+ Reply to Thread
Results 1 to 9 of 9
01-31-2009 01:04 PM #1
Anyone know how to test the coolant fan relay?
The problem is a High Speed Fan Run-On After Warm Up
Within a few minutes, the high speed fan begins running when the engine reaches normal operating temps of about 180-185’F. After it activates, and it will not shut down until the ignition was turned off. After restarting, the fan will reactivate within two or three minutes.
* 4.0 liter '97 XK8
* New Water Pump (metal impeller);
* New Thermostat (w/ New Metal housing);
* Not overheating; engine temp 180'F. at the tsat housing (laser thermometer readings)
*There are no "Check Engine lights",
* Low speed fan seems to work OK before the high speed fan problem begins
The coolant temp sensor was changed with no result. (old one ohm’s out OK)
Cleaned the contacts on the A/C pressure sensor (on the AC line behind the radiator)
Pulled the Fan Relay but I do not know how to test it ??
Are there any ECM tests for cooling fan operation?
01-31-2009 04:00 PM #2
Re: Anyone know how to test the coolant fan relay?
Just a guess, but I have tested similar relays from my other Jag (1985 XK8 lumper). I simply isolate the + and - from the power input side, and then put a 12V test light or ohm meter on the other two leads and energize it by connecting the + and - to a battery or 12V current source. If it goes back to open circuit every time you cut the power, perhaps you have some kind of short in the ECM. I believe motors can act like power generators once power is cut and perhaps the spinning motor is generating a small current back through the wiring that is being picked up by the ECM which has a short on the board that is somehow closing the circuit. You might also consider detaching the cables to the ECM and cleaning them, and making sure there isn't any material in the fuse box that is causing a short to another active circuit. You could switch the suspect relay with an identical one in another electronic circuit like the fuel pump relay to see if you still get the same symptom. If this is a multiplexed circuit, perhaps it could also be affected by a weakening battery that is failing to maintain the right mathematical values dedicated to the fan circuit, but that is pure speculation on my part. I read somewhere that there is less wiring on these cars than you would normally need because they are able to run multiple circuits through the same wires. That would of course require the kind of electronic signal splitting that a computer is capable of (ECM).
01-31-2009 04:17 PM #3
After closer reading.....
I have another suggestion. To the best of my knowledge the fans come on independently at different temperatures. That would suggest that either your temp sensor transmitter or the ECM receptors are screwed up. I suspect the ECM receives a varying amount of current from the sensors depending on how hot the coolant is. The sensor is probably a thermocouple designed to vary the current it generates based on the heat it senses, so if in fact the fans are designed to switch on and off at different temps, it looks more like an ECM problem or some wierd breakdown in the thermocouple that shoots out a max signal after two minutes of running. The fact that they come on shortly after starting the car but go off after you turn the key off suggests that the relay is a long shot.
01-31-2009 10:43 PM #4
The sensor function is based on its resistance
As it heats up, resistance is lowered and the ECM reads that data. The JTIS supplies temp and ohms levels. Both coolant temp sensors tested within spec. consequently I am confident the sensor is OK.
When coolant temp exceeds a certain value, the ECM activates the fans immediately after engine shutdown. Fans shut down when coolant temp is sufficiently reduced.
I am suspecting the relay is receiving the proper signal from the ECM, but is turning on the high speed fan instead of the low speed fan. That is why I am searching for a relay troublshooting test.
02-04-2009 02:25 PM #5
Update on the Cooling Fans....inside.
I decided to pull the thermostat and check it out in a pot of boiling water. The T-stat begins to open at 84’C. and is fully open at 96’C. (Approx 180’F. to 190’F.).
I also checked the resistance values of the coolant sensors against temperature. There is a chart in the JTIS which provides relative values. Both items checked out OK.
Finally, I noticed the lower radiator hose had been installed backwards and was rubbing on the fan shroud. There was a worn spot which could have compromised the hose. They sometimes collapse, and restrict flow. So I changed the two major radiator hoses.
After performing these three procedures, I refilled with antifreeze and checked for leaks. The hose temps on the plastic coolant manifold (immediately after where the coolant leaves the cylinder heads) registered about 10-15 degrees cooler.
Temps there were measure at 185-200 when the high speed fan would activate. They are now down around the 180 mark measured from the same spot. The low speed fan now activated at about 185; but to my surprise, the high speed fan did not come on at all.
I am guessing the T-stat was intermittently failing… and/or the hoses were constricting flow. Or just the fact that I was pulling parts and moving things around, I could have reestablished a corroded connector, or jiggled a stuck relay.
I did learn the temp sensor and/or ECM must be very sensitive to temp changes, because the measured temps were only 10-20 degrees cooler. either way, the car seems to be running as normal. This is with the original coolant sensor and fan relay installed. Go Figure. Word is once a T-stat has been overheated, they are not to be trusted again. So I am going to replace the T-Stat as a precautionary measure.
... And keep an eye on the temp gauge. gordo
02-04-2009 03:05 PM #6
Interesting, Check possible early warning
I had to replace my 18 month old thermostat this past week. I usually drive short distances, but long enough to kick on the fans. About two weeks on a longer trip my low coolant light came on. I thought that was odd because before this drive, the temp gauge needle had been running just over the halfway mark and everything was operating normally. On this trip it actually was resting just BELOW THE HALFWAY MARK. Very counterintuitive for overheating! I refilled the coolant and all was well for about a week. The temp gauge returned to just over the halfway mark once the car was warmed up. Then last week on a typical short trip, it again was reading just under the halfway mark, and again the coolant light came on. I opened the hood and felt the radiator hoses. The drivers side hose was relatively cool and the passenger side was hot. I squeezed the hoses and small amounts of coolant managed to escape from the radiator end of the drivers side hose as the temp went way up inside the hose. The last time the thermo was replaced, I put spring a screw type hose clamp on just that union because the eariler bad thermostat blew off the hose at that location dumping all the coolant onto the street. I suspect that if the spring clamp was on I could have blown the hose off again. After replacing the thermostat the car functions correctly again AND THE GAUGE AGAIN READS JUST OVER THE HALFWAY MARK AGAIN! If you or other posters have seen the same thing, perhaps this is an accidental discovery of an earliy warning that the thermostat is going intermittent, long before the low coolant light comes on. Or maybe I just have a magic gauge...
02-04-2009 07:30 PM #7
The gauge is not linear... it is center-weighted
On my car, the gauge at 1/4 is moving past 135'F. and reaches the center at about 160'F. It remains on that same centered halfway point until the water temp begins to pass 190 or 200'F. At 230 it was passing the 3/4 quarter mark.
I think the reason is the gauge is controlled, not by the coolant temp sensor, but by the ECM. The coolant temp sensor IS linear in the sense its resistance (in Ohms) is in direct relationship with an increase in water temp. Going from .15 ohms at 212'F to 5.00 ohms near freezing.
The ECM measures the resistance output of the sensor and holds the gauge in the center weighted position as long as the water temp is within a safe, or normal, operating range. If the coolant temp begins to move outside than range, then the ECM instructs the gauge needle to also begin moving in that direction.
The temp gauge is not an idiot light, as some have described it; but it is not a real time read of your engines exact coolant temp either. gordo
02-04-2009 07:58 PM #8
I didn't switch out the ECM which may explain why
an identical stimulus from the two different engines' temp sensors produced the same reading.
The car is a 2000 so perhaps it is a difference between our ECMs, or I have a very unusual temp gauge. Last summer I had an overheating incident due to blown cooling fan fuses. (I think the blown fuses were caused by an auto repair shop shorting the circuit during the repair process, because it has never recurred.)
During that overheating incident I saw the linear increase as you describe. However my gauge's counter-intuitive readings have appeared twice on low coolant alerts where it appeared the engine was running cooler than normal (just under halfway vs. just over) preceding the alerts.
Stated differently, it sounds like your gauge is always dead center regardless of whether the T-stat is beginning to stick, unless the temp starts to exceed the norm.
Perhaps my system is allowing coolant to bleed out somewhere I can't detect during sticking periods, letting air in and lowering the linear reading. However, I think I'll stick with the one screw type clamp to guard against being stranded from the hose popping off the next time my new T-stat starts to stick. If there is a next time where the gauge is just under halfway, I suspect it will again be time to replace the T-stat. Thanks Gordo...
02-05-2009 12:10 AM #9
2 diff sensors will provide different readings
The sensor is a thermistor, a type of resistor with resistance inversely proportional to its temperature. The ECM uses the relative change in values to establish the signal to the gauge.
The JTIS describes the uses of hysteresis in setting the temp readings: A phenomenon sometimes observed in the behavior of a temperature-dependent property of a body; it is said to occur if the behavior of such a property is different when the body is heated through a given temperature range from when it is cooled through the same temperature range. consequently, the reaction of your gauge to heating up is different from the reaction of your gauge to cooling down. gordo
Users Browsing this Thread
There are currently 1 users browsing this thread. (0 members and 1 guests)