Greetings from Croatia 626 GE 2.5 V6

[JokeCro]

Guys, can someone tell me what exactly is the radiator fan switch/coolant fan switch and where is it located? I can't find a manual for it. I installed a new ECT coolant sensor, a cap (since the old one started leaking), a new relay, but it still turns on the fans late.

[Kampfkuchen]

Did you try my hint with the diagnostics mode? There is no thermostate switch for the fan. It is controlled by the ECU. I think the low-speed fan doesn't work. Maybe you have to replace the fan. If you have a multimeter, check the fan for open circuits.
Are the fuses fine?

I will send you the service manual.

[JokeCro]

Did you try my hint with the diagnostics mode? There is no thermostate switch for the fan. It is controlled by the ECU. I think the low-speed fan doesn't work. Maybe you have to replace the fan. If you have a multimeter, check the fan for open circuits.
Are the fuses fine?

I will send you the service manual.

I've been away for a while.

I think the fans are working, because when I came in from town the other day, I left the car running for a little while longer, the needle started to climb above half, and after a few seconds the fan turned on, but only one at maximum speed. Then I turned on the air conditioning, then the other fan started working.
The fuses seem okay, I checked.

[commander_keen]

Yeah, correct, I already ruled that out:

The second fan is for the air conditioning condenser.

The motor cooling fan in the V6 has two fan speeds (low-speed and high-speed),
while the 4-cylinder cooling fan has only one speed.

So while there is no "low speed fan", there is indeed a low speed mode for the single engine cooling fan.

And I would agree that the low speed mode seems to have an issue, because I think the temperature needle should not move while only low speed temperatures are reached.

So you must check the relais (there are three of them, where only one is for low speed mode) and maybe the temperature sensor.
Target behaviour:
When the engine coolant reaches 97°C, the engine ECU must turn on only one relais. (fan in low speed mode)
Only when the coolant reaches 108°C, the engine ECU switches on relais no.2 and no.3. (fan in high speed mode)

Therefor you need to check if only one relais is being turned on at first. Because if two relais are turned on directly after fan standing still, then you are indeed missing the low speed mode.

If you are able to remove the temperature sensor, you can test it in boiling water. Kampfkuchens manual should have a table with temperature vs. Ohms.
But be aware that it is unlikely that the sensor failed completely. Because the engineers at Mazda were clever, as they provided a failsafe: If the engine coolant sensor fails, the fan is put to permanent ON. (obviously cannot be your case)

So the sensor will either most likely be okay or report wrong values to the ECU.
=> I would recommend to start testing the relais, as it's the easiest test. (and more likely root cause?)

If you never experienced two different fan speeds now although the temperature needle rises, either the ECU has a problem and doesn't switch on relais no.1 or it tries and the relais is defective. (both cases should be testable with engine hot by a simple multimeter I think?)
You should also be able to connect the low speed pins of the fan to the battery + and - directly with some wire, to see if low speed mode of the fan itself works.

[Mazda RX-7 SA]

If the engine coolant sensor fails, the fan is put to permanent ON. (obviously cannot be your case)

So the sensor will either most likely be okay or report wrong values to the ECU.
=> I would recommend to start testing the relais, as it's the easiest test. (and more likely root cause?)

I expect that there are 2 different Sensors: one for the dash (mostly one pin connector) and a second for the ECU (mostly two pin connector).
Why I do expect this? Because in the Xedos with same V6 it is in that way. I think in the GE with 4-Zyl. it is the same.

Just as a side info.

[Kampfkuchen]

No you are right, I think we discussed this a few posts ago. The V6 has two separate sensors and the one from the dash is not the same the ECU sees. So my first thought was a drift in one sensor, but this is very unlikely to come overnight.

I'm 99% sure, there is no working low-speed as the high-speed comes at a higher temperature. Why I suggest checking the low speed from the fan.

[JokeCro]

There are two more sensors in that location, one with one wire (for the needle on the dashboard) and another with two wires.

What is that second sensor actually for?




I put a new one on the third one.

[JokeCro]

Yeah, correct, I already ruled that out:

The second fan is for the air conditioning condenser.

The motor cooling fan in the V6 has two fan speeds (low-speed and high-speed),
while the 4-cylinder cooling fan has only one speed.

So while there is no "low speed fan", there is indeed a low speed mode for the single engine cooling fan.

And I would agree that the low speed mode seems to have an issue, because I think the temperature needle should not move while only low speed temperatures are reached.

So you must check the relais (there are three of them, where only one is for low speed mode) and maybe the temperature sensor.
Target behaviour:
When the engine coolant reaches 97°C, the engine ECU must turn on only one relais. (fan in low speed mode)
Only when the coolant reaches 108°C, the engine ECU switches on relais no.2 and no.3. (fan in high speed mode)

Therefor you need to check if only one relais is being turned on at first. Because if two relais are turned on directly after fan standing still, then you are indeed missing the low speed mode.

If you are able to remove the temperature sensor, you can test it in boiling water. Kampfkuchens manual should have a table with temperature vs. Ohms.
But be aware that it is unlikely that the sensor failed completely. Because the engineers at Mazda were clever, as they provided a failsafe: If the engine coolant sensor fails, the fan is put to permanent ON. (obviously cannot be your case)

So the sensor will either most likely be okay or report wrong values to the ECU.
=> I would recommend to start testing the relais, as it's the easiest test. (and more likely root cause?)

If you never experienced two different fan speeds now although the temperature needle rises, either the ECU has a problem and doesn't switch on relais no.1 or it tries and the relais is defective. (both cases should be testable with engine hot by a simple multimeter I think?)
You should also be able to connect the low speed pins of the fan to the battery + and - directly with some wire, to see if low speed mode of the fan itself works.

New ECT coolant sensor installed.



I also replaced the first relay that was cracked with a second new relay.

[JokeCro]

If the engine coolant sensor fails, the fan is put to permanent ON. (obviously cannot be your case)

So the sensor will either most likely be okay or report wrong values to the ECU.
=> I would recommend to start testing the relais, as it's the easiest test. (and more likely root cause?)

I expect that there are 2 different Sensors: one for the dash (mostly one pin connector) and a second for the ECU (mostly two pin connector).
Why I do expect this? Because in the Xedos with same V6 it is in that way. I think in the GE with 4-Zyl. it is the same.

Just as a side info.

Yes, there are actually 3 different sensors in that location, one with a single wire (probably for the needle on the dashboard) and two sensors on each side near the coolant cap.
I replaced one, the green ECT one.

[commander_keen]

Interesting, the older GEs seem to have an additional cooling fan sensor, which was later replaced by full control via ECU and only one sensor.
For these older models additionally to that failsafe:

If the engine coolant sensor fails, the fan is put to permanent ON. (obviously cannot be your case)

There is another one:
If the cooling fan sensor fails, the fan will be turned on if the engine coolant sensor (ECT) reports 60°C or higher.

I'm unsure what it was actually good for, if it was removed later anyway. (maybe the two different fan speeds were controlled by the two sensors seperately in the early days?)

Have you solved your low-speed / high-speed issue with the sensor and/or relais now?