Hi all. I'm looking for some help diagnosing and fixing my wife's car (2018 Mazda6). It's having an issue in the past couple days where the electronic parking brake (EPB) won't engage or release when asked to with the switch. However it will engage/disengage the brake automatically when the car is parked.
The timing of this coincides with a weird bump we felt in the car wash. I heard a bang and the car kinda jumped in the track. It drove normally afterwards but the EPB warning light started showing on the way to dinner later that evening. The message screen says "see dealer" and my code reader doesn't seem to be able to access that particular module. Very helpful all around.
This is the newest car I've worked on by far, so I'm a little lost. Are we stuck with a dealer visit?
EPB's are pretty simple, there are two wires going to each caliper. i'd take a peek under the car and look for damage to the wires leading to the calipers.
In reply to AngryCorvair (Forum Supporter) :
That's reassuring. I did a perfunctory jiggle of the plugs on the calipers so I know at least one end is good.
In the interest of humor, I should probably add that I was in the driveway trying to figure out how make the parking brake automatically disengage for like an hour until my wife came over and said I needed to have my seatbelt buckled first. Lo and behold she was right. Her car was blocking in the other two and parked too close to the fence to get the passenger side wheel off so I was getting pretty annoyed.
In reply to thatsnowinnebago :
yep, "driver presence" is required for auto release. depending on the suite of sensors on the vehicle, one way to "determine" (with some certainty) the presence of a driver at the controls of the vehicle is to look for things like "is the driver's door closed" and "is the driver's seat belt buckle engaged" etc.
I finally crawled under the car today and everything looks perfect. Not a scratch or scuff anywhere. Tugged the wires too and nothing was loose. What should I check next?
Time for a trip to someone with a diagnostic tool that can read EPB faults.
AngryCorvair (Forum Supporter) said:
In reply to thatsnowinnebago :
yep, "driver presence" is required for auto release. depending on the suite of sensors on the vehicle, one way to "determine" (with some certainty) the presence of a driver at the controls of the vehicle is to look for things like "is the driver's door closed" and "is the driver's seat belt buckle engaged" etc.
If you are trying to drive a newer Mopar with the door open so you can see where you are placing the vehicle, it will not slam the shifter into Park as long as you have the brake depressed even a little.
I assume they are looking at the pedal position sensor data and not brake fluid pressure data, but IIRC both have to go through network, so maybe either/or, or both?
aside - it is a little amusing what internal logic tests will determine is wrong. Not that I think I could do any better than some poor engineer trying to brainstorm fault testing while writing code that has to fit in a limited ROM space in a small amount of memory, and it has to be done last week. Mazdas will set a code for a failed brake fluid pressure sensor when the brake pedal switch (not position sensor, Mazda is too cheap for that) fails. It sees brake pressure when the pedal "isn't applied", you see. A parts hanger might throw in a $2000 ABS unit before checking a $50 switch.
In reply to Pete. (l33t FS) :
The effort that goes into brake controls software development is probably 5% on performance and 95% on fail safe and diagnostics. I'd say that the Mazda logic you noted is incomplete, rather than flat-out wrong.
In reply to AngryCorvair (Forum Supporter) :
Me too. It isn't wrong, just biased. A more thorough check might cross reference vehicle deceleration (accelerometer or speed sensor based?) versus detected pressure, or a few other ways I could think of that may or may not require additional hardware, memory, or clock cycles.
I'm also thinking about some sarariman sleeping under his desk because he's being voluntold to work an extra 8 hours of unpaid overtime per day to get a project pushed through.
In reply to Pete. (l33t FS) :
"sarariman"
I learned a new word today.
thatsnowinnebago said:
In reply to Pete. (l33t FS) :
"sarariman"
I learned a new word today.
I had to say it a few times until I got it
To close the loop here, the dealer diagnosed this problem as a bad switch for the EPB. $300ish later and it's fixed.
I suspect the failure had to do with someone (who definitely wasn't me) spilling coffee all over it a while ago.
