoswitch. Again, on the driveway, all was good - heated to 3/4, fan kicks in - cools to about 1/4.

And, again, all was good until the test drive. Admittedly, it took a little longer, but the temp hovered around 3/4 throughout the drive and then hit red as I was reversing back onto my drive. Again, when I parked, the fan was not running (at least, not at the half-speed setting - with the engine off, but ignition on). It should have been - because the temp was into the red.

It strikes me that I have one of the following issues:

(1) The aftermarket water pump is not up to the job - the car never did this before the water pump job. For reference, I have used one of these: http://cgi.ebay.com/ebaymotors/NEW-PORSCHE-944-944S-S-924-924S-S-WATER-PUMP-W-GASKET_W0QQcmdZViewItemQQ_trksidZp3286Q2em63Q2el11 77QQhashZitem140278766075QQitemZ140278766075QQptZM otorsQ5fCarQ5fTruckQ5fPartsQ5fAccessories

(2) A possible issue with the fan relay

(3) Maybe the cooling system suffered some sort of issue, whilst it was empty for about a month - during the long-term belt job

(4) Of course, it could just be the temperature sender

(5) Possibly a blocked rad - although, having flushed this with a hose, it seems to run freely enough.

I must admit that I'm at the end of my tether with this (and, I have a really slow fuse!).

I'd really appreciate any words of wisdom, or shared experience, which anyone else has had with this sort of issue.

Oh, and now the car has started trying to kill me on wet roundabouts - during the test drive today, entered a wet roundabout at about 20 (which didn't feel fast) and found myself looking back the way that I had come from - so now I have a nice front air dam repair to do also.

Where's the petrol and the matches!!!!!!

[JonA]

A few questions:
Do you have heat inside the car? One indication whether all air is out.
Are the cooling fans running in the right direction? These are DC motors and can run backwards with the polarity reversed.
Is the radiator getting hot? Are you getting flow thru it?


This may be a simple problem, don't think big problem until you eliminate all the small stuff.

Jon

[Porschamaniac]

Hi Jon,

Yes, plenty of heat inside the car - so I guess all is good with the marix (and possibly indicates no air lock).

That's a good call about the fan running in the right direction - is it possible to fit the connector the wrong way around?

Yes, radiator is getting hot - bottom hose as well. I even took the precaution of testing the thermostat before I fitted it.

Cheers,

Ian.

[Techno Duck]

Im not sure if its possible to reverse the leads on the fan. The pre '85 cars simply have a fan switch with spade connectors on it, the newer post '85 fan switch has a molex connector instead. Im not sure what effect would reversing the leads on that be, but i dont think that would make a difference at all. All the fan switch does is complete the circuit, so it shouldnt make any difference.

The fact you say the temperature slowly rises up tells me air is not your issue. Also since you have heat, that is a good indication also. If the cooling system was airbound you would typically see normal temps, then the a sudden almost immediate spike on the gauge.

I would say double check the temperature gauge is reading correctly, from your description i would say it is, but its worth a shot at checking and its easy to do. Honestly i do not believe this is a gauge issue, grounding issues usually cause erratic gauge readings and you seem to have fairly consistent readings. Or atleast you do not mention erratic readings.

I think its possible the waterpump pulley is slipping on the shaft. This is possible with rebuilt pumps and from the picture it looks to be this is rebuilt pump. I say that because this pump casing does not have the turbo water fitting on the front end. All of the new pumps are superseded to the newer style turbo pump, n/a's simply use a blockoff plate in this location. You can check this by removing the front timing cover and making with a paint pen or white out (whatever you have) the orientation of the pulley on the shaft. See the picture below, i put a green line where the pulley and the shaft meet. If you start the car up and get your overheating issue stop the car and check the marks are still lined up. If they are not, the pulley is slipping on the shaft, this can be corrected by putting a tack weld between the pulley and shaft. The other possible failure is the impeller falling off the shaft. But this does not sound like your issue because then the car would overheat pretty quickly.

[Porschamaniac]

Thanks Techno Duck - that makes great sense.

I guess that the thermo switch can be connected any way - but I do wonder if I've managed to get the connector on the fan motor the wrong way around. Presumably, this would try to push air through the rad, rather than pulling it - fine on the driveway, with no airflow over the car - but when it's moving forwards, the fan would effectively be pushing air one way, whilst it is naturally trying to flow the other. Result - dead air all around that part of the radiator.

Having said that, I think the spinning pump pulley is more likely, so I know what I'll be doing next weekend (come rain or shine!).

Best regards,

Ian.

[Techno Duck]

As far as i remember all of the 944 series regardless of year used a molex connector for the fans, so its basically keyed to go in only one direction.

[Porschamaniac]

Yes, that makes sense.

More likely cheap water pump methinks!

[msc247]

I just recently experienced the same issue with my 84' I flushed and bled and changed just about everything in the cooling system until I finally thought I had it figured out, and it still would overheat. I then tried to drain the block by removing the drain bolt and guess what? nothing came out. Now I was at what we or just I would call a certain "block in a hot place". I decided to pull the head to try and investigate the cooling problem. What I found was about a half pound of some kind of sand like substance blocking the head and the cylinder walls, I assume someone down the line tried to add some type of stop leak or head gasket repair. However after flushing the block and the head, putting it back together all is finally good. Now I am having idling issues that i have to track down. Anyways GOOD LUCK!

[944Spec_bound]

Did you ever verify it actually is hot, or if it's the infamous bad ground scenario causing a false temp reading? This would explain the erratic fan operation, the thermofan switch is seeing the real temp, but the guage is showing hot. I'd buy a cheap infared thermometer and see what the temp actually is when it's in the Red.

[944 boy]

+1 for verifying actual temp. The fans do not need to run when your driving around, I made it 2000 miles without fans working until I hit traffic and found out that the fans were not operational.

I recently let the car idle to get up to temp and it took driving and idling for 20+ minutes before I even hit 1/2 mark. I never did even get the fans to kick on, still don't know if they are working.

Never thought of the slipping pump but it makes sense, I would check that out too. Good luck and don't give up yet!

[HondaDustR]

I do remember a thread where someone's fans were actually wired wrong somehow... It would be passable at highway speeds, fine in stop and go traffic, but would seriously overheat at something like around 35 mph. Check for proper spinning direction.

Sometimes things get weird if there's some air in the system, but not enough to have outright vapor lock. The thermofan switch sometimes doesn't come on if the fluid level in the radiator is low enough, since it is located near the top. The temp sender is also located near the top and toward the front of the water jacket and might also suffer inconsistent readings if there's enough air. Successfully filling and bleeding these things is a misterious art. Once you figure out a trick that works, it's a piece of cake, but until you do, it's a serious PITA!

The clogged block idea sounds very possible. Some good shops offer something like a power flush that uses a machine to rapidly pulse the flow back and forth, or something to that effect, that might dislodge and clean out the block if that's the problem.

Seeing as it worked fine before, it's probably some pesky air, the pulley is slipping on the shaft on the WP, or the new timing belt has broken in and stretched, resulting in it now slipping on the WP pulley. New WP's usually tend to be stiff from the new bearings and seals, take a little while to free up, and would consequently need a little more driving force to spin. A loose cam belt would contribute to this problem. Don't forget a new cam belt has to be tensioned harder than normal for the first 1500 mi. to compensate for initial stretch. Not too insanely hard or it might kill the WP bearing...or the belt.

[Slam]

+1 to belt slipping on WP pulley.

[Porschamaniac]

Thanks for all the replies guys. I'm definitely going to check the pump / pulley.

Just a thought, but if I do need to tack weld it, is the rule of thumb to disconnect both the alternator and DME to prevent damage (from the welding current)?

Thanks again.

Best regards,

Ian.

[Makis]

Originally Posted by Porschamaniac

Having noticed all the signs of a failed thermostat (hot engine, cold radiator) I resolved to replace the aftermarket part that I fitted with the new pump, using a genuine Porsche part.

One other possibility would be the headgasket is gone. Exhaust from the engine enters the cooling system causing air locks and stoping the water flowing. This will then cause the overheating. If the radiator is cold the cooling fan will stay off. Should be easy to check the headgasket.

[Pops944]

Disconnect the DME, alternator, and the ground at the battery. This should protect your electornics. Sometimes it seems like overkill but better safe than sorry.

[HondaDustR]

Originally Posted by Makis

One other possibility would be the headgasket is gone. Exhaust from the engine enters the cooling system causing air locks and stoping the water flowing. This will then cause the overheating. If the radiator is cold the cooling fan will stay off. Should be easy to check the headgasket.

Ooo...good point...

Wait, I know what happened. The head gasket is gone causing overheating which boils coolant out of the overflow tube and onto the road where the water on the road helps spread it, making a very slippery patch, and the rear tires hit it and lose traction. Just a wild scenario... I hope not. It would be nice if it was just belt tension.

[SuperXRAY]

Maybe the picture just doesn't show correctly, but my water pump gasket looks nothing like the one in the picture from the auction...is it just me? Doesn't look like the gasket extends to the upper exit cavity of the block.

Check your gauge readings to verify your gauge is accurate. The gauge reads, according to outside sources (and verified on my girlfriend's '83 944) like:

First block on gauge is 40-68c, from left to right.
First tick is 80c.
Second tick is 89c.
Third tick is 97c.
Red block is 105-120c.

You have to verify the readings by taking temp readings from the sensor area (upper block under the intake boot and crankcase venting lines). I do this by warming the car up, then shutting it off and using a RayTek IR Temp Gun. It takes longer because the car has to cool, but is more accurate. If there is air in the system, you will not get accurate readings.

If that all turns out well, check your thermostat. I noticed the Wahler thermostats take considerable pressure to open the bypass valve. This affects heat inside the car. If you don't get heat (depends on thermostat, but slightly warm) by the time the temp reaches 68c, then there is either air in the system (again), or, in my opinion, you could have a failed thermostat. You can remove the thermostat and perform a static test by heating it in water. With an aftermarket, such as what I am using, there is hot air in the car before the 68c mark. When the Wahler thermo was installed, there was only slightly warm air until the 80c mark. Makes a big difference in the winter!

If all checks out well to this point, take the front timing cover off and check the belt tensions, as already pointed out about the belt slipping. Slipping of the timing belt could cause catastrophic failure, as it will heat the belt by friction until the belt fails - bent valves.

If all that is good, then I'd guess the head/head gasket is leaking. Cars do not always burn coolant when the gasket leaks, though it is common. They also don't always mix coolant and oil. If your temp gauge fluctuates from high to low, then your best bet is to take the head off the car. If there was every ANY kind of 'sealer' used to temporarily fix a cooling problem, it's very possible it is rearing its ugly head.

Hope this helps, even though most of it was said before. I referred to outside sources that I found browsing the articles here on Pelican and other places. Not sure if I should mention their names?

The suggestion or scenario that said you slipped in the roundabout because coolant dumped out is not uncommon in reality. It doesn't take much antifreeze dilluted with water (such as would be on the roadway) to cause ice-like traction loss.

One other thing I want to note is the understanding of thermostat ratings to what the sensor sees. The thermostat opens a nominal amount when it reaches its rated number, such as 82c (180f). However, since water flows towards the top of the radiator, it travels through the engine block and head, heating it up. Using a 82c thermostat could lead to 100c water temps at the sensor.

[Porschamaniac]

Thanks for the reply SuperXRAY.

You are right; the gasket supplied with the pump was (for some reason) only a partial gasket. I actually fitted the pump with the correct full gasket.

At the weekend, the first thing that I am going to check is belt tension. It was actually a very cold day when I fitted the belt (1 or 2 degrees) and I wonder if this has resulted in a 'false' reading, also.

After that, I think it's going to be a Dye Test on the colling system to see if there arde any combustion gasses present - followed by marking the water pump and checking if the pulley is slipping on the shaft.

I guess that it is possible there could still be air in there - but I have really done all that is humanly possible to erradicate it.

As you say, it wouldn't take a lot of coolant on a wet road to make you lose traction - and, indeed, the slide was almost identical to the feeling that you get on ice.

Thanks again - I'll let you all know the outcome.

Cheers,

Ian.

[SuperXRAY]

You are right; the gasket supplied with the pump was (for some reason) only a partial gasket. I actually fitted the pump with the correct full gasket.

GREAT...that had me really concerned. though I couldn't imagine if someone didn't use the correct gasket that they would make it more than a mile or two before the motor would seize.

At the weekend, the first thing that I am going to check is belt tension. It was actually a very cold day when I fitted the belt (1 or 2 degrees) and I wonder if this has resulted in a 'false' reading, also.

I hear you. was 12f here when I did all the front work. It's too bad garage has electric, concrete floor with epoxy finsh, insulation, water, and air lines...BUT NO HEAT...yet!

The timing belt I installed is experience-tightened. I hate this, but I had no other option. So far so good, going to check it at 250 miles. No whining though. I even had to modify the water pump because it uses, what I know now, the newer style belt guard. Stupid me, but oh well.

Hope all goes well,
Mitch

[HondaDustR]

The temp of the motor does noticably affect the cam belt tension. The factory manual specifies tensioning to be done at something like 68 degrees, but setting it correctly at 2 degrees would make it tighten up too much at full temp. How much I don't know.