Not just from being d/c'd from their unit. It's usually from a power surge/short or something of that nature that causes them to go off. A cut wire/ground causing a short somewhere, or a surge from the bags getting sudden power is usual cause. Delayed deployment isn't something that happens in modern cars anymore from what I've heard - but it's best to play it safe in a situation of this nature.

It's safe, you just need to make sure that the power is discharged from the system first. Last I checked, the bags get their (most) power through the control unit.

I don't know, to be honest. Again, I'd just give it the recommended time. After that period unplugging/plugging the various components shouldn't hurt/cause issue.

I'm rather over cautious though. Anytime I am working on my car I tend to disconnect the battery first if it's something that is going to take more than a second. I dislike the idea of accidentally frying various electrical bits in cars nowadays. Also if I have the doors open for a period, or anything of that nature. It's just a habit I got into at the shop that my first boss taught me so that we didn't kill people's batteries or fry their systems from welding or whathaveyou. I'd like to think it's saved me some money, as I've seen tech's fry things for the sake of not taking a precaution that takes all of 10 seconds to do (less you're waiting for airbag, 30m 10s).

 

To disconnect the driver side airbag, all you have to do is disconnect the plug in the steering column. Than put a 2 amp fuse in the plug leading to the control unit to fool the computer that the bag is still connected and to prevent the SRS light to come on. I did this when I changed my steering wheel.

As far as the passenger side, I'm not sure yet, but I want to find out as I plan on taking my car to the track. I think the same method will work for both sides.

 

Necro bump!

I'm considering putting in a 15A (or 7.5A not sure yet lol) rated toggle to cut power to the SRS-ECU so I can turn the system off and on as needed for street/track use.

I'm not an electrical engineer, so I have very limited knowledge of wiring and schematics.

Does anyone know which wire I need to splice to insert my toggle switch?

It looks like power is delivered to the ECU via 13 or 16?



Looking at the schematics I think it might be one of the places highlighted:







Theoretically, this should work just the same as an ignition switch. Should be as safe, too. With the car off, I set the circuit to open with the toggle and then start the car normally (SRS light will come on). When I'm ready to allow for the SRS system to operate normally on street I would do the same process again. With the car off, close the circuit with the toggle and then start the car normally.

Thoughts?

 

Curious on this as well for future endeavors. Also not a EE or engineer at all but please confirm with someone who knows what they are talking about prior to splicing, granted it's you we are talking about and you're more OCD and organized than me so I'm not sure why I'm even warning you

One thing to keep in mind is that the air bags probably use a capacitor of some sort to store the energy needed for the air bags to deploy so breaking the circuit may hinder the charging of the capacitor but not the discharging if that makes sense. But like I said I'm not well versed on the topic so hopefully someone who actually knows what he is talking about will chime in.

 

A friend pointed out that the SRS-ECU has a back-up capacitor. I might have to wire in a switch for the two airbags and the two seatbelt pre-tensioners...



He did in fact confirm that 13 & 16 are the power wires, if it matters.

 

Maybe I'll have to move this out of Evo Electrical / Audio / Security to give it more visibility...

 

That sounds like a really, really bad idea. If you were unhooking the battery before flipping the switch maybe, but that would seem to defeat the purpose.

If you want a quick disable just open the glove box and unplug the passenger air bag connector that is right there. When it is unplugged (or any SRS component is in an error state) the system is deactivated as a safety precaution. Depending on when the ecu was built it may or may not leave the SRS light on afterward, however that would be the same as putting a toggle to the power which would also leave the light on afterward on an affected unit.

 

I'm all for doing this with extreme caution. I really don't want to screw up this system nor do I want to put my life at risk. I have two kids and a wife.

Conceptually, the ignition works in a similar fashion as a toggle in this case. I would think putting an additional break in the circuit this way would work the same, as far as the system is concerned.

The back-up capacitor would defeat this, of course. I was thinking about maybe inserting the break at the ground instead of the power, but the circuit board or case for the SRS-ECU might have an additional ground via the bolts/chassis.

The entire system or just that airbag? Since safety systems are designed for redundancy, I would assume that the rest of the system should continue to work in order to protect the other person in this case. It seems like some is greater than none when there is a fault in the system.

I don't mind the SRS light, but some reading seems to reveal that when the SRS-ECU is reconnected the code clears and it's business as usual. This is not the case with anything else throwing a fault code in the system, which requires a MUT III to properly clear IIRC.

Thoughts?

 

Looks like the capacitors discharge in ~10 minutes.

http://www.airbagsystems.com/switch%...azda_Miata.pdf

 

Yeah if you are letting the system shut down before touching the switch that is fine. My concern would be forgetting/being in a hurry etc and flipping it not realizing the key is on, or a passenger pulling a WHAT'S THIS DO etc. The last thing you want to do is send a power spike to the ecu. Cutting power is the preferred way over ground.

If you are doing it with the system down it shouldn't be an issue. I just worry about the what ifs.


The entire system is supposed to shut down if any one device is reading out of spec. This is why it constantly monitors all resistance values, to make sure it has perfect conditions for operation. If it can't be certain that a particular value is correct, the system is shut down to avoid unknown behavior. Since it can't tell if (in the example of the passenger bag) the bag is bad, the wiring is bad, or the ecu driver is bad, the entire system goes down. If the fault was from wiring being jacked up and the system were allowed to operate, it could cause incorrect deployments; if the ecu is internally puked, it could malfunction on other portions of the system too. It is safer for it to just shut down.
While much older systems we would see a deployment even though the system was apart (passenger bag goes off with aftermarket steering wheel), we don't see it any more. And on top, have had more than a few greasy attempts at suing because somebodies boy racer kid put a grant steering wheel on and wrecked his car and no bags went off etc.
By design, this system should shut down with any component disabled..... any time the light is on basically.


With regard to the light clearing.... it is hit or miss depending on the ecu; I don't know if that is intentional or not, as supposedly they are meant to hold -all- codes permanently requiring a scan tool to clear. I have run in to more than a few cars where this did not hold true even though an identical model beside it did. For example a buddies 03 OZ has a passenger bag fault come and go and the light comes and goes with it. But when I'm pulling these recall bags out I have to use a scan tool to clear the fault from it being unplugged and me having to cycle the key on.... it won't go out on its own. Whether the exception is a build date change, an ecu that was sent out for repair in the past and lost that trait, who knows.

If you are cutting the power to the SRS on this model it should turn the light back out since the ecu didn't have power to record the fault to begin with. If this were an X, then that would be a different story as the SRS fault would report in ETACS as well and set a permanent fault. 8/9 is dumb, the light coming on from the ecu being unpowered is just a knee jerk reaction as the ecu turns the light off, and can't since it is unpowered etc.

 

Interesting. It sounds like my assumptions were fairly close to the mark.

I wonder if there is a way to see if the capacitor is fully drained. Maybe an LED in the circuit that illuminates when there is power? Trying to K.I.S.S., but not knowing if the system can still trip (or not) when on track isn't re-assuring.

I'm going to get an SRS-ECU off of ebay and open it up.

 

The less modifying you do, the safer you are. I personally wouldn't touch anything internal.
It is pretty safe to call them disabled in 30-60 seconds, though the "Safe" instruction is usually 2-5 minutes depending on the manufacturer. Even at that, it doesn't really matter as it isn't like you are going to be throwing the toggle seconds before an impact or something, you'll be disabling the system then going racing or whatever. If you are concerned about it setting a fault while the system still has power in the capacitors, that won't happen. Once supply and back up voltage are cut, it is incapacitated. The capacitor is just reserve discharge power to drive the igniters.

 

From outside of the SRS-ECU, I don't see a way to detect the charge level of the capacitor or a way to discharge it as it is safeguarded by a diode and ends at the airbag circuitry.

 

Are you suggesting that the diode will prevent the capacitor from discharging if power is removed and the circuit is closed via ignition?

 

No, I'm saying you can't speed up the process by grounding the capacitor because the diode forces the connection to be a one way only connection. If the diode wasn't there, you could short the capacitor to ground and thus remove the wait time but because it's there, you have to wait for it to discharge naturally.