Mr.T

Was the fan roaring on high speed at 237 F? EDIT: Just read your last post, that it's running at max above 226 F.

156 F at the radiator sounds low. It's a bit of an art using an IR temp instrument, best I can say is experiment on known temperatures like the outlet hose that should be the same as the ECT sensor (237 in the prior test) with good coolant flow. Maybe coolant flow is actually low, lots of temperature drop across the radiator, but insufficient heat is removed from the engine. A good place to measure temperatures is at the radiator inlet and outlet.

john77

Got them at the dealer.$46 per side.

john77

Here is what I have found at 235 the radiator intake reads 186 and the outlet reads 162 and at 226 intake reads 185 and the outlet reads 153.
This dose not make any sense to me.

Mr.T

Presuming the IR instrument is fairly accurate, it points to the ECT reading being higher than actual. I know the ECT sensor is new, but maybe it's bad, or perhaps there's a problem with the ECT connector, wiring, or even the PCM.

Here's a couple of possible crosschecks:

• It the morning the ECT should read ambient.
• During warmup, feel the hose-radiator inlet for heat when the thermostat opens and note the ECT at that time. The ECT should be close to the thermostat rated opening temperature at that moment. If the ECT reads significantly higher than the thermostat rated opening, the ECT is probably reading too high (just like the IR thermometer is showing).

john77

Sorry, I had to work the last couple of days and could not work on the Jeep. Anyway, ambient temp in the garage was 84 and the ECT shows 96. The thermostat is rated to open at 195 with my hand on the radiator inlet hose I discovered that the ECT reported 215 when the thermostat opened, so I think we found the problem. I am working today but will try and replace the ETC this afternoon and let you know how it goes.

Mr.T

Cool -- Pun intended

john77

Swapped out the ECT and same thing, Jumped onto the highway and hit 237 and by the time I was at the stop light on the off ramp it had dropped down to 221.
Do you have any other suggestions?

Mr.T

A more precise way to verify ECT accuracy would be to put the ECT sensor into a cup of hot water with a thermometer, wiring connected, key to run, then read the ECT with a scan tool and the thermometer in the cup at the same time.

Presuming the ECT sensing is similar to the 3.8L, the causes of reading 10-20 degrees high look like a partial ground on the ECT sense wire side (positive), a partial short between the ECT signal and ground wires, or an internal problem with the PCM itself. The place I'd start is at the ECT connector itself, clean, dry, and look over the wiring.

Side note, a larger ECT error will likely set a code (P0116); On the 3.8L the PCM compares the ECT with the IAT (Intake Air Temp) after a certain amount of non-running time, and sets a code if there's greater than 50F difference between IAT and ECT -- Which is obviously more error than you suspect.

john77

Followed your directions to the letter on the test and the temp was the same on the thermometer as it was on the scan tool. So at least we know that the PCM and the ECT are working correctly.

Mr.T

OK -- That's a big plus. Maybe it's worth looking at the fan in more detail by graphing (or monitoring) ECT and fan speed. At idle, block the front of the radiator with some cardboard and verify the fan appears to increase speed like the scan tool shows.

Shutdown when the fan is maxed out. Be careful about damaging the fins on the rad, shutdown rather than try to remove the cardboard with the fan sucking it against the rad. Double check the shrouds and air deflectors too. Re-start hot, cardboard removed; verify the fan is running high speed and moving lots of hot air.