Johnsrestos

What are your gauges reading on a really hot day over 100 with the A/C on full while driving around town or sitting still in the sun on a 32 valve car? My 89 S4 gets up just above 3/4 on the temperature gauge with 101 degree weather in the sun with A/C on high. A/C air is coming out ice cold too. It never got to the edge of the red zone though, even while sitting in the sun idling. No difference between sitting at a light or cruising at 40+. Harder acceleration seems to push it a little higher, but not into red. I measured with a laser and it was around 200-210 at the manifold and sensor. Upper hose was as hot as 255 in going into the radiator, but 195 or less coming back out. Is this normal? All new hoses, new stock thermostat, water pump, head gaskets, valve job, and I pressure tested the system for leaks. Fans and fan thermostats are working fine. I power washed the radiator and condenser plus made sure all fins were straight when engine was out. It has the aftermarket all aluminum radiator, not the stock one with plastic sides. The A/C fan was replaced at some point with an aftermarket fan that has more blades. Flaps are no longer hooked up and blocked open. Belly pan is missing and I did read that those help cool the engine. In the 90's I am back down to 3/4 or a little less. I am used to a 944 that never goes over half so it is a bit worrisome. Lower thermostat will just make the car take a little longer to warm up, especially in the winter. It lowers the floor, but not the ceiling so that is not a solution for this situation.

FredR

With respect to the gauge on the dash the indicated temperature should be about a needles width to the left of the last white line. For those of us with Sharktuner kit that will equate to a temperature of 98C being transmitted to the engine management system from the two channels on the temp 2 sender located on the water bridge. The cooling system will typically take out about 8 degrees C with a return temperature of circa 90C.

Whatever you see on a 944 is irrelevant but I can pretty much guarantee the actual temperature the 944 operates at will be pretty much the same given that in essence it is basically half a 928 motor.

Where your post goes a bit sideways is when you introduce the temperature of 255F without context. I have to assume that what you are trying to say is that this is something that has just started to happen whereas previously it was not the case as the system behaved as described in the first part of your post -is that a correct interpretation?

However if the dash panel gauge is now well into the red as in full scale deflection and you have an alarm of some kind showing [not sure what happens on the non digi dash models] the explanation for this would be that the water pump has chucked the plastic impeller of the drive shaft. If that happens coolant will not circulate and it is credible that what you measure in the return line will cool down as indicated. if this is what you are experiencing then do not run the engine like this until the issue is resolved. I doubt this is the case as the temeprature would not drop in lower ambients.

Assumnig you are using a temperature gun do not meaasure the temperature on the hoses and read from metal surfaces. Also calibrate your temp gun by measuring the temperature of boiling water at sea level that should read 100C.

Johnsrestos

Thank you for your reply. 100C was right about where my laser was measuring, but needle was above 3/4 so perhaps it is reading slightly off. I will have to go back and review my gauge and sender resistance notes I made right after my engine restoration to check resistances and gauge measurements. Do you happen to know what the sender resistance should be at 100C? When it is slightly cooler out it the gauge reads just a little under 3/4. Water will boil at 96C where I live so I can check calibration with that and some other thermometers. Also, at 4000' elevation engines run hotter so 102C would be equivalent to 100C at sea level. Disregard the other temperatures because that was hoses which you said are irrelevant. It has never even touched red and I have digital dash which bugs me even if the washer coolant is low so I'm sure it would really be alerting me if it was too hot.

FredR

John,

I do not know the resistance values for the gauge sender other than they are different from the temp2 sensor. For some weird reason [age?] I had it my head you had an 87 model so yes the 89 is indeed a digi dash and if you have not seen an alarm then i would think you have nothing to worry about.

Elevation does not cause the engine to run hotter per se but for sure you will be closer to the limits of the system to control the engine temperature given the thermal conductivity of air is about 10% lower at 4000 ft ABMSL. The fans are not maxxed out until the coolant return temperature hits 95C and that corresponds to coolant leaving the engine at approx 103C. The radiator has a little margin in it design wise but not a whole lot and just as I had to learn to optimise cooling because of high ambient temperatures I would think you may well be advised to do the same things to help optimise control the system due to elevation.

The things you can do to improve cooling are:
!. Remove the slats - such improves air flow through the system a little. They are easy to remove and can be reinstated if needed. Have you ever seen a GTS with those things fitted?
2. No idea what your winter low temps are but in most locations a 25% coolant mix is perfectly acceptable- i run with 33% mix and if i had any sense i would run with 100% water to avoid the cylinder head corrosion problem i have advised the list about with appropriate anti corrosion chemicals- see the next point.
3. You should also run with a bottle of Redline water wetter in the coolant. Not only does this help improve heat transfer across the boundary layers in the block and the radiator it also contains anti corrosion chemicals
4. Sounds as though you may also have the Spal 12 inch twin fan setup installed like i have- that also helps a tad.
5. If the system is really struggling one can also take the PWM system out of the control loop and power the fans directly given I found that with the PWM system the fans run a bit slower at max output of the control system.
6. The radiator capacity can slo be increased slightly to improve the overall heat transfer coefficient but we are talking about a couple of per cent or so. Still every little helps.
7. The later models with external engine oil coolers help not sure if the 89 model has this or not. For the sautomatic models i also recommend taking the tranny cooler out of the radiator and just use the external cooler. i suspect Porsche ran the ATF through the radiator to help warm it up quicker in cooler climates strange as that may seem.

Do not pay too much attention to where the dash panel gauge needle sits whatis more imp[oprtant is to look out for any signs of different behaviour [higher temeprature indicated]. As stated in my previous post the needle will tpyically try to sit just to the left of the last white line on the gauge but there are differences example to example but note that both of the 928's I have owned behaved identically. That the lines are coloured white is because that is convention considered to be OK. When the needle goes into the "red zone" that is the time to start worrying. When I purchased my first 90S4 the needle would go beyond the last white line when the ambient temps hit 44C. After optimising the system never seen that. That being said during the hot season I tend to store the car away during July and August and when i drive it I tend to do so after 4pm but it will run OK in midday heat. I have done a number of runs between Muscat and Dubai in 44C heat in years gone by. These days i avoid Dubai like the plague simply due to the traffic volume.

Forget lower temperature thermostats- they do not work properly during cooler weather and these engines like to run at the temperatures they were designed to operate at

Bulvot

This post has my findings for the thermistors, including the dash gauge, temperature curves and actual measurements: https://rennlist.com/forums/928-foru...er-things.html

You can use that to precisely test your own coolant sensors and the dash gauge.

In my opinion, your temperatures are normal for the 928.

With a properly working coolant and fan system, including a proper coolant reservoir pressure cap and a 50/50 mix, your coolant boiling point is about 260 degrees Fahrenheit. That's a fair amount of head room.

In similar weather conditions to what you've stated, I've observed my coolant temperatures reaching as high as 219F after sustained low RPM (low coolant pump speed), high engine load, repeated short intervals of low boost, and moderate vehicle speed while climbing a long uphill mountain highway. That temperature was only briefly hit before the fans caught up and brought the temperatures back down. At a sustained idle in those temperature conditions, or while cruising on level ground, I am seeing around 200 to 205.

I have a lot of data since I log every drive with my ECU, and the thermistor calibrations are accurate so my #'s are not ballpark. They are exact. I have found that shooting the coolant bridge with an IR temperature sensor only gets you in the ballpark, but isn't accurate enough to do more than say whether it's way too hot, or way too cold (e.g. 250 or 150). Anywhere in the 200ish ball park just means it's close. If you have a nice IR gun, you can adjust it for the surface you're shooting, or hook up the thermistor for a better value, but it's still not going to match the thermistor since it's not in the coolant.

For reference, I'm running the Hans slim profile fan shroud with dual SPAL 12" 30102038 puller fans (1,640 CFM each at 0 static pressure, 431 at 1 inH2O), on/off fan controller triggered by a sensor in the OEM position in the radiator, CFS radiator, waterless coolant (worse heat transfer so it is prone to raising engine operating temperature), intercooler heat exchanger in front of AC condenser (so it adds heat to the stack), single SPAL 10" 30100435 pusher fan in front of heat exchanger (1,350 CFM at 0 static pressure), stock coolant thermostat, and a turbo (which adds heat to the intake, removed by the intercooler and put into the radiator stack).

One significant observation that I've made is that the puller radiator fans run a lot more on the 928 than people realize. They are critical to maintaining reasonable coolant temperatures at vehicle speeds much higher than on other cars with better air flow into the radiator stack.

I can't speak for the 944, and it probably matters a lot what the vintage is, but we've got a 924S with the 944 motor and the temperature gauge definitely moves up and down in sync with weather conditions, fan status and type of driving. Is it possible that your model year of 944 has a coolant temperature gauge with a large flat spot in its curve to make the temperatures appear more stable and worry the driver less? Our modern cars all seem to do that, even though we can see the temperatures rise and fall on our plug in engine management tools.

Vargleblaarg

I'm new to 928s, but as I've been fixing up my '87 5-speed I've noticed it runs warm (at least indicated temps) compared to other cars. As Bulvot noted, modern cars often fake the gauge readings. I know for a fact my late model truck does this.

I live in the metro Phoenix area and on a 116 degree day recently the car was hovering just below the red mark but never touched it. IR gun readings at the water bridge were around 230 degrees when I stopped. This was with the AC on (which at that temp doesn't work all too well). In slightly cooler (still above 100 degrees) it seems to do much better (just a needle line above 3/4 and 215 on the IR gun). That is to say I think there is just an upper limit to the cooling capacity of the stock components in such extreme heat.

I've recently completed a great deal of work including a new water pump, hoses, t-stat, coolant, among other service items. The fans are both operational as well. I also replaced the rear t-stat seal and tried two different thermostats (180 and 195 degree). The cooling flaps are still in place and operational but I plan to fully delete them and test again. To my eye even fully open seems like it would block some airflow. I do think the lower temp t-stat helped a bit but I also replaced the rear t-stat seal at the same time, so it's possible that helped more. Looking at the design of the water bridge I can see how some coolant can bypass the radiator and circulate through the block if the t-stat is not fully open and if there are problems with the rear seal.

I would love to setup a remote sensor on the inlet and outlet of the radiator to see how it's doing at speed. As you noted, the temps climb regardless of driving conditions. When I shoot the IR gun at the top part of the radiator after a drive it's never above 200 which seems to indicate that the radiator has sufficient cooling capacity, but the flow of coolant into the engine somehow still ends up a little on the warm side.

FredR

Forget what other cars may or may not do- the 928 uses a pure analogue instrumentation system that accurately measures temperature from ambient conditions to alarm level conditions.

As with any thermal system the 928 cooling system is designed to dump a fixed amount of heat at a given temperature with the ac unit running. The fan programme is modulating up to a temperatre of 95C in the coolant return line that approximates to about 103C in the coolant as it leaves the water bridge. After that temerature is controlled by the coolant heating up some more creating more driving force relative to the air temperature being used to drive the cooling process. At some point the high temperatuer alarm will sound- unfortuantely Porsche in their innfinite wisdom do not publish such datsa as I am aware but as I recall it is something like 110C or a tad more. If you want to convert these numbers into British dinosaur units multiply by 9/5 and add 32 to get degrees F.

In my experince over here where ambient summer time temps typically reach 45C here on the Gulf of Oman coastline that is about the limit of design control. On examples [pre-89 S4?] where the oil cooler is in the radiator they struggle a bit more compared to those models with the oil cooler underneath the radiator like my late 90S4 and my 92 GTS.

For someone likke yourself living in Phoenix I have mapped out the things one can do to help optimise cooling performance. The stock radiator can only be improved upon very marginally and the driver for the CSF radiator is replacement cost compared to stock.

Did any off you chaps out there ever stop to wonder how it is that the very first 928's producing 240 BHP have exactly the same cooling system as the GTS and how those building monster motors with outputs in excess of 600 bhp all function with the same cooling system design not to mention supercharged/turbo examples turning out nigh on 1000 bhp?

Bulvot

OEM coolant warning is 115C (239F). Clearly marked on the thermistor itself and referenced in more detail in my link above.

Practically speaking, the 928's cooling limiter is air flow through the heat exchanger stack. Even at highway speeds the fans are necessary on very hot days.

In my opinion and experience, the original poster's observations are normal and safe.

Chester89

My 89 flaps are fully open when the car is shut off. Closed when warming up. How do I go about keeping them open? Is it as simple as removing the fuse or relay? Thanks.

Johnsrestos

Thank you all for the detailed comments. One thing I have seen in various comments here and elsewhere is how the belly pan contributes to cooling. Unfortunately, mine does not have either belly pan. They make a front one in aluminum and the plastic rear one is available, but the front aluminum one does not come with factory "vents". You are supposed to reuse your own, which I do not have. How important are those vents? Can I run without them and will the belly pans really cool it that much or waste of time and money? I always hate removing those things on vehicles to do work on them so if we are talking another degree or so on highway only it probably isn't worth it. Those pans cost more than a new radiator. Also, if I was to disconnect the transmission cooler lines, what do I do with them and how do I plug it off?

Mrmerlin

why do you want to disconnect the trans cooler lines?



NOTE I would not remove the cooling capability from the transmission ,
even if its plumbed into the radiator to warm the fluid to operating temp,
your car has two oil coolers for the transmission,
and you should still have an external cooler.

Johnsrestos

Fred suggested it and he is the hot climate 928 expert.

"7. The later models with external engine oil coolers help not sure if the 89 model has this or not. For the automatic models i also recommend taking the tranny cooler out of the radiator and just use the external cooler. i suspect Porsche ran the ATF through the radiator to help warm it up quicker in cooler climates strange as that may seem."

FredR

John,

Much as I appreciate the accolade of being perceived as an "expert" I prefer to think of myself as a well experienced 25 years 928 owner with a professional engineering background capable of analysing any engineered system. As an FYI I analysed why 928S4's were suffering from TBF [due to fretting corrosion]back in 1999 and regularly shearing torque tube shafts and was the [joint] first 928 owner to implement a solution [Loctite] to the problem costing no more than a few cents! I also analysed why 928 cylinder heads corrode and head gaskets degrade the way they do.

I advise owners like yourself of the mods I implemented many moons ago [AKA around 1999] after careful consideration and study of various suggestions that were made by very credible members of this list at the time. I do not do things "***** nilly" just because someone mentions such in dispatches as it were and it is up to the individual owner whether or not they think it is appropriate to follow the same path.

When studying how the cooling system works I found that whereas the engine gets up to working temperature in about 3 minutes the lube system takes up to about 10 minutes to get fully warmed judging by how the oil pressure gauge settles at idle and given that when fully warmed up it indicates about 1.5 barg as per the dash panel gauge. The coolant return temperature at circa 90C is a viable source for cooling engine oil that typically operates in the range of 110C to 120C but cooling via the return tank will of course heat up the coolant [not good on a marginal system]. Given the cooling system is marginal at 44C with the ac running I was somewhat surprised to learn that the earlier versions of the S4 cooled the engine oil in that manner but my 90S4 had the external cooler under the radiator thus that made sense to me even though this is not without its own issues. In your case with the 89 model my grey matter does not tell me whether it has the external cooler- for some reason i have it in my head that this was a MY90 modification like the PSD system but I am not at all sure of this. Thus if your model has the in-radiator cooler then retrofitting the external oil cooler package would be a viable modification should such be needed - [hopefully already incorporated into your 928S4 during initial build].

For the auto tranny the modus operandi surprised me somewhat in that it seemed there was no control temperature wise- it simply routes the pressurised ATF in the torque converter back to the tranny sump via the hot end of the radiator and then the external air cooler. The strange thing [to me] was that auto tranny ATF systems normally operate in the range of 70C to 90C so why would Porsche route it via the inlet tank of the radiator that operates at 98C when it will act as a heater? The only logical explanation for this I could come up with is that perhaps they used it to help warm the ATF circuit up quicker in cold climates and once up to operating temperature it is probably at or around the same temperature as the coolant. if i were ever to return to my native UK I would probably leave the system as Porsche designed it but given i have no cold weather to worry about I prefer to keep the ATF system and the cooling system separate to avoid the possibility of a wrecked transmission due to ingress of coolant due to coil failure as we see happen from time to time. Thus my custom made alloy radiator has no coils in the end tanks. If you have cold winters in Phoenix then I might not implement this mod that thermally speaking is at best marginal and may even be of no thermal gain [with hindsight].

With respect to the panels under the engine the stock item in the front section is plastic and the rear section is alloy. The front panel is quite often toast the first time one drives over a speed bump with standard suspension units and invariably they end looking like Rocky Balboa's face after 12 rounds with Carl Weathers! As i understand these panels do two things:
1. Aerodynamically they help pull the car to the ground at speeds in excess of about 90 mph. Exactly how this happens i am not sure - I suspect it is because they deflect air as it passes under the car and stops it from creating resistance against the components in the engine under bay that it would otherwise impact and maybe velocity changes at speed cause a differntial pressure that pulls the car down slightly thus a venturi type effect creating slightly reduced pressure on the underside of the car.
2. The flow of air over the panel corrugations will to some extent tend to suck hot air out of the engine bay rather than compete with it and thus help support the two puller fans work more efficiently as seems to happen.

With respect to the vents i kind of smile wryly when I see this topic discussed on the list. Everyone seems to know "they are needed" to help support longevity of the stock hydraulic engine mounts but no one has ever presented a scrap of evidence to verify whether this is the case or not. Personally I have a feeling this is a "red herring" . I run with the Volvo mounts brother Roger supplies - they have been in there for the last 8 years or so and no issues to date running without those things in our very hot climate. To be fair I do not do much mileage so that probably does not mean much. The air temperature flowing through the engine bay heats up to about 75C or so I reckon and whereas heat will be emitted from the exhaust manifods whether those ducts can do anything of value is at best a "contentious" issue. Given the airflow through the engine bay I doubt they have much in the way of value but I cannot prove that one way or the other but I am open minded on the subject.

Bulvot

Fred, I'm going to add a couple of more data points. As you know, I have an aftermarket ECU which precisely monitors the temperature of the coolant and oil. It also can accurately infer engine volumetric efficiency through precise monitoring of fueling, air intake and exhaust gas oxygen content.

-I define the oil's operating temperature as the temperature at which it stabilizes. Because it stabilizes at different temperatures depending on ambient conditions.
-Oil temperature stabilization coincides with VE stabilization. Even the WSM states that air/fuel mixture adjustments should be done after oil temperature is warmed up (not coolant).
-With regular 50/50 coolant and a mid-80's F ambient temperature, oil takes ~10 minutes to reach full operating temperature with light load driving on flat roads (CSF radiator and using the integrated oil cooling loop).
-With waterless coolant and the same ambient conditions, oil takes ~30 minutes to reach full operating temperature with light load driving on flat roads (CSF radiator and using the integrated oil cooling loop). No difference in how long the coolant temperatures take to stabilize.
-Oil temperatures, once stabilized, are noticeably lower with waterless coolant versus standard 50/50 coolant.

The above results are very repeatable and have similar relative changes that coincide with changes in ambient conditions.

I've tried many different cooling stack configurations, so I have a lot of empirical data from which to make observations.

FredR

Chris,

i always enjoy reading what you have to say regarding your exploits and your attention to detail is exemplary.

What you say about the stock cooling system aligns with my findings and no need for further comment in that regard.

What I did find surprising was when I read you were using waterless coolant. I have no personal hands on experience of such but based on what I researched about that stuff some 6 years ago when i was investigating what I had seen and experienced cylinder head corrosion wise was enough to put me off the stuff for ever and a day. In an engine specifically designed to use the stuff fair enough maybe but not our 928 motors or so I concluded.

To be fair I have no idea whose brew you deploy but the stuff I looked into was a product called Evans Waterless Coolant and I recall a very specific article commenting how the cylinder head temperatures ran much hotter with this stuff circulating. Now you advise that it takes the stuff you use 30 minutes to warm up the oil and that the oil temperatures run lower - do you perceive this to be a good thing or a bad thing? Then as i thought about what you said a bit more it seemed obvious to me that if the cylinder heads run hotter with this stuff then it is also reasonable that if it is not removing the heat needed like the stock system does then it would seem reasonable that it will take longer to heat the oil up. That it takes 30 minutes to heat the oil up would be sufficient to put me off using the stuff irrespective of any other considerations given most of my driving activity these days are fun runs that last about 30 minutes or so. Of course if it is not picking up heat in the same way then it is reasonable that it will run cooler but that is not neccesarily a good thing either. Then of course there is the issue of specific heat and waterless coolants do not come vaguely close to water in this regard and even a 50/50 glycol mix performs better as i recall. Your motivations for using this stuff appreciated and please feel free to enlighten me if there is anything i have misundertood.

The other major conclusion i formed when looking at the cooling system expained something that mystified me before I looked into it. My conclusion was that once an engine configuration is made the cooling needs remain nearly constant irrespective of the power that engine derives. Historically it seems that most folks who comment on such seem to think that the reason the 928 engine can take whatever is thrown at it is because Porsche in their infinite generosity sized the cooling system to handle way more power than it was designed for and thus why it can handle 1000 bhp motors like the one Carl Faucett built for Bonneville! Needless to say this is thermodynamic nonsense. Gasoline burns at around 1000C no matter how much of the stuff is burnt and the heat needed to be removed by the water bath is that required to maintain the designed for average metal temperatures. The thing that controls this is the resistance to heat transfer at the boundary layer and thus why the average metal temperature is only marginally higher than the coolant temperature . As far as i know this boundary layer resistance does not change by much if anything at all - when more fuel/air is fired the amount of heat going out through the exhaust system most certainly increases. Needless to say more power going through the big ends and mains will heat the engine oil more and if the oil runs through the radiator it will stress the radiator more. It would be interesting to know if such aligns with your findings having turbo charged your motor- I am pretty sure it will. The only thing I am a bit uncertain about is whether the increased combustion chamber pressures exert more pressure on the piston rings and whether that increases the frictional loading on the cylinder walls. I suspect there may be a small increment but nothing spectacular.