• NAXJA is having its 18th annual March Membership Drive!!!
    Everyone who joins or renews during March will be entered into a drawing!
    More Information - Join/Renew
  • Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

1990 electric fan diode?

cherokee4fun

NAXJA Forum User
Location
Bend, Oregon
So,
On my 1990 with a 4.0, the electric fan won't come on when the motor temp comes up over 220, but it works with the AC. I did install a new temp sensor in the radiator,and the fan comes on when I short the sensor lead.

So, is it possible the relay is bad or the diode assembly is bad. I don't know what the diode is, but there is this orange connector looking thing, which might be the diode. Any idea how to replace that? I assume the relay is good because the fan comes on with the ac.......

Any advice is greatly appreciated,
Thanks,
Bob
 
You can test the diode simply by using an ohm meter, it should pass electricity one direction but not the other, kind of like a one way street.

I doubt that is your problem, if you can jump the temp switch wires and the fan runs. That isn't a temperature sensor in the radiator it is a temperature switch. It also switches at the bottom left, coolest part, of the radiator. While the temperature gauge senor is at the top rear of the head.

Simple solution is to piggyback two wires off of the temperature switch wires (follow the wires up the harness a ways to save some wire) and run the two wires into the cabin and hook them to a simple toggle switch (what i did with my 88). Then you have the option of turning the fan on early (before the motor heats up) or if you forget the temperature switch will eventually turn the fan on. The OEM temperature switch turns the fan on a little late IMO, at a round 217 degrees by my tests, no matter what the book says, I've tested three different thermal switches.

On my 87 (that died a long time ago) I found a temperature switch in a fiat and put that in there, same threads, that switched clised at a lower temperature. Most of the temperature switches have the switching temperature stamped on the face that goes into the radiator, sometimes Fahrenheit sometimes Celsius.
 
If you need to test a diode with an ohm meter, make sure you are in the diode test mode. There will usually be diode schematic symbol on the meter mode switch. In diode test mode the meter provides a bit of voltage (about 1.5) for the test (called bias voltage). With out bias voltage it will look like the diode is an open circuit.

If your meter doesn't have this you will need to use a 1.5v battery with the meter in volt mode to test the diode. It will show voltage with the battery one way but not the other if it's good.

Marc
 
I had that problem and later discovered that the radiator was nearly clogged, and the coolant flow was too slow. As a result the cold side of the radiator did not get hot enough to turn the E-fan switch in the radiator on! The engine got up to 230 F but the radiator exit temp was under 185-190 F which is the cut on temp of that sensor/switch in the radiator. Also, trapped air in the top of the radiator will cause the same problem!

Get all the air our of the radiator on a Renix is PITA!
 
You can test the diode simply by using an ohm meter, it should pass electricity one direction but not the other, kind of like a one way street.

I doubt that is your problem, if you can jump the temp switch wires and the fan runs. That isn't a temperature sensor in the radiator it is a temperature switch. It also switches at the bottom left, coolest part, of the radiator. While the temperature gauge senor is at the top rear of the head.

Simple solution is to piggyback two wires off of the temperature switch wires (follow the wires up the harness a ways to save some wire) and run the two wires into the cabin and hook them to a simple toggle switch (what i did with my 88). Then you have the option of turning the fan on early (before the motor heats up) or if you forget the temperature switch will eventually turn the fan on. The OEM temperature switch turns the fan on a little late IMO, at a round 217 degrees by my tests, no matter what the book says, I've tested three different thermal switches.

On my 87 (that died a long time ago) I found a temperature switch in a fiat and put that in there, same threads, that switched clised at a lower temperature. Most of the temperature switches have the switching temperature stamped on the face that goes into the radiator, sometimes Fahrenheit sometimes Celsius.

8Mud is back! Howdy old buddy!

The last two I bought (US) actually switch on at about 185 to 190 F at the sensor I tested them with IR), but that leads to an engine temp of 190 to 220 F, depending on the radiator, clutch, water pump and engine condition.
 
If you need to test a diode with an ohm meter, make sure you are in the diode test mode. There will usually be diode schematic symbol on the meter mode switch. In diode test mode the meter provides a bit of voltage (about 1.5) for the test (called bias voltage). With out bias voltage it will look like the diode is an open circuit.

If your meter doesn't have this you will need to use a 1.5v battery with the meter in volt mode to test the diode. It will show voltage with the battery one way but not the other if it's good.

Marc

A simple ohms test will show high resistance one way +/-, and low resistance the other way (polarity reversed), -/+, on a good diode. A blown diode will normally be infinite resistance both ways.
 
8Mud is back! Howdy old buddy!

The last two I bought (US) actually switch on at about 185 to 190 F at the sensor I tested them with IR), but that leads to an engine temp of 190 to 220 F, depending on the radiator, clutch, water pump and engine condition.

Good thinking about the clogged radiator.

Back for visit, thought I'd get into the mix a little.

I grant you my method was likely flawed, but my results were consistent. I put the temp switch into a pan of water, hooked up to a ohm meter. and put it on the stove with an electronic temp meter sensor tied to the face with copper wire. I slowly turned up the heat until the switch closed and then checked the temperature. At least one of my switches was stamped 207, don't know if that was a part number or not. Using my method all switched around 217 +/- a degree or two, like I said my test method was likely flawed.

They may have finally gotten smart and changed the specification for that thermo switch, my tests were done a long time ago.
 
I have seen and read of off spec sensors, so I don't doubt your data, I suspect the ones you bought were the wrong spec. I have three aftermarket dash gauge sensors that were off by at least 100 F right out of the bloody boxes! Your electronic temp sensor may have been out of calibration, but then again so could mine, LOL. But I used 2 different IR models to verify mine, and at one time used a thermocouple to verify them IIRC.

And I have one radiator switch that drifted up 15 degrees over 2 years in use. So who knows.
 
Last edited:
I have seen and read of off spec sensors, so I don't doubt your data, I suspect the ones you bought were the wrong spec. I have three aftermarket dash gauge sensors that were off by at least 100 F right out of the bloody boxes! Your electronic temp sensor may have been out of calibration, but then again so could mine, LOL. But I used 2 different IR models to verify mine, and at one time used a thermocouple to verify them IIRC.

And I have one radiator switch that drifted up 15 degrees over 2 years in use. So who knows.

I agree. It has not been determined that the sensor is functioning.
 
Thanks for the information,
I did replace the sensor in the radiator, and if I short out the sensor, the fan comes on. After reading the replies, and reading the factory manual, I will first try replacing the relay, as it sounds like the diode is not a big deal or a part that fails that much,

The couple of things that I did not try, is to short the sensor, and check for voltage at base of the relay, or just change the relay, as I assumed it was good, since the fan kicks on with the AC and by shorting the leads on the sensor. But after looking at the wiring diagram, there should be voltage at the terminal when I short out the sensor, and I have not verified that yet.

Sounds like no one has issues with the diode in these rigs

Thanks,
Bob
 
if I short out the sensor, the fan comes on. After reading the replies, and reading the factory manual, I will first try replacing the relay, as it sounds like the diode is not a big deal or a part that fails that much,


Your test confirmed the relay is good, why would you replace a good relay?

Either the new sensor is not turning on at 185-190 F like it should, or the radiator is clogged up, and not passing coolant fast enough. This results in colder, slower moving coolant reaching the sensor (slower moving coolant volumetric flow gets cooler by the time it reaches the exit!!!!!), and delay the turning on of the e-fan!!!! It is also the reason Jeep moved the e-fan temp control sensor to the thermostat housing on the late HO models!

My money is on a clogging radiator!!!!!
 
Last edited:
Depends on how bad the radiator is, and what kind it is. If it is 4-5 years old or older, and is a cheap plastic tank, aluminum radiator, you can buy one of them delivered to your door for $80-$100, so they are not worth trying to repair. If it is a solid brass tank, brass core HD 2 row, take it to a radiator shop and have them rod it out for about $70 tops. A new 2 or 3 row CSF brass radiator from www.radiatorbarn.com is about $169, last time I bought one.

Try disconnecting the upper and lower hoses, and use a garden hose and rag seal, to push water from outlet (bottom) to the top (inlet) thus backflushing it, if junk comes out, run it tell it flows clean, then reverse the flow, if pressure backs up on the top side, and the flow out the bottom is slow, it is still scaled up, and needs replacing, rodding out, or possible descaling with a cleaning solution (from the parts store, heavy duty radiator flush, follow the instructions on the bottle). But if the radiator is old, and a cheap plastic/aluminum one, just replace it, don't try to use HD radiator flush cleaner, as it it is likely to leak after wards.
 
Cool,
Thanks for the advice.
I assume this is the original radiator, as everything looks original, so it most likely should just be replaced. Do many folks have much luck flushing it out? I would guess I could try flushing it with out removing it from the rig......

Is it possible that there might be a air bubble in the radiator?

Thanks,
Bob
 
One other thing,
I did take the old sensor and did the stove test, and it did switch on and off, so I wasted $20 on a sensor....

This rig has not be used very often over the past 5 months, as I have a couple other Jeeps and I use this as a ski Jeep. Last winter and summer, it always ran cooler than my other Jeeps with 4.o, but would heat up when I would stop a fast food place after skiing, and I noticed that the fan did not kick on, so I just assumed it was the $20 radiator sensor......
 
Huge air bubble on a 1990, close Renix system is a huge possibility!!!!!

I had a new 3 row radiator in mine, and the first start up the top 1/2 of the drivers side of the radiator was room temp while the engine was 220 F!!!! About 1/2 way down, top to bottom, it went from 120 F to 180 F over a 1" span!!!!! The top half of the radiator was full of air, and that screwy P trap hose on the radiator inlet does not bleed the air properly, especially with a closed thermostat. I gave up and installed an inline filler neck on mine near where the hose connects to the radiator. Made burping & refilling a piece of cake!

So yes, check the drivers side temp top side and top half to see if an air bubble is blocking flow! It will also keep the Temp switch from getting hot enough to turn on if the air bubble is large enough.


One other thing,
I did take the old sensor and did the stove test, and it did switch on and off, so I wasted $20 on a sensor....

This rig has not be used very often over the past 5 months, as I have a couple other Jeeps and I use this as a ski Jeep. Last winter and summer, it always ran cooler than my other Jeeps with 4.o, but would heat up when I would stop a fast food place after skiing, and I noticed that the fan did not kick on, so I just assumed it was the $20 radiator sensor......
 
If it is the original radiator, it is 21 years old, time for new one probably, but if you have the time, you could take it to a radiator shop and have them check it out (overnight, 48 hr deal usually). Might just need a deep cleaning in the vat, and a flush. If they pull the tanks off and rod out the core, they may solder off and close any bad tubes, reducing cooling capacity.

As cruiser said, they do not flush well. By the time they are acting up, usually many tubes are blinded off with scale, no flow, and therefore do not flush. Taking the tanks off, and giving the tubes an enema (LOL) with a flat rod is the only way to clean them once they get that bad.

But since it is a renix, check for a half filled radiator with a massive trapped air bubble first for sure!!!!
 
They do not flush well.

But they do LEAK well, when they blow!:cheers:

My buddy pulled up today in his 95 grand C, and the heater hose split wide open just as he pulled in the driveway, puked about 2.5 gallons all over the place (water pump worked too well, LOL), hose looked like a watermelon split in half, open length wise about 6" along the length, like some one had cut it with a razor blade! It was weird as hell.
 
Back
Top