When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Been working on my 82 GSL. My factor temperature gauge doesn’t work. The temperature sensor was replaced but the gauge still doesn’t work. So, I installed an aftermarket temperature sensor and gauge. The sensor is installed in the thermostat housing. The car also have have electric fans.
When the car is running at idle, the temperature is between 185-195. While driving it goes up around 220. This is after 5 minutes. If I turn the A/C on it could go above 250. I also took temperature reading at the radiator, thermostat housing and the engine. The temperature that I’m getting at those points are right at 200.
Do you guys think that I have a faulty coolant sensor or should I be worried. What else should I be checking?
Depends on how your electric fans are triggered? Do you use a temp sensor to turn them on? If the fans are on all the time they will cause heat up once you go faster than 25-30 mph because they start impeding air flow.
Depends on how your electric fans are triggered? Do you use a temp sensor to turn them on? If the fans are on all the time they will cause heat up once you go faster than 25-30 mph because they start impeding air flow.
They’re on all the time. What’s the best type of temperature sensor to use? Are they difficult to install?
The running fan will not make temperatures go up when you are at speed. The improvement of airflow is fairly small due to the operating fan at speed, but it definitely will not reduce the airflow through the radiator.
You can check your existing temp sensor, by sticking it in boiling water. I did this under the hood of the car using a small camp stove. Kind of awkward, but showed that water boiled at 232 degF. So yes, obviously my sensor/gauge was reading 20 degF too high at that condition. If you are more comfortable pulling the sensor and gauge you can obviously do the test on the stove. My gauge was electric, so I installed a resistor in the circuit to correct the temperature in the important range. The sensor is not linear, so outside the temp range of about 180-220 degF, the temperature is not so accurate, but I don't really care.
It sounds like things are running hot though. Assuming the temp reading is close:
what is the condition of your radiator? Pretty new or old and gummed up?
Do you have the proper thermostat installed? The kind that blocks the coolant bypass flow when it opens?
Stock oil cooler installed and thermostat not bypassing the cooler?
Your aftermarket gauge is inaccurate. The pressures in your cooling system would blow the radiator out if it was actually reaching 250F.
Boyle's Law states that Pressure, Temperature, and Volume are related. Volume of the coolant in the system is fixed, apart from the Radiator Cap and the Overflow Bottle. As Temperature increases, Pressure increases as well. With your coolant temp at 250F, the Pressure would quickly surpass ~20psi which the Radiator Cap regulates, and the coolant would steam into the overflow and out of the system. You'd quickly run out of coolant, the engine would overheat, and the bearings would seize.
Most internal combustion engines are designed to operate effectively at temps centering around 200F for a sustained period of time, yes, that temp can rise above 212F - where water boils under standard air pressure, but your antifreeze additives prevent freezing below 32F and prevent boiling up to about 225+F, which is also why having antifreeze of the proper mix is so important during summertime.
As someone suggested already, you can test the gauge by putting it in boiling water and it should read ~212F (*depending on your altitude above sea level), but that still doesn't let you test how far off it will be where it matters, i.e., when it's nearing the overheating point. Sustained operation at temps over 225F will quickly lead to oil breakdown, loss of engine oil viscosity lubricating the bearings, and will cause severe engine damage.
I agree that it is likely that the gauge is probably off. Testing in boiling water will give you accurate results in the range that counts though. If the coolant is running as high as 212 degF you have a problem that you need to address. It should not really ever be going over the temperature of the thermostat rating. Above that there is nothing controlling the temperature and every bit of additional heat into the engine makes the temperature increase. My race car (only 150HP at the wheels) runs 190 degF happily for hours on end (literally), even with ambient temperatures over 100 degF.
LongDuck - Actually Boyle Law applies to gases only, not liquids. It is a special case of the Ideal Gas Law, which is PV=nRT. It is actually the vapor pressure of the liquid that is governing. The vapor pressure at the boiling point is equal to 1 atm (or ambient pressure). More or less as temperature increases, the vapor pressure increases. That is a short cut explanation, but... Anyway, the table below shows the boiling point for different pressures in the radiator vs water/coolant ratio. Yep, 250 is pretty much pushing it for not boiling in a radiator, but it isn't a slam dunk that it will be boiling. Depends alot on the pressure rating of your radiator cap. Again, if you are anywhere near 212 degF on a regular basis, I recommend looking at the integrity of your cooling system.
Carl
A sensor issue is sounding like the most likely possibility. Do as Carl said and test the sensor in some boiling water (boiling water cannot exceed 212F at atmospheric pressure, so you're looking for the gauge to read that). That'll answer the sensor/gauge question at least.
As for the fans, depending on the setup, an engaged cooling fan might cause a restriction in air flow at speed. Generally speaking however, poorly designed shrouds are often the real offender. Depending on the vehicle use case and fan size, a shroud might not even be needed. Above idle and low speed operation, bad shrouds can pose notable restrictions to airflow. This is why many OEM shrouds have flaps which allow air to flow through them freely at speed.
For controlling the fan, I honestly suggest a radiator push-in probe. Quick and simple, and they work far better than most folks seem to think (just get a good brand, I suggest Derale or Spal). Don't bother getting a controller with a temp readout, it's not very useful since your radiator isn't the same temp as your engine (ideally). A proper thread in sensor is a better option, however it must be installed on the radiator side of the thermostat (preferably in the radiator itself). The issue a lot of folks run into is that they see the threaded sensor and look for an existing threaded connection - usually the factory choke temp switch on the water pump housing...
Problem is that the water pump housing location reads engine temp. For example, lets say you have the controller set to turn the fan on at 180F. When referencing engine temp, the radiator fan will run until the engine hits 170F (most simple controllers use a ~10F delta). No matter what temp you set it to, the engine will always be fluctuating in temperature by this relatively large amount, and the thermostat can't do anything to prevent it. The better solution is to have the controller reference the radiators coolant temp, which is essentially what clutch fans do. This way the fan maintains a temperature range of maybe 150-200F at the radiator (depending on sensor location), and the thermostat controls how much coolant is bled from the engine loop to maintain the proper operating temp of 180F. I say there's no point in having a temp readout on a controller used in this manner as the actual temperature will vary depending on the setup and sensor location. What you instead need to do is adjust the controller such that the thermostat is able to maintain engine temp correctly without over or undershooting, and that just takes a little trial and error.
My apologies for the bit of a side track there. Hopefully the information is useful if you decided to get a fan controller.
Last edited by Benjamin4456; 06-29-22 at 07:56 PM.
Reason: cut bad info, elaborated on controller sensor position
The running fan will not make temperatures go up when you are at speed. The improvement of airflow is fairly small due to the operating fan at speed, but it definitely will not reduce the airflow through the radiator.
You can check your existing temp sensor, by sticking it in boiling water. I did this under the hood of the car using a small camp stove. Kind of awkward, but showed that water boiled at 232 degF. So yes, obviously my sensor/gauge was reading 20 degF too high at that condition. If you are more comfortable pulling the sensor and gauge you can obviously do the test on the stove. My gauge was electric, so I installed a resistor in the circuit to correct the temperature in the important range. The sensor is not linear, so outside the temp range of about 180-220 degF, the temperature is not so accurate, but I don't really care.
It sounds like things are running hot though. Assuming the temp reading is close:
what is the condition of your radiator? Pretty new or old and gummed up?
Do you have the proper thermostat installed? The kind that blocks the coolant bypass flow when it opens?
Stock oil cooler installed and thermostat not bypassing the cooler?
Good Luck
Carl
The radiator (3 row aftermarket), radiator cap, and thermostat are all new. The oil cooler is an aftermarket cooler. It’s new also. What type of resistor did you use.
Based on all the new stuff it sounds like your sensor/gauge is probably off. I am assuming that you installed an electric gauge/sensor? If so, to get the resistance correct you will need to know the resistance of the sensor at different temperatures and the resistance the gauge is expecting. I am using a Marshall Instruments gauge and I was able to find that information on their website. From that info, it was clear that the sensor was clearly not the right one for the gauge. It was just easier to put a resistor inline to get the proper indication than to swap out either the gauge or the sensor.
This picture shows how I set up the boiling temperature measurement. To prove to the manufacturer that the supplied sender was incorrect, I actually measured the resistance of the sender in hot oil at various temperatures on the kitchen stove (my wife was at the movies with friends and had no idea so I am still here to tell the story). The second picture shows the sender/sensor resistance data from the manufacturer's website.
The other question though about cooling is how close is the aftermarket oil cooler to the stock cooler? Does the amount of cooling surface seem to be the same? Is it mounted in the same position as stock below the radiator so no air can bypass? Even with an undersized oil cooler, I would be surprised if your coolant would get as hot as you are indicating driving it on the street.
I also forgot to ask. I assume the thermostat is a bypass type. If not, you lose a lot of cooling capacity because not all the coolant is going through the radiator. Bypass Thermostat. As the thermostat opens, the disc on the bottom blocks the bypass port Non-Bypass Thermostat
Yes, my thermostat is the bypass type. Also, my fans and the radiator came as a set. Yes, there’s 2 fans with each have their on shroud around them. I’m going to test the temperature sensor 1st. Thank you guys for all the knowledge.
Based on all the new stuff it sounds like your sensor/gauge is probably off. I am assuming that you installed an electric gauge/sensor? If so, to get the resistance correct you will need to know the resistance of the sensor at different temperatures and the resistance the gauge is expecting. I am using a Marshall Instruments gauge and I was able to find that information on their website. From that info, it was clear that the sensor was clearly not the right one for the gauge. It was just easier to put a resistor inline to get the proper indication than to swap out either the gauge or the sensor.
This picture shows how I set up the boiling temperature measurement. To prove to the manufacturer that the supplied sender was incorrect, I actually measured the resistance of the sender in hot oil at various temperatures on the kitchen stove (my wife was at the movies with friends and had no idea so I am still here to tell the story). The second picture shows the sender/sensor resistance data from the manufacturer's website.
The other question though about cooling is how close is the aftermarket oil cooler to the stock cooler? Does the amount of cooling surface seem to be the same? Is it mounted in the same position as stock below the radiator so no air can bypass? Even with an undersized oil cooler, I would be surprised if your coolant would get as hot as you are indicating driving it on the street.
Cheers,
Carl
I finally had time to work on the 7. I have a new temperature gauge and sensor installed now but I’m still having the same issue. Where do I go from here? How do I figure out what type of resistor I need? I guess I will contact the Manufacturer of the sensor and gauge.
Did you ever test the gauge/sender combination in boiling water to see what the indicated temperature is when the sender is in boiling water? If so, what were the results?
Beyond that. Is the new gauge and sender the same model? What brand and model is it? The manufacturer should be able to provide the temperature vs resistance curve for the sender (which should also be what the gauge is expecting). Once you get the information about what the gauge is expecting for resistance at temperature, you need to see what resistance your sender has at temperature. If you can't get the temperature/resistance curve and the "boiling temperature" test shows that the gauge is not reading correctly, you can figure out what size resistor to install by installing a potentiometer in series between the sender and the gauge. Adjust the potentiometer until the gauge reads 212 degF and then measure the resistance of the potentiometer. That will tell you how much resistance to add to get the gauge accurate around 212 degF, which is an OK range, since that is +/- 20 degF where the gauge will operate.
Once you know that the gauge is accurate or not (you may have already done the boiling water test) I can give you some more specific advice on how to proceed (maybe even do the potentiometer test for you).
06-28-22, 01:36 PM
