Here in the 21st century, I'm sure you as an engineering type appreciate that even a skilled amateur grease monkey can't do much on a car.

I'm thinking adealercould hook your car up toa diagnostic, and see if it is in fact the Limiter switch which is turning your compressor off (or keeping it off).

Its possible the Limiter switch is faulty. Its alsopossible yourAC system is overcharged - too often, in response to a customer comment about bad AC, the tech will add more refrigerant regardless of whether the system needs it - which can make matters worse. Too muchrefrigerant in therewould cause the pressure to build up to a bad level quickly, especially if its hot and the driver is stomping the gas pedal. They could try letting some refrigerant out of the system - itwill notharm the compressor to be a little under.

Another possibility isthe refrigerant/oil mix in your AC system is wrong (there isa little bit of synthetic oil in with the refrigerant gas,purpose of which is tokeep the seals and valves within the system lubricated - if they put too much oil in, it could cause a pressure problem).The solution would be to purge the system of the currentrefrigerant/oil, and refill it paying special attention to getting the mix correct.

Finally, as others have noted,more than a couple of M3 owners find the AC system to be a little weak. I have just bought mine, it is not yet summer here, so I don't have first hand experience with mine. You may unfortunatelyjust have to deal with it.

I hope you come back and tell us how things turn out for you.

 

I went to an A/C specialist yesterday. The problem is that he doesn't have the diagnostic adapter for the computer..he should get it in 1-2 weeks.
We went for a drive (he's driving), part of it a short up-hill climb. He put a thermometer on the air exit. We saw the temps cycles between 4.5 to 6 degress celcius, when pushing the gas pedal in the up-hill climb we felt the compressor cuts off, temp went up to 8.5c and it kicked back in. Since it was a not-so-hot day (~21c) - and the up-hill was too slow and short we didn't really felt hot in the car.
Never the less - I described to him what happens on a hot day, that I'm getting hot and sweaty if I push the gas too much - he said that this should never happen. Even in cars that has open throttle cut-off, the compressor should kick back in after few seconds, it should never make the driver sweat under no condition.
When I told him that the Mazda diagnostics guy previously checked this and just said I'm driving too fast, he was totally shocked.
He checked the refrigerant volume (390g, spec is 450), pressures, the fan, didn't find anything wrong.
We are going to do the check again on a real hot day, and hopefully with the computer connected.
He suggested it might be faulty compressor temp sensor or refrigerant pressure system, or just bad computer program - which in that case is a problem - we'll have to find a smart way to fool the computer, since Mazda are not going to give any solution for that.
Anyway - at least now I have a proffesional guy that acknowledges that there's problem, I'm sure that we'll find the cause. I promise to keep updating. Thanks everyone for the help so far.

 

I'm happy to say that my problem has been solved.
I decided to make a bypass on the A/C relay control - so I can force the compressor to work when I want it to.
I added a safety feature - the bypass works only if Hi/Lo pressure unit is connected (normal pressure).
The implementation is quite simple - all you need is a 5 poles relay (3 contacts and 2 for control), a 2 contact switch to install in the dash.

Normally I would drive without bypassing the computer, in hot days, when I'm pushing the car - I will turn it on for a limited time. The problem is I might cause the evap to freeze, is it possible under these conditions ?

[IMG]local://upfiles/5101/A360D2F99BE2415686F047BA23BA33FB.jpg[/IMG]

[IMG]local://upfiles/5101/59AB0112DF60453BBAD12C7118559387.gif[/IMG]

[IMG]local://upfiles/5101/4502B5D99CA6486083B42B73B32CC6BC.gif[/IMG]

[IMG]local://upfiles/5101/54107E6380B14C8DB106C58CD5994345.gif[/IMG]

[IMG]local://upfiles/5101/EA24E6A4E5A24D87A279446BE41F38F4.gif[/IMG]

 

the relay...

[IMG]local://upfiles/5101/A7F90163EE294B3799F940F48DB5919E.gif[/IMG]

 

Yes, quite possible. Especially if you forget the bypass circuit is on.
A guy on the Mazda3 forums, screename billm found a similar fix a while back only he has it controled by an electronic timer instead of a manual switch. I guess his method pushes the the A/C to the max, then shuts it down for a limited time so it won't freeze up.
He's created a circuit and sold a few, and it seems like all those that have tried it have had no other problems... including icspots.

 

Yes, I read all that post, but it's a different solution to a different problem.
He (and all the others with manual control A/C) had a cycling problem - the compressor cycled on and off too much (too much off time) with no relevance to the ambient conditions.
The main issue was cooling the car in the first minutes - when you want 100% time compressor - so he made a circuit to set the compressor on for ~10 minutes, and go back to normal cycling.

My problem is different - It's possible that I take a 2-3 hours drive in a hot day and most of the time the compressor will be cut-off since the throttle is open. So I need to bypass the computer during all that time. I'll just keep in mind to turn it off from time to time...
Is there any warning sign that the evap is about to freeze ?

 

A-ha! Well, in my own defense, it has been a while since I read through that whole thing.
And in that case... Nice Fix!!
About your question: Hmmm. Again it's been a while, and I'm no expert when it comes to auto a/c, but iirc the system simply stops cooling because it's the venturi (like GM cars) or expansion valve (which ever system we have on these things?) is frozen shut, or over, and won't allow coolant to flow through the evaporator any longer. In effect, the system shuts down though the compressor may still run/cycle. With modern electronic control over everything these days, I'm sure they've come up w/a way to shut down the compressor under those circumstances, but that's just a guess on my part.
As far as detecting it ahead of time, I'm not sure if there's a way to tellbefore it's too late or not?
There must be a modern (lol) tech on here or one of the other forums that can answer that one for you.

 

I know how the system in the mazda 3 works - I've been reading all the schematics in the last 2 weeks...
There's a thermistor located on the evaporator, the A/C control unit sense the temperature and shuts down the compressor under a certain limit.
This is how it normally works - but when I bypass the computer - it can't shut down the compressor...
so it might freeze.

 

No offense intended sir. I was not trying to patronize you, and apologize ifI came off that way.
Clearly you know more about the electonical aspect of this system than I do. From your question it just seemed to me that the physical/mechanical aspects of it may have been a bit illusive leading to your question in the first place. I was only trying to give the the general benefit of my own experience w/auto a/c... though by my own admission, is somewhat outdated.
I wonder if instead the thermister might be replaced or modified (a resister in series for example?) to allow the system to stay on longer?
As far as detecting a freeze up before it happens, as I'd said before, short of the system stopping after a freeze up, I am not sure. Perhaps if some one else w/more, updated experience will chime in they could answer that one better than I?

 

Non taken, of course.
Alterning the thermistor will not help me, since the problem I have is with the computer (main computer, not AC controller). The problem is it cuts-off the compressor when I press the gas pedal. The only way to change that is by changing the program in the computer. Only Mazda can do that - and they won't, since they claim it's suppose to work this way..go figure.
Thanks for your help anyway, I guess I'll just see what happens