Sad but true, five hours later the car had the same problem again. They checked again and it seems a problem with either the motherboard or the keys that are no more recognized. They are checking it again.

 

Update of the situation.
The Mazda center is unable to reproduce the problem: the car is starting every time they try. Therefore, they are going to substitute nothing at all, since they cannot see the problem. They told me it could have been a remote-controlled gate or some other radio-frequency device that had caused the immobilizer to panic. It could be, since the first time the car blocked it was parked near a remote controlled gate, but it was not the first time I parked it in there. The second time it was even in another city, so I don't give this radio-frequency theory too much credit.

The Mazda technician told me that the mechanical key should always unlock the immobilizer, and I don't believe this is true since (i) the user manual refers to the fact that the immobilizer could remain locked and that (ii) it seems to me that the manual key is a 100% mechanical one, so I would be very scared of a thief in such condition...

I've suggested them to inspect the key-noob contacts, and in the case to change the controller of the key, but they refuse to do it since they don't find the problem. I have also to mention that the warranty on the car is still active for a month, and I'm scared if they don't find the problem now they will discover once the warranty has expired....

Finally, both the times the car was locked it was the end of the week, so the car remained off for a couple of days and both the times it started again by itself. So the pattern seems to be that the immobilizer get panicing and after a couple of days it recovers by itself.

Any thought? Suggestion on how to deal with Mazda center?

 

First off there are no updates for the security system. All they may have done is tried reprogramming the remotes. Battery spike???? Seriously? Sounds like a backyard mazda shop! The hard key that is removable from the card key is also programmed and has a microchip in it so it will only start the car if its programmed. I've fixed a few of the mazdas that have had the same issues your having. It needs the immobilizer module replaced. The module has a fault making it forget the keys for a period but eventually remember them again. Your warranty may b up in a month but that doesn't matter u have 1 year (in the USA) from the first time the condition was looked at by the dealership. They have to continue warranting a pre-existing condition.

 

This is what I was thinking too. At least, having a new immobilizer module will prevent us to search for a problem in the module itself. I don't understand their no solution approach: "here's the car, it has no problem, bring it here when there's again the problem".

 

After one more week at the local mazda center, this is the situation.
The machine never exposed the problem, so the mazda technicians gave back the car to my local dealer. No more than 12 hours later, before I physically went to take back the car, the defect shows up again, so the car was locked again. Two mazda technicians inspected the car finding 11 errors that, told to me, were transmission errors they have never seen and that required a more deep analysis. After 48 hours the mazda center told me that, despite the errors, being the car working again there was not enough information to make an accurate diagnosis, and therefore they could not do nothing at all.

Now, I cannot imagine that, being my car date 2008, there are no unit testing, stress testing or alike to do on the cpus and other devices to ensure they are working, and I cannot see why there is no decision to start substituting any device, as the immobilizer controller, to see if the problem can be solved. Moreover I cannot understand why, having 11 errors, a mazda center cannot make a diagnostic report; what is the purpose of the error memory if it cannot help?

My local dealer proposed to see if there is a way to exclude the immobilizer at all, stating that the anti-thief circuit will continue to work. I don't like this approach very much, since I payed for every single bit in the car.

The fact is: 3+ weeks and I still don't know what they are going to do, and how long it would take.

Any comment, idea, suggestion is strongly appreciated.

 

My local Mazda Center refuses to check in again the car, stating the car has no problem and until it locks so that they can attach the computer to it while still locked, they cannot do nothing at all.
The fact that the car has locked and they found 11 errors does not count at the moment.

I had no other option, in the meantime, to take the car to another non-mazda center, to see if they can get rid of the problem.
This is the fourth week I don't have my car.

 

This is what we have found so far: if the car stays at the sun, and the temperature get higher, the car is blocked and the immobilizer shows the problem. After the night, getting cold, the car starts again.
Therefore it seems there is a kind of relationship between the temperature and the problem, even if the first two times I got the problem it was in the early morning (and hence after the night).

By the way, it seems that the PCM is the guilty part and will be changed. As far as I understand the PCM cannot communicate with the ECM (that I don't know what it is) when the problem arises.

Anyone has experience on this?

 

At the end of the 6th week I got back the car, and it was working!
I was told it took so much time because apparently it is not a common car, and in particular it is not common here in italy, and therefore there is not enough knowledge about this kind of problems. However, I guess that substituting the board at the very beginning should have solved the problem without need for any further investigation.
I hope the problem will not raise again, and I'm curious to know what could have damaged the board/PCM /I mean, could it have been the weather, the sun exposure, or what?).

As a side note: once get back I noted that the front collision sensors were unbalanced, and apparently this has nothing to do with the PCM substitution, even if I had to re-asset them.
Second thing I noted is that before the PCM substitution, once I was pressing the key and getting the green KEY light in the control panel I also heard a beep; now no sound is emitted. I believe this is due to some software upgrade and it could be a programmable option, but I don't care very much. Anyway, this difference could also make possible to know if the previous PCM was a well known buggy one.