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.
With the recent thread revival, now seemed like a good time to share this information. This will be surprising to some and potentially helpful to others, but after weeks of troubleshooting and conferring with Dale, I have updated information on how the cooling fan logic actually works.
For context, I completed the FC thermoswitch mod and the fans still didn't kick on until 105C with the ECU input. As a result, I went on a deep dive; I pinned out the harness, checked grounds, checked power to each relay pin with the fans on and off, bench tested the thermoswitch, and spent countless hours reading threads, wiring diagrams, and consulting the experts. Anyway, here’s what I’ve got!
Conventional wisdom states the cooling fan inputs are all completely independent. Meaning the ECU, AC, or thermoswitch can each turn the fans on low.
In stock configuration, the ECU will be the first input to turn the fans on low at 105C. Next, the thermoswitch will close at 108C to turn the fans on medium. With that understanding, many have opted to install the FC thermoswitch to trigger the fans at 97C instead of 105C.
While good in theory, installing this component in stock configuration will not alter the fan trigger temperature at all. The wiring diagram depicts the ECU and AC inputs as independent, while the thermoswitch is wired to only work as a second (or third) input.
In other words, relays 1, 2, and 4 have constant large current power any time the ignition switch is on (or all the time for the recall harness). Relay 3, on the other hand, only has the large current supply when the fans are running. So when the small current going through relay 3 is grounded by the thermoswitch at 97C (closing the relay), the fans don’t turn on because there is no large current present to activate them.
I pinned out the wires and confirmed they agreed with the diagram; constant power only arrived to relay 3 once the fans were running. As a final test, I hit the AC input to actuate the fans, then grounded relay 3 and the fans went to medium. So as it turns out, the wiring diagram was correct (albeit confusing), and the thermoswitch only works as an input when the fans are already on.
To further muddy the water, the thermoswitch mod would work correctly for someone with an alternate fan input lower than 97C (aftermarket management, fan controller, etc.). But for a stock car where the fans don't come on until 105C, the FC thermoswitch won't change the trigger temp at all. It will however, go to medium if the fans are running at/above 97C from another input.
I feel this isn't apparent to most who do the mod because, a) they already have a lower trigger temp, b) they don't have an aftermarket temp gauge or, c) they just assume the fans come on sooner as intended.
There is one trick in stock configuration to get the fans on sooner. The ELD will generate a fan input depending on electrical load; for example, any time the headlights or rear defroster are on, the fans will engage at 100C instead of 105C. So when I’m driving around in traffic and see the temps creeping up, I’ll hit the rear defroster button to keep the temps below 100C. It’s a bit jenky, but certainly an effective solution until the PFC is installed and I can modify the ECU input. Hope this helps!
The ground wire with the eyelet on the far left side of the picture is a ground. The one on the right side with the black single connector is the test connector. It should not be hooked up in normal use.
is there a way to get fans to come on high when ac is on? I'm running a vmount and the gap between the two heat changers is pretty large. I think the higher fan speed can high keep my high side ac pressures lower.
Not easily. There may be a way to wire it to do that but it would be tricky. You'd also be over-cooling the engine at idle most likely and putting the engine below thermostat opening temps.
May want to look instead on putting an aux fan on the AC condenser.
Not easily. There may be a way to wire it to do that but it would be tricky. You'd also be over-cooling the engine at idle most likely and putting the engine below thermostat opening temps.
May want to look instead on putting an aux fan on the AC condenser.
Dale
I thought about using a aux fan then I saw how much off the open room it would take between condensor and intercooler.
I was thinking using the ground wire that turns on relay 1? (ac relay) to also close the relay that the fc thermoswitch would close if it hit the extra hot temp.
Great information..not sure if ill go the PFC route, ill probably use the AC or defrost. I just got my FD and that was one of the first things I noticed, felt like they came on just too late.
Damn, I replaced my FD thermo switch earlier this year since the tip was falling apart. Now I have to buy the 89-91 FC turbo thermo switch. Since I have a 12/95 car I can not remove the relay, at least I can use the lower temp setting.
Damn, I replaced my FD thermo switch earlier this year since the tip was falling apart. Now I have to buy the 89-91 FC turbo thermo switch. Since I have a 12/95 car I can not remove the relay, at least I can use the lower temp setting.
Thanks Dale!
Wait, I also have a 12/95 car. Is something different about them?
The fan recall harness is built into the ECU harness on the '95 cars and also on cars that have had the harness replaced. This is the harness pictured above in post 127.
Wait, I also have a 12/95 car. Is something different about them?
I do not want to fudge up this thread, but yes slight changes to....
wire diagrams
underbody braces
we buy stuff for the 93-95 years but somewhere late 95 our cars changed a tiny bit. 12/95 cars are Version 3 or 4 cars, you have to look at the VIN (post 34 in link)
Hi, new FD owner dealing with an overheating issue. First off, thanks Dale and every else contributing, for all of this great info. I've read this thread and several other fan threads while troubleshooting to identify my problem, but I have still have some questions. Some background on my car and the overheating problem:
- 94 stock ecu, DP / exhaust, otherwise mostly stock
- aftermarket water temp gauge hooked up at thermostat housing
- new OEM wiring harness at some point (but I don't believe I have the Fan MOD from the recall, I had to turn the key to ON to perform the relay tests)
- normal temps on highway cruising (~190F steady no spiking)
- when idling, fans will not come at all unless A/C is activated. (Side note, triggering electrical load via headlights /rear defrost has no effect on fan speed under this condition)
- Idle from 180F and watched it climb past 220F, then past 230F then I turned the A/C to activate the fans out of precaution
After some reading I tested the fan relays by using the grounding method. I was able confirm that the individual relays are working and was able to replicate all three fan speeds.
My question - Is it safe to assume my issue is a bad water temp sensor (that feeds the ECU) ? Or also a bad thermoswitch / connection? Next logical step to replace one or both of these sensors? I'd opt for an FC thermoswitch while I'm in there.
Hi, new FD owner dealing with an overheating issue. First off, thanks Dale and every else contributing, for all of this great info. I've read this thread and several other fan threads while troubleshooting to identify my problem, but I have still have some questions. Some background on my car and the overheating problem:
- 94 stock ecu, DP / exhaust, otherwise mostly stock
- aftermarket water temp gauge hooked up at thermostat housing
- new OEM wiring harness at some point (but I don't believe I have the Fan MOD from the recall, I had to turn the key to ON to perform the relay tests)
- normal temps on highway cruising (~190F steady no spiking)
- when idling, fans will not come at all unless A/C is activated. (Side note, triggering electrical load via headlights /rear defrost has no effect on fan speed under this condition)
- Idle from 180F and watched it climb past 220F, then past 230F then I turned the A/C to activate the fans out of precaution
After some reading I tested the fan relays by using the grounding method. I was able confirm that the individual relays are working and was able to replicate all three fan speeds.
My question - Is it safe to assume my issue is a bad water temp sensor (that feeds the ECU) ? Or also a bad thermoswitch / connection? Next logical step to replace one or both of these sensors? I'd opt for an FC thermoswitch while I'm in there.
Thanks
Looks like you're on the right track, might as well go for the FC thermoswitch upgrade while you're there.
With the recent thread revival, now seemed like a good time to share this information. This will be surprising to some and potentially helpful to others, but after weeks of troubleshooting and conferring with Dale, I have updated information on how the cooling fan logic actually works.
For context, I completed the FC thermoswitch mod and the fans still didn't kick on until 105C with the ECU input. As a result, I went on a deep dive; I pinned out the harness, checked grounds, checked power to each relay pin with the fans on and off, bench tested the thermoswitch, and spent countless hours reading threads, wiring diagrams, and consulting the experts. Anyway, here’s what I’ve got!
Conventional wisdom states the cooling fan inputs are all completely independent. Meaning the ECU, AC, or thermoswitch can each turn the fans on low.
In stock configuration, the ECU will be the first input to turn the fans on low at 105C. Next, the thermoswitch will close at 108C to turn the fans on medium. With that understanding, many have opted to install the FC thermoswitch to trigger the fans at 97C instead of 105C.
While good in theory, installing this component in stock configuration will not alter the fan trigger temperature at all. The wiring diagram depicts the ECU and AC inputs as independent, while the thermoswitch is wired to only work as a second (or third) input.
In other words, relays 1, 2, and 4 have constant large current power any time the ignition switch is on (or all the time for the recall harness). Relay 3, on the other hand, only has the large current supply when the fans are running. So when the small current going through relay 3 is grounded by the thermoswitch at 97C (closing the relay), the fans don’t turn on because there is no large current present to activate them.
I pinned out the wires and confirmed they agreed with the diagram; constant power only arrived to relay 3 once the fans were running. As a final test, I hit the AC input to actuate the fans, then grounded relay 3 and the fans went to medium. So as it turns out, the wiring diagram was correct (albeit confusing), and the thermoswitch only works as an input when the fans are already on.
To further muddy the water, the thermoswitch mod would work correctly for someone with an alternate fan input lower than 97C (aftermarket management, fan controller, etc.). But for a stock car where the fans don't come on until 105C, the FC thermoswitch won't change the trigger temp at all. It will however, go to medium if the fans are running at/above 97C from another input.
I feel this isn't apparent to most who do the mod because, a) they already have a lower trigger temp, b) they don't have an aftermarket temp gauge or, c) they just assume the fans come on sooner as intended.
There is one trick in stock configuration to get the fans on sooner. The ELD will generate a fan input depending on electrical load; for example, any time the headlights or rear defroster are on, the fans will engage at 100C instead of 105C. So when I’m driving around in traffic and see the temps creeping up, I’ll hit the rear defroster button to keep the temps below 100C. It’s a bit jenky, but certainly an effective solution until the PFC is installed and I can modify the ECU input. Hope this helps!
A little belated, but thanks to @Dale Clark for the thread and info and @6speedrx for this extra information. I have the FC thermoswitch in my 1995, with OEM ECU at this time. So if I understand it correctly, the fans are coming on at 105°C with the ECU control, but because the 97°C thermoswitch is present they are coming on at medium speed rather than low...is this correct?
Also, it is great to learn about the ELD activating the fans when the temp is at 100°C. I had noticed this effect before, randomly, but I always assumed it was just another FD grounding idiosyncrasy. It's nice to learn it is actually by design!
Just a quick update on my progress - managed to change the ECU water temp sensor (NF02-18-840) and FC thermoswitch (PN41-18-840). I could have reached the ECU temp sensor on top without removing the alternator, but the thermoswitch is buried pretty deep. Tested the fan functions by idling up to temperature while keeping an eye on my temp gauge and ear open for the fans (had my hood propped b/c they can be hard to hear).
- everything turned off (lights, A/C), fans now come on at 220F as they should
- parking lights on, fans come on at 210F as they should
- turning on A/C for either of these conditions kicked the speed up as expected
It was a little difficult to tell which speeds the fans were going at (whether I was in 1-2 or 2-3) as I was by myself. But since all my relays were working, and the fans are behaving as expected in multiple speeds, Im confident my issue is fixed. Thanks for everyone's contribution in this thread!
I have some hopefully useful information to add to this post.
Later JDM models are completely different and you can pretty much disregard everything from the OP.
Later models don't use the thermoswitch for triggering - everything is performed by the ECU by detected coolant temperature and so the FC thermoswitch modification will do nothing.
From version 5 onwards (1999-2000, FD3S-5xxxxx and FD3S-6xxxxx and up), Mazda lowered the 105'C threshold to 101'C. Furthermore, they added a new 97'C threshold for low fan speed operation when electrical accessories are operating (fan, rear defroster, park light, headlight). On 280ps models (Type R, Type RS, Bathurst R, Type RZ, Spirit R Type A & B), the fans will operate at full duty when the coolant temp is higher than 101'C for a 35 minute period when 4 high-load or high speed events are detected within a 90 second period. They call this "circuit mode" or track mode. Presumably this is to address the driver being in a more spirited mood and to help address heat soak.
We should take from this that Mazda recognised the previous fan programming was not fit for purpose. This accompanied a range of other cooling system design changes that came in for 99 on models especially on the 280ps versions.
Full table below. I will work on an English version....
As promised above, here is my English translation for fan operating conditions for version 5 and 6 RX-7s:
The manual includes an explanatory note to say the relay no. 1 is no longer triggered by the A/C switch, but is instead operated by a circuit driven by the PCM (1 H terminal). This enabled the PCM to trigger relay 1 and switch the fan to high speed for high load conditions even when the A/C is not operating. As explained in my earlier post above, relay no. 3 is also triggered by a dedicated PCM circuit (4 L terminal) - the thermoswitch is redundant.
The PCM controls these relays according to the above table based on input from the A/C circuit, water temperature sensor, TPS, crank triggers, electrical load inputs and MAP sensor.
Last edited by KYPREO; 01-14-24 at 10:59 PM.
Reason: Additional notes on relay triggers
04-18-19, 01:35 AM
