Solution To Flickering Fan Speed 2 Or 3 A/C PART 2

Sorry it took solong to reply... Thanks for the info. I managed to get a new replacement Switch. It works in all positions but 3 intermittently now. at first when I connected the new switch, it only worked on position 1 with a couple light tugs of the wires they all started working. I am taking that as a bad sign. :( The Controller board.... Do you know where this is located? If i follow the wires form the Switch, do they go directly to it, or?

Also, was anyone elses old AC switch discolored and Yellowed? Mine was. The new one is super white.

The controller board is inside the climate control unit. When you had it out to change the fan speed switch the controller board was literally in your hand. The edge of it is where the connector with a lot of thin wires plugs in.

If the new switch did not work in position 4 then you must have had a bad connection. Pull the thing out again, tighten up the connectors slightly by squeezing them with pliers, and then put them back on. (Battery unplugged while you do this!) It should take a little oomph to push the connectors onto the spades, if not, they are not tight enough.

My switch was not discolored. I worry about that symptom - it could be an indication the blower is going and drawing too much current. Or not, that would also happen if the connections loosened up and there was arcing at the contacts. I'm not sure how much current the blower is supposed to draw at speed 4 (straight to ground, look in the shop manual), but if you have a hefty multimeter that can measure many amps, then you can test the blower. The multimeter leads must be pretty heavy too, or they can get too hot, melt, or potentially even catch fire.

You want to measure the current between the vacated blower speed 4 spade and ground, ideally with a spade on the lead going to the former, and a clamp of some sort going to a chunk of metal under the dash. I would hesitate to measure the current with the "pointy" leads on most multimeters because the contact might not be good and the resulting arc could be bad for the leads, the spade, or you. Remember that when an ammeter is wired this way the blower will start the instant the car is turned on - it is wired as if fan speed 4 was selected.

Edit: here's a video showing how to do it at the blower on a different car.
http://www.youtube.com/watch?v=pK3dmfBjRI8
 
Last edited:
so glad i fould this. this issue has been driving me nuts. so happy i fould this im about to go outside on this damn hot ass weather and work on the car lol
 
The controller board is inside the climate control unit. When you had it out to change the fan speed switch the controller board was literally in your hand. The edge of it is where the connector with a lot of thin wires plugs in.

If the new switch did not work in position 4 then you must have had a bad connection. Pull the thing out again, tighten up the connectors slightly by squeezing them with pliers, and then put them back on. (Battery unplugged while you do this!) It should take a little oomph to push the connectors onto the spades, if not, they are not tight enough.

My switch was not discolored. I worry about that symptom - it could be an indication the blower is going and drawing too much current. Or not, that would also happen if the connections loosened up and there was arcing at the contacts. I'm not sure how much current the blower is supposed to draw at speed 4 (straight to ground, look in the shop manual), but if you have a hefty multimeter that can measure many amps, then you can test the blower. The multimeter leads must be pretty heavy too, or they can get too hot, melt, or potentially even catch fire.

You want to measure the current between the vacated blower speed 4 spade and ground, ideally with a spade on the lead going to the former, and a clamp of some sort going to a chunk of metal under the dash. I would hesitate to measure the current with the "pointy" leads on most multimeters because the contact might not be good and the resulting arc could be bad for the leads, the spade, or you. Remember that when an ammeter is wired this way the blower will start the instant the car is turned on - it is wired as if fan speed 4 was selected.

Edit: here's a video showing how to do it at the blower on a different car.
http://www.youtube.com/watch?v=pK3dmfBjRI8

WOW! Thank you. Everyone helping out.. You are great. I will see if my father-in-law has what we need to do thses tests. In the meantime, he's an update:

I was doing so research regaurding fuses and the AC (which ended up not being the case). While doing this inspection, I tested the AC switch again..... It is now working on ALL positions; fans speeds and compressor light. Hmmm. I am completely baffled. maybe it is just loose connect as you said. I still want to run the tests though. I dont need my car bursting into flames. I just paid the thing off.
 
well the white plug was all burnt a bit so just took all the wires out of it. cleaned them out a bit put them on the switch made sure they were all tight and good to go and bam all levels again. makes me feel like i have a new car again lmao. also at the same time cleaned the fogged nasty looking headlights too. ah making progress slowly.
 
The controller board is inside the climate control unit. When you had it out to change the fan speed switch the controller board was literally in your hand. The edge of it is where the connector with a lot of thin wires plugs in.

If the new switch did not work in position 4 then you must have had a bad connection. Pull the thing out again, tighten up the connectors slightly by squeezing them with pliers, and then put them back on. (Battery unplugged while you do this!) It should take a little oomph to push the connectors onto the spades, if not, they are not tight enough.

My switch was not discolored. I worry about that symptom - it could be an indication the blower is going and drawing too much current. Or not, that would also happen if the connections loosened up and there was arcing at the contacts. I'm not sure how much current the blower is supposed to draw at speed 4 (straight to ground, look in the shop manual), but if you have a hefty multimeter that can measure many amps, then you can test the blower. The multimeter leads must be pretty heavy too, or they can get too hot, melt, or potentially even catch fire.

You want to measure the current between the vacated blower speed 4 spade and ground, ideally with a spade on the lead going to the former, and a clamp of some sort going to a chunk of metal under the dash. I would hesitate to measure the current with the "pointy" leads on most multimeters because the contact might not be good and the resulting arc could be bad for the leads, the spade, or you. Remember that when an ammeter is wired this way the blower will start the instant the car is turned on - it is wired as if fan speed 4 was selected.

Edit: here's a video showing how to do it at the blower on a different car.
http://www.youtube.com/watch?v=pK3dmfBjRI8

In addition....Hows this for weird? All positions work fine on with the key turned to accessories. But, upon starting the car, the compressor light goes out in position 3 and blows warm. I tested it 5 times. My head is gonna split open with all this frustration.

some one told me to try a very light amount of dielectric grease on the connectors. Good idea or bad?
 
Last edited:
well the white plug was all burnt a bit so just took all the wires out of it. cleaned them out a bit put them on the switch made sure they were all tight and good to go and bam all levels again. makes me feel like i have a new car again lmao. also at the same time cleaned the fogged nasty looking headlights too. ah making progress slowly.

Lucky. I seem to be having more of an issue. ugh.
 
** ANOTHER UPDATE **

ok, so soldering lasted 3 FULL YEARS without a glitch until last week. I cleaned the contacts on the blower fan resistor and no flickering for a day or 2. Then, yesterday, I decided to crimp the connectors on the blower fan resistor and see what happens. The light stayed solid for a few minutes. Looking at a tiny chip on the resistor pack (I now know it's the thermal fuse), I noticed it was incredibly cruddy. Knowing that this could become conductive and cause problems, I decided to gently remove the crud. Well...as gentle as I was, the fuse popped right off! Rather than by-passing the fuse, I went to Mazda this morning, got a new resistor and popped it in. SOLID light on EVERY setting and (I find this strange) the A/C air was MUCH colder in just 2 minutes driving that it ever was after hours of highway driving. So, I believe it is a 2-prong problem. Soldering the connectors to the switch solved the problem 3 years now until the thermal fuse on the blower resistor became flaky or was making intermittent contact. Any comments? Anybody else have this experience? THANKS!
 
Any comments?

A couple.

1. No way in heck changing the resistor should change the temperature of the air coming out of the vents. All the resistor does is control the fan speed, it should have absolutely nothing to do with how well the compressor is working. In position 4 there is no current going through the pack except for the tiny amount from the controller sense line. The only way changing the resistor could change the air temperature would be if the controller was cycling the compressor differently based on the voltage it uses to determine if the blower is on, while still leaving the AC LED lit in both cases, and it would have to be a really terrible design to do that.

2. The flickering is temperature dependent. The higher the ambient temperature the more likely it is to fail. So if 2/3 aren't working in the afternoon (95 degrees) and you change the resistor in the morning (60 degrees) and 2/3 work, it is probably because of the lower temperature, not the new resistor package. The air coming out of the AC will be colder if you run it on a colder day to start with. This is the same reason one can sometimes get speeds 2/3 to work by first running at speed 4 until the car is cold, and then switching to a lower speed. If the lower speed can hold the temperature in the car low it will continue to work. If not, the LED will go off, the fan will start blowing warm air, and the car will heat up rapidly, quickly moving above the upper temperature limit for 2/3 to work.

3. Under the best of conditions with the stock setup there is only around .05 volt difference on the sense line to the controller that determine if 2/3 work or not. It doesn't take much of anything to get on the wrong side of that limit. That's what my alternative fix avoids, it gives the controller an unambiguous on/off signal.
 
** ANOTHER UPDATE **

ok, so soldering lasted 3 FULL YEARS without a glitch until last week. I cleaned the contacts on the blower fan resistor and no flickering for a day or 2. Then, yesterday, I decided to crimp the connectors on the blower fan resistor and see what happens. The light stayed solid for a few minutes. Looking at a tiny chip on the resistor pack (I now know it's the thermal fuse), I noticed it was incredibly cruddy. Knowing that this could become conductive and cause problems, I decided to gently remove the crud. Well...as gentle as I was, the fuse popped right off! Rather than by-passing the fuse, I went to Mazda this morning, got a new resistor and popped it in. SOLID light on EVERY setting and (I find this strange) the A/C air was MUCH colder in just 2 minutes driving that it ever was after hours of highway driving. So, I believe it is a 2-prong problem. Soldering the connectors to the switch solved the problem 3 years now until the thermal fuse on the blower resistor became flaky or was making intermittent contact. Any comments? Anybody else have this experience? THANKS!

Can you post pictures? Is the resistor you changed the one that is right behind the Glove Box in plain site?
 
Pasadena, I completely agree that the resistor should have NOTHING to do with how cold the air gets, but, strangely enough, it does in my case. And it's obvious. Very. I recall going on long trips in the summer and it would take like a half-hour for the air to get to peak cold. Then I'd notice (after maybe an hour or so) that although the A/C light was on solid, the air from the vents was warm. The compressor was not on. In fact, it would never come back on for a few hours. I suspected maybe a hi/lo pressure trigger was tripped at the compressor. Anyway...
You're right about the new resistor NOT solving my problem. As it got warmer out, I started the car and the A/C light was flickering from the start. The car was an oven when I opened the door. So, yes, heat is an issue. If you realise this, then I'm not sure why you're resistant to my soldering solution. You MUST keep a few things in mind. One of them being that heat expands metals. Like the conectors to the fan switch. Even a few degrees warmer and metal will expand a few thou of an inch. Now, remember...people are complaining that their wires are BURNING hot. In fact, some of them have melted wires! So, while I agree with you that connectors like this SHOULD give MANY years of trouble-free connections, introduce extreme heat, and the rules don't apply. Don't forget that these very connectors work on the principle of SQUEEZING. So there's pressure exerted to open the female connector by the flat male connector. Introduce extreme heat, and that heat (along with the pressure) will easily make the female connector open-up. Also, don't forget the following:

- this problem developed over time. This is KEY.
- I've tried several fixes (new switch, re-soldering circuit board, etc) which have not worked
- crimping the connectors tighter worked for several weeks
- drivers experience and have proof of EXTREME heat at the wires which MUST therefore also be transfered to the connectors
- Me soldering the wires gave me trouble-free operation for 3 whole years
- Many others have tried this solution and have verified that it works
- If the heat is so extreme that it can melt wires, it is plausible that the reason my issue has returned after 3 years is because the solder I used may have melted or softened, thus weakening the connections

Pasadena, I don't want this to sound disrespectful in any way (as I have appreciated your work and help over the years), but never over-look the obvious. Have you even TRIED my solution? The solution is given. There's no need to re-invent the wheel. It has been tried, it has been tested. It has been verified by many drivers. It has withstood THREE years. You cannot simply over-look this and say that it just can't be a connection issue. Again, under 'normal' circumstances, I'd agree. The MAIN issue is heat due to excessive voltage. And this is a flaw that Mazda should be fixing at THEIR expense. Christ...MELTING WIRES!! They should fix this regardless of warranty. It's a hazard! As for YOUR solution, it's impressive and wonder if your wires also get extremely hot. I don't see why not, but I'm curious. THAT is the main problem. In the end, if my solution AND your solution BOTH work and BOTH still result in melting wires, we haven't really solved anything. I should drop an email to a USA Mazda dealer I buy parts from on occassion. See what his take on the 'new' board is. I seem to recall people getting climate control boards for a 2003 and STILL have this issue despite a newer revision number. So tired of this stupid proble. Gonna re-solder this week.
 
Pasadena, I don't want this to sound disrespectful in any way (as I have appreciated your work and help over the years), but never over-look the obvious. Have you even TRIED my solution?

Not to the extent of soldering the wires. But I did pull the connectors off, crimped them a bit more to make them tighter, and put them back on. It didn't make any difference. My fan switch doesn't have any signs of overheating.

Assuming there is nothing else wrong with your car, if you put the fan in position 3 on a hot day (so that the AC doesn't light), then run a wire from ground (bare metal on the console will do) to the connector for fan speed 1 at the resistor pack (just poke the wire in, this is the thinnest wire) the AC LED will immediately come on and the air will blow cold. Do NOT do this if the fan is off or on position 1, as the wire will become the primary path to ground and it will probably spark. Anyway, that's the quickest demonstration that the issue is with the sense line going to the control board.
 
In short, is this what you've accomplished with your fix? Clever, clever. I would be very interested in trying this. But to be completely honest, I'm so incredibly swamped right now, that just re-viewing your steps and materials list overwhelmed me. I have not had the time to sit and appreciate the work you've put into it. You really know what you're talking about. Which leads me to my next question, seeing as how it's an issue with the sense line, why was this not an issue when the car was young? Or was it? I only got mine in 2006 and it only started having this issue in 2007 or 08. And does an actual board exist that corrects the issue? As for your test, I'll definitely try it soon! So, I can just leave the fan in position 2, 3, or 4 while plugging-in a wire? Or, I'm assuming just either 2 or 3... THANKS!
 
Fan Speed Fix

Thanks to everyone here for this post. I recently bought a 2002 Mazda Protege and the AC only worked on setting one. I did a little research and stumbled across this thread and was able to fix it myself. Although I didn't solder my connections I did have to completely remove the wiring harness due to corrosion. I found some female connectors with a hard plastic coating that is designed to withstand up to 221 degrees. The plastic is hard and hopefully will keep the contacts from expanding. I squeezed them all onto the AC fan control and man they are a tight fit. If for any reasson this doesn't work I will go back and use the solder method, but I believe this should do it. I'll keep you posted. Specifically would like to thank Weborific!

I tried to attach pics from Photobucket but it didn't work.
 
Last edited:
Which leads me to my next question, seeing as how it's an issue with the sense line, why was this not an issue when the car was young?

I think this is a case where there wasn't enough margin in the original design and as the cars aged the smallish voltage changes that appeared over time due to component aging, on the order of .1 or .2 volts for positions 2 and 3, were enough to push the sensor line over the edge. Also it looks like somewhere along the way they added a diode between the fan speed switch and battery ground, but didn't compensate for the extra .5-.7 volts on the existing controller board. Electrically it looks like that diode exists, but it isn't documented in the circuit diagram in the workshop manual.
 
Sorry it took solong to reply... Thanks for the info. I managed to get a new replacement Switch. It works in all positions but 3 intermittently now. at first when I connected the new switch, it only worked on position 1 with a couple light tugs of the wires they all started working. I am taking that as a bad sign. :( The Controller board.... Do you know where this is located? If i follow the wires form the Switch, do they go directly to it, or?

Also, was anyone elses old AC switch discolored and Yellowed? Mine was. The new one is super white.

Yes old one was yellow, slightly newer one was white. I am pretty sure they are actually different colors though as there can't be that much discoloration. Both of them had the same flickering issue though, so revised or not, it is a problem.
 
Pasedena, why couldn't we just put a relay on the A/C compressor itself. It definitely sounds like the current from the A/C clutch is causing the voltage drop across our spotty A/C control unit connections.

If we just put a relay on the A/C clutch solenoid it would reduce the current draw through the control unit to milliamps and therefore reduce the voltage drop to almost nothing. The current would just get pulled from the nearest 12V hot wire.

No one would have to rip out the control unit to do this fix and there would be a lot less soldering.

Does this sound plausible? I am willing to give it a try, but would need to know what kind of relay to purchase. I guess I could just go measure the current draw of the A/C.
 
Back