A/C switch problem, getting worse
Junior Member
Joined: Apr 2012
Posts: 4
Likes: 0
Do you know where the resistor is on the mazda 6?
Junior Member
Joined: Jun 2011
Posts: 2
Likes: 0
Update and Improved fix
Based on some recent "discoveries" on my part I wanted to post back an update because anyone with this issue knows exactly how annoying it can be in the summer months. Now I think I have an additional fix that provides even better results. I will apologize now for the long-winded explanation. If you are in a hurry just skip down to 'SOLUTION STARTS HERE'.
First, let me thank darkMattress for the excellent teardown link to take out the center dash and the recommendation for the metal m10 nut replacement. Both were extremely helpful! Okay, onto the update...
So, in a previous post I suggested that tightening the plastic nut that sits beneath the fan speed **** will eliminate the issue. While tightening the nut had dramatically improved the situation in my case, the outdoor temperatures were still relatively mild (low 80s). However, as the weather began getting hotter as summer approached (say 90+ and parked in the sun all day at work), the A/C began exhibiting the same bad behavior of not turning on until the car's interior cooled back down again. I found that the switch was working great while driving into work after sitting in my garage all night where it was cool but leaving work after the interior heated up, the issue started again. So not completely satisfied I began experimenting by tightening the nut further. What I found was that this made things even worse where the A/C would now disengage at blower speed 4 and only worked at speed 1. This temperature sensitive behavior from the morning to the evening was really interesting to me so I turned my attention towards the switch itself.
This time, I took the switch out completely of the car, and disassembled it. The switch has a very simple construction, and you really should not be afraid to take it apart. But in that simple construction lies the problem. Pictures of what I'm about to describe would be vastly more helpful, but unfortunately I don't have any right now so I will try to describe things as best I can.
The switch is essentially 2 plastic halves that snap together. One half connects to the electrical wiring and the other half connects to the plastic **** for adjusting the blower speed. Inside the switch, on the end of each half is a metal plate. The metal plate on the wiring half is divided into several isolated segments. On the other metal plate are 3 slightly raised metal prongs that make contact with some of the metal segments on the other half. As you turn the switch, the 3 metal prongs make contact with different segments for each fan speed position and this is what causes the blower to change speeds since different electrical paths are being created as you turn. When the switch is assembled, each metal plate should be tightly compressed against the other and that's what completes the electrical circuit.
Now we're getting to the good stuff! As I said earlier the metal plates should be *tightly* compressed to make *good* contact. At first I thought it was just too much tarnish on the metal plates impeding the electrical signal so I got some Brasso and cleaned the metal so it was nice and shiny again. Unfortunately that made no difference; I experimented further. The metal plate on the electrical plug end is fixed in place and cannot move. However, the plate with the metal prongs is pressed against the other using 3 small springs. One spring presses against a ballbearing that sits in a series of 5 shallow grooves in the plastic -- this is what causes the switch to "click" precisely into place as you change blower speeds (this will be important later). The remaining 2 springs simply help push the 3-pronged metal plate into the other metal plate so an electrical circuit is completed.
--- SOLUTION STARTS HERE ----
Still reading? Good. So my theory was that the electrical connection between plates was poor because they were not compressed tightly enough and uniformly enough. As the car's interior heated up, the springs would expand, but not necessarily evenly and this uneven expansion directly let to (at least) 1 prong not making a solid connection since pushing more on one side of the plate would have the effect of lifting up on the other side. If one of the metal prongs (not sure which one exactly) made no contact, then I reasoned the A/C was going out and the light was off. If it was just barely making contact then the A/C light would flicker. As the ambient temperature fluctuated, the length of the springs fluctuated, and the contact of the metal prongs fluctuated. So to fix the problem, you guessed it, I stretched out all 3 springs -- that's it. Not by much mind you (20% or so). Enough to know that good contact was being made, but not so much I couldn't get the switch back together.
The results? The A/C is rocksteady at all blower speeds. I haven't had a single A/C issue in 2 weeks so far of 90-100 degree weather; I actually cannot remember the last summer that happened.
So now for a little postmortem diagnosis:
Why did having the A/C blower speed between positions make the A/C work again? As I stated above, one of the springs pressed against a ball bearing which sat in shallow grooves thereby giving the switch discrete positions (off,1,2,3,4). By having the switch between 2 positions, the ball bearing was between 2 grooves making the spring more compressed and thus pushing more on the metal plate causing the electrical contact to be reestablished.
So why did tightening the nut help at all? Well, I think that as the nut tightened down, it was bending/warping the plastic casing of the switch just enough to help push the springs a tiny bit closer so the metal plates made contact. Suffice it to say, with the springs now stretched, that nut can be completely loose and everything works great.
I hope somebody gets use out of this. As always, I will post back if anything changes. And again, if this helps solve your A/C problem -- or even if it doesn't, please post back here and let others know.
First, let me thank darkMattress for the excellent teardown link to take out the center dash and the recommendation for the metal m10 nut replacement. Both were extremely helpful! Okay, onto the update...
So, in a previous post I suggested that tightening the plastic nut that sits beneath the fan speed **** will eliminate the issue. While tightening the nut had dramatically improved the situation in my case, the outdoor temperatures were still relatively mild (low 80s). However, as the weather began getting hotter as summer approached (say 90+ and parked in the sun all day at work), the A/C began exhibiting the same bad behavior of not turning on until the car's interior cooled back down again. I found that the switch was working great while driving into work after sitting in my garage all night where it was cool but leaving work after the interior heated up, the issue started again. So not completely satisfied I began experimenting by tightening the nut further. What I found was that this made things even worse where the A/C would now disengage at blower speed 4 and only worked at speed 1. This temperature sensitive behavior from the morning to the evening was really interesting to me so I turned my attention towards the switch itself.
This time, I took the switch out completely of the car, and disassembled it. The switch has a very simple construction, and you really should not be afraid to take it apart. But in that simple construction lies the problem. Pictures of what I'm about to describe would be vastly more helpful, but unfortunately I don't have any right now so I will try to describe things as best I can.
The switch is essentially 2 plastic halves that snap together. One half connects to the electrical wiring and the other half connects to the plastic **** for adjusting the blower speed. Inside the switch, on the end of each half is a metal plate. The metal plate on the wiring half is divided into several isolated segments. On the other metal plate are 3 slightly raised metal prongs that make contact with some of the metal segments on the other half. As you turn the switch, the 3 metal prongs make contact with different segments for each fan speed position and this is what causes the blower to change speeds since different electrical paths are being created as you turn. When the switch is assembled, each metal plate should be tightly compressed against the other and that's what completes the electrical circuit.
Now we're getting to the good stuff! As I said earlier the metal plates should be *tightly* compressed to make *good* contact. At first I thought it was just too much tarnish on the metal plates impeding the electrical signal so I got some Brasso and cleaned the metal so it was nice and shiny again. Unfortunately that made no difference; I experimented further. The metal plate on the electrical plug end is fixed in place and cannot move. However, the plate with the metal prongs is pressed against the other using 3 small springs. One spring presses against a ballbearing that sits in a series of 5 shallow grooves in the plastic -- this is what causes the switch to "click" precisely into place as you change blower speeds (this will be important later). The remaining 2 springs simply help push the 3-pronged metal plate into the other metal plate so an electrical circuit is completed.
--- SOLUTION STARTS HERE ----
Still reading? Good. So my theory was that the electrical connection between plates was poor because they were not compressed tightly enough and uniformly enough. As the car's interior heated up, the springs would expand, but not necessarily evenly and this uneven expansion directly let to (at least) 1 prong not making a solid connection since pushing more on one side of the plate would have the effect of lifting up on the other side. If one of the metal prongs (not sure which one exactly) made no contact, then I reasoned the A/C was going out and the light was off. If it was just barely making contact then the A/C light would flicker. As the ambient temperature fluctuated, the length of the springs fluctuated, and the contact of the metal prongs fluctuated. So to fix the problem, you guessed it, I stretched out all 3 springs -- that's it. Not by much mind you (20% or so). Enough to know that good contact was being made, but not so much I couldn't get the switch back together.
The results? The A/C is rocksteady at all blower speeds. I haven't had a single A/C issue in 2 weeks so far of 90-100 degree weather; I actually cannot remember the last summer that happened.
So now for a little postmortem diagnosis:
Why did having the A/C blower speed between positions make the A/C work again? As I stated above, one of the springs pressed against a ball bearing which sat in shallow grooves thereby giving the switch discrete positions (off,1,2,3,4). By having the switch between 2 positions, the ball bearing was between 2 grooves making the spring more compressed and thus pushing more on the metal plate causing the electrical contact to be reestablished.
So why did tightening the nut help at all? Well, I think that as the nut tightened down, it was bending/warping the plastic casing of the switch just enough to help push the springs a tiny bit closer so the metal plates made contact. Suffice it to say, with the springs now stretched, that nut can be completely loose and everything works great.
I hope somebody gets use out of this. As always, I will post back if anything changes. And again, if this helps solve your A/C problem -- or even if it doesn't, please post back here and let others know.
Last edited by the.omega.glory; Jun 20, 2012 at 03:44 AM. Reason: clarification
Junior Member
Joined: Jun 2012
Posts: 1
Likes: 0
bought a new switch last week from Onlinemazdaparts.com
I went from AC working on 1 and (sometimes) 4 to it now working on 1,2,and 4. Progress...
I'm gonna try taking apart the old switch and fiddling with the springs as mentioned above.
I'll post back.
I went from AC working on 1 and (sometimes) 4 to it now working on 1,2,and 4. Progress...
I'm gonna try taking apart the old switch and fiddling with the springs as mentioned above.
I'll post back.
Junior Member
Joined: Jun 2012
Posts: 2
Likes: 0
Crimped wires. Tore switch apart and filed contacts. Installed a bb into each prop spring. Then twisted the connecter and switch before reinstalling it has worked scene my last post. Speeds on all 4 with constant green light from ac.
Junior Member
Joined: Sep 2013
Posts: 13
Likes: 0
It is the switch and nothing else. This is my 4th protege and they all had the same problem. Has nothing to do with A/C charge or resistor. Don't trust me.. Next time make a fist and rap on the side of the console with the side of your fist and watch the light come back on. And have you noticed that sometimes when you hit the gas and it was flickering that the light will come on steady. It switch is faulty and when you hit the gas and it is in the flicker mode the added voltage from the alternator turns the light steady.
I work on F-15 avionics and I have ripped two of them apart on my bench and using power supplies troubleshot and tested them. The contacts become worn and instead of making contact on the entire contact area it only makes partial contact thus not getting the proper voltage causing it to flicker or go out. When you thump the side od the A/C panel you jult the contacts and they will work.
Do yourself and just buy a switch....
I work on F-15 avionics and I have ripped two of them apart on my bench and using power supplies troubleshot and tested them. The contacts become worn and instead of making contact on the entire contact area it only makes partial contact thus not getting the proper voltage causing it to flicker or go out. When you thump the side od the A/C panel you jult the contacts and they will work.
Do yourself and just buy a switch....
When I turn on my speed switch and push in the AC button I don't have any light flicker on 1, 2, or 3 speeds at any speed, ambient temp, or rough road. I'll report back after I replace the resistor.
Junior Member
Joined: Sep 2013
Posts: 13
Likes: 0
Replaced the resistor and I have AC and correct fan speeds on all four settings.