Water temp differance between Engine - Top radhose

[Vee]

S13 CA18DET @ 260HP

Something odd-ish that I noticed. I have been doing a few top speed runs that made the stock water temp gauge climb towards the red border for the first time. However the aftermarket one wasn't showing over 85c (185f). The temperatures given by two different sensors were quite far apart.

As soon as the stock gauge reaches a notch before red, I back off. The defi reads 85c. I can cruise 180-190km/h fine but over that more power is needed and that is when the engine starts getting warm.

Now having looked at this thread and incredible input by John Bennet: http://www.sxoc.com/vbb/showpost.php...4&postcount=64

It looks as the water temperature measured before the thermostat by the stock sender (replaced sender and stat 10k ago) should be around 105c (221f), so it differs by about 20c from the radhose temperature. Can this be true?

I have measured the Defi with a water boiler and it read 100c (212f) precisely, so I'm thinking that the reading from the top hose is accurate. I also think - based on mr. Bennets findings - that the stock gauge is accurate. But where does this 20-25c difference comes from, what am I missing?

Could it be an air lock? I doubt it as the gauges all work fine the rest of the time, and I also purged the cooling system a few times. Well.. purged, I took the stock sender out and filled it up till coolant came out and let it idle with rad cap off a few times.

What are your views guys?

--

Ok I had a thought while writing this;

The "overheating" happens during 200kp/h + action. That goes paired with a lot of cooling air. Can it be the front of the radiator hose (where the accurate sensor is) gets cooled enough to drop the temp 15c? Cause the airflow is extreme, plus the visco fan is blowing to the side as well, right at the sensor housing...?

--

[orion]

One question: Is the DEFI sensor properly grounded...?

I have seen, with Autometer gauge sending units, incorrect readings when using one of those radiator hose adapters. They are basically "un-grounded", and the sensor required good grounding for proper readings.

This usually calls for running a ground wire to the body of the adapter.

- Brian

[Vee]

Yup, it is grounded with a seperate wire. The defi seems precise as I had the sensor in a water boiler and it creapt up steadily to 100C precise and held as long as the water boiled.

[orion]

^^^ Cool, just checking.

[Matt93SE]

Depending on where the temp sensors are lcoated, you can definitely see a 15C change in temp from one end of the engine to the other. I don't have time to read that article, but suffice it to say placement of the sensor is important to get accurate readings.

At those speeds, you can also be dealing with some cavitation of the coolant as well. I've had times in the past where I had to jack the front of the car up quite high in order to bleed the cooling system of all the air. It definitely helped after I removed the heater core, but I still have issues with air in the system on occasion.

then again, I haven't changed the coolant in the car in 2-3 years.. probably would be a good thing to do, eh?

[Vee]

Both sensors are 30cm/12" length apart, so not far. That is why I'm concerned.

[cdlong]

Got an engine bay pic for those of us not familiar with the CA18? They're only separated by 12" of hose, right? It could be actual cooling, but given the flow rate in the hose and how well it transfers heat (not very), I'd say probably not.

[Vee]

Looks like this:

[Equinox]

Doesn't the hot coolant from the engine flow out via the upper radiator hose though? So the reading in the Defi should be higher than the reading from the stock coolant temp. sensor.

I usually have a 5°C difference between the temperature in the upper coolant elbow and the stock coolant sensor while on throttle at the track.

Maybe your thermostat is somehow closing itself at high speed?


(SR20 I know but the system is similar)

[Vee]

Yup, it does. It comes from the OEM sensor / Thermostat housing and travels thru the upper hose into the radiator.

I can only think of why the temp in the Defi is lower, is because rammed air travels over the hose/sensor housing (which is metal btw) first and cools it down a tad. Now I would'nt lose any sleep over 5c but 15 to 20 got me thinking. But I talked to a know-quite-a-lot-about-cars-n-****-guy, and he told me it is likely, at 200kph the airflow is strong enough to cool 20c. So I am leaning towards it a bit...

[greenman100]

wrap the sensor in something insulating, like a towel. repeat said speed run. note changes.

[WorkInProgressK]

OP, your talking about high speeds but at what RPM. If its high, I might just say the flow is stopping since your engine water pump is probably cavitating. That would explain why the cold entry of water is actually getting warmer.

[greenman100]

OP, your talking about high speeds but at what RPM. If its high, I might just say the flow is stopping since your engine water pump is probably cavitating. That would explain why the cold entry of water is actually getting warmer.

If the flow is slowed (or stopped) to the point that the engine is warming up the coolant near itself through conduction, you'd see temps a lottttt higher than he's seeing.

[WorkInProgressK]

Proof? ^^

[greenman100]

As a general rule of thumb, gas engines created approx 2x their shaft HP in the form of waste heat. Say half of that is dissipated through the cooling loop. That's 260HP, or 193kW. We'll assume the other half of the heat energy goes out the exhaust, or is conducted away from the engine block.

There is an elementary equation from basic thermodynamics that states that the rate of heat transfer (Q) equals the mass flow rate (M) times a Constant (the specific heat of water) times the Delta T (fluid temp out minus fluid temp in):

Q = M x C x Delta T

In other words, the rate of heat transfer is directly proportional to mass flow rate.

In our case 193,000 = M x 4.18 x Delta T.

Now, we know that the lowest possible temperature in an warmed up engine is the thermostat temp, so lets call that 170F, or 77C. We know the hottest temperature in a warmed up engine is about 250F, or 121C. Delta T is 121-77, or 44C. Solving for M, we find a mass flow rate of:

1049 g/s, or 1.1kg/sec, or ~1.1L/sec, or .26Gallons/sec, or 15GPM

That covers forced convection. If you drop the flowrate to zero, or nearly zero, that that 193kW has to pass through conduction in the coolant loop. Lets look at that. Water has a thermal conductivity of 0.6W/m-k in liquid form. Lets also assume there's 0.3m between his measurement points, and that a coolant hose has an area of 25cm2.

Then, if both sensors are true, and he's seeing a 20C change, we get the following through conduction:

20*0.6*0.025/.3=1W.

I guess you could assume that you're conducting through both hoses. Fine. 2W conducted. Not quite enough to cool an engine creating 193kW. In fact, if you wanted to pass 193kW through conduction of through two coolant hoses, you'd need 193,000 degC of delta T between the sensors. Which is considerably hotter than the surface of the sun.

For a sanity check, the Meziere electric water pump for the SR20DET is rated at 20GPM. Not far off from the 15GPM I calculated to be required at full load.

I'd theorize that the bigger issue here is sensor error. I have a strong feeling the stock sender isnt linear like that one poster thought, and I'd bet the DEFI sensor is experiencing cooling from its housing.