1992 Airbag Module

[bicknell]

My air bag module adventure has taken an unexpected turn. I had found this post by a Mazda guy who re-capped his airbag module and got it working. That is within my skill level, so I thought I would give it a go. I do some hobby electronics, I've made my own circuit boards and soldered components before. Well, I finally got the air bag module out this morning and......mine is totally different.

The part number on mine is E90B-14A624-BA. Same part number as the Mazda guy, except his has a "90" above it (I assume 1990 manufacturer) and mine has a "92" on it. The board looks totally different. I've attached high resolution pictures below.

There are 15 electrolytic capacitors on my board. None are bulging, one shows some signs of corrosion under it, the 4 big ones I can't really see under without removing. That's a lot more work! The thermal fuse is also in a different package, it has on it part number F20F-14A094AA. That lead me to this thread
and this bronco guy video. Same thermal fuse, but yet another different board layout. I also found explorer guy who's board looks just like mine, he just replaced the fuse.

I think I could replace the thermal fuse and re-cap the board. It's more work than I expected, and a bit harder than I expected, but I think it's in my capability. My concern though is two things. One I see corrosion far from the capacitors, on the pins to the IC in particular and on a couple of transistors. There is also some corrosion on the back of the board (lower left in the piucture) that doesn't look too good. Also, there is this black stuff all over the place. Like the blob just to the right of C17. I am not familiar, is that corrosion or some other form of deterioration?

I did send pictures to www.airbagsystems.com to see if they can repair it, and/or if they have replacements they have tested without the corrosion. If they can do either for a good price I'll send it off to someone with more experience.

If anyone with more experience wants to give me an idea how bad they think the corrosion is, and if it is worth repairing I would appreciate the opinion. Any other suggestions on how to get it repaired or replaced would also be quite welcome.

 

[bicknell]

Collecting some more info here. Called my local dealer's parts department, had them search the computer.

E90B-14A624-A is an engineering number, not a part number.

The last valid part number was F2AZ-14B056-A. I then noticed stamped very faintly on the outside of the blue case on mine is F2AF-14B056-AD. Searching for those lead me to this post in a Miata forum which suggests the Mazda part number was NA01-67-7F0A for the same part. Over in this thread we see the Mazda part number, but also F0OF-14B056-AA and F0OF-14B056-AB. That helped me find a reddit thread which adds part number F1DF-14B056-AA to the mix.

Based on what I have read, the same module was used in:

1992-1994 Aerostar, Capri, Crown Victoria, Econoline Van, Grand Marquis, Tempo, Topaz
1994 Festiva/Aspire
1992-1993 Mustang
1990-1994 MX-5 Miatas

And I think variations on the part number, but otherwise compatible modules might be in a number of other things, like Broncos and F150s in the 89-1995 timeframe. Anyway, there's a bunch more part numbers to search for, but there are still few options out there.

 

[AeroCoupe]

I just used www.car-part.com and it cross references all the modules that will work. Average cost is around $30. You want to make sure that the car you get it from did not have an air bag deployment.

 

[bicknell]

Good info. I may have to just buy some and check them out. Seems like many have bad, or going bad capacitors, and/or corroded boards. I bet most of the pull-a-parts don't open them to look. Might have to just take some chances.

 

[bicknell]

I found some folks who repair PCBs and asked about these pics. That splotchy black is corrosion caused by failure of the solder mask. If it was one small spot it could be fixed, labor intensively. But with this much on the board I’ve been advised to not even try and repair.

Looks like I’ll be ordering one or two and seeing if I can find one working and in better condition, and maybe one with the 5 capactiors so I can go ahead and recap it.

 

[bicknell]

I used car-part.com to find places that had more than one, and were located in the desert south west. I'm hoping I can find boards in better condition if the vehicles lived their whole life in that climate. I've sent off a couple of requests for quotes, we'll see what comes back.

 

[Mustang5L5]

Those caps are exactly the same ones used on the ECU. I’ve pulled off a few ones that looked perfectly fine and found them to be bad. I wouldn’t be surprised if these failing modules were all in need of a general recap

 

[bicknell]

Those caps are exactly the same ones used on the ECU. I’ve pulled off a few ones that looked perfectly fine and found them to be bad. I wouldn’t be surprised if these failing modules were all in need of a general recap

Agreed. While the troubleshooting guide suggests shorted wiring is the likely cause, at this age I'll bet 99% of the deaths are due to capacitors, and/or other board corrosion. If I can get a working board I will probably recap it even if it works, given how wide spread the failure is on these modules.

To compare I've been working on my radio cassette deck hiss issue on the side. Looks like it is the cassette mech, I found a guy who repaired them for 30 years. I have a part coming. Anyway, inside the radio the circuit board looks like it is brand new, not a bit of corrosion or tarnish anywhere, no caps showing any issues. The radio also sounds fantastic.

I think the airbag boards used cheaper parts, or maybe just a bad supplier. Mustang, Miata, Bronco, F-150 forums are all full of dead modules. Insert grumble about the cheap parts being used on the SAFETY SYSTEM for the AIRBAG.

 

[Cheapskate207]

What’s the issue with you airbag system? I just repaired mine on my 1994 f150. The light was solid, I removed and found the thermal fuse to be bad. I replaced it and it’s good to go.

 

[bicknell]

My car has "Code 51" indicating the thermal fuse has been blown. I can verify that from visual inspection as well with the box removed. At a minimum it needs to be fixed on this box, but then the question is what caused it to blow the thermal fuse in the first place?

The troubleshooting manual suggests the most common cause of that code is shorted wiring in the car. I'm pretty sure that's not the case. Three starts with bad wiring and it blows the fuse as a protection feature.

For years, lots of people online have reported the boxes going out with code 51, and the reason is always bad capacitors inside the box. When the capacitors leak it causes the same look to the box as bad wiring. The fix tends to be replacing all of the capacitors and the thermal fuse. I had intended to pull mine out and put new capacitors on it as a first step. When I pulled it out though I noticed significant corrosion all over the board which is not often reported. Since the rest of my car is in great shape (garage kept its whole life) I think I got a board from a bad batch. Because of the corrosions I've changed plans to locate a replacement box -- possibly one that still needs capacitors.

 

[Cheapskate207]

Oh ya, I see the corrosion now that I’ve zoomed in. I got a pack of 10 thermal fuses on Amazon for $5. It might be worth throwing one in and seeing if you get any other codes.

Does the horn work? And cruise control, if you have it.

 

[bicknell]

Horn, cruise, everything else works perfectly. Well, except for the tape deck. See my other thread.

No corrosion or cap problems in the radio. Inside of the radio looks brand new.

 

[bicknell]

I sent high res pictures to http://www.airbagsystems.com/ and they believed they could repair my module. I went ahead and shipped it off to them, it should have arrived late yesterday.

I still plan to obtain at least one more module and if necessary (and they do a good job) get them to repair it as well so I have a spare. These things are so hard to get, and it's only going to get worse.

 

[AeroCoupe]

Let us know if they can fix it as I have a spare that I know works and one that does not so I would send them the one that doesn’t work for a rebuild.

 

[bicknell]

I got the module back today. Cost was $100 for the repair, $12 for return shipping, and I think I paid like $11 to ship it to them. I've included after pictures below. I consider the cost quite fair for what they did, and the turnaround time was awesome. 1 day to repair, they then send an invoice (so you don't pay if they can't repair) and shipped the same day I paid the invoice. Two day turnaround, basically.

From a visual inspection it appears to me they replaced:

• All 4 large capacitors on the right.
• The black metal oxide varistor in the middle with a red one.
• The thermal fuse with an SW-128T.

They also claim they "programmed" it. It's not clear to me these have any real programming, but they do somehow remember 3 bad startups to blow the thermal fuse, and so perhaps they reset that counter?

Now, here's the part I'm kinda of two minds about. The thermal fuse in the unit originally is a NTE-8139, which is activated at 141C. It's paired with a resistor in the case, when the unit has detected the three bad startups it turns on the restore which cooks the thermal fuse. The thermal fuse they installed is a 240C fuse. It's no longer in a case, and the resistor is not particularly close to it. It seems to me like they are intending for it to never blow.

On the one hand, this seems a bit like defeating a safety device. Clearly this was designed in for a reason. On the other hand, these things are impossible to find, hard to repair, a pain to get in and out, I don't really want it disabling itself after 3 bad starts. If there is a problem I want it to tell me so I can troubleshoot the wiring/issue and not kill itself.

I'm probably going to use it as they delivered it, figuring they know more about them than I do. I still need to install it in the car and make sure it works.

 

[bicknell]

No joy in mudville. I plugged it up. The good news is the airbag did not immediately deploy.

I am getting code 13 and 53. Code 13 is "Front crash sensor or air bag circuit shorted to ground ", and code 53 is "Diagnostic monitor internal fault". Looking at the schematic I would think 13 means it thinks pins 17, 18, or 19 are shorted to ground? The other side, 20, 21, 22 look like they are connected to ground in the sensors. I guess my next move is to break out the multi-meter and see if any of them are shorted to ground.

Video showing light pattern: https://youtu.be/11veUlpBmOM

 

[bicknell]

Did some diagnostics with the multi-meter:

• Pins 2, 3 and 5, CK57, all show good grounds.
• Pins 20 (CK 622), 21 (CK620) , and 22 (CK618) all show connected to ground.
• Pins 17 (CK 617), 18 (CK619) and 19 (CK621) all show open to ground.
• 17-22, 18-21, and 19-20 all show a resistance of 1180 Ohms. Like, exactly.
• Pin 13 has 12.6 volts, so fuse 7 is good.
• The Air Bag indicator light flashes, so fuse 18 is good.

I'm seeing no evidence to suggest the wiring is the problem.

 

[bicknell]

Alright, an update on my saga. I sourced an airbag module from a junkyard in Arizona. It came out of a crown vic but it's the exact same board. Sent it off to the airbag systems people who replaced the capacitors and a few other parts. Got it back today, put it in the car and success!

In the end I can absolutely recommend http://www.airbagsystems.com/, they did nice work. Obviously your module has to be in repairable shape, and my first one was too corroded. I suspect the capacitors are going bad on all of these boards, I almost recommend preemptively getting them replaced before they corrode too much and fail. No one is making new modules.

One minor detail. When I put back together the climate controls the hot/cold blend control in the middle turns a bit further to the right (all the way to the end of the red line) than it does to the left (about 1/4" short of the end of the blue line). That seems to be full hot and full cold. From how the mechanism looked and went together I don't really see how I could have messed it up when it was apart. That said, I figured I'd ask 1) are other people's the same, and 2) does anyone know of a way to adjust?

 

[tbarney]

Now, here's the part I'm kinda of two minds about. The thermal fuse in the unit originally is a NTE-8139, which is activated at 141C. It's paired with a resistor in the case, when the unit has detected the three bad startups it turns on the restore which cooks the thermal fuse. The thermal fuse they installed is a 240C fuse. It's no longer in a case, and the resistor is not particularly close to it. It seems to me like they are intending for it to never blow.

On the one hand, this seems a bit like defeating a safety device. Clearly this was designed in for a reason. On the other hand, these things are impossible to find, hard to repair, a pain to get in and out, I don't really want it disabling itself after 3 bad starts. If there is a problem I want it to tell me so I can troubleshoot the wiring/issue and not kill itself.

I'm probably going to use it as they delivered it, figuring they know more about them than I do. I still need to install it in the car and make sure it works.

I'm a Miata guy, and I just stumbled on this thread researching some airbag module stuff. But I couldn't pass by this without commenting.

They have absolutely defeated a major safety component of the airbag system. It's not 3 bad starts. The thermal fuse is intentionally blown when the diagnostic module thinks that one of the crash sensors has failed. This is to prevent an accidental deployment if another one fails. Here’s how it works: It only takes a signal from one crash sensor to activate the resistor below the fuse, but it takes two or more to activate the inflation signal which goes thru the fuse. Since it takes a while for the fuse to blow, it is ok when a very fast inflation signal goes thru. However, if one crash sensor is activated, say by a sensor fault, there is no inflation signal. The resistor gets warm, and if the fault it not corrected, it gets hot enough to trigger the fuse. The fuse is blown after some delay, and the airbag is effectively deactivated preventing an accidental deployment later. If it did not blow the fuse when it got a single invalid inflation signal, then a second invalid signal at another time would act with the first fault, and possibly trigger the airbag unexpectedly.

In the Miata community, many of have found that our crash sensors are fine, but deterioration of the module circuit board makes it think that sensors have failed. But the safety logic can't tell the difference, and the fuse gets blown anyway.

With the resistor not touching the fuse, there is no way for it to transfer any heat. So, an invalid signal (including damage to the circuit board) would not activate the thermal fuse as designed. This leaves you open to random airbag deployment in the case of system faults.

You may be able to correct this by bending the leads on the thermal fuse to insure contact with the resistor. Some modules have a thin copper wrap around the resistor and the fuse to insure heat transfer. Other designs don't use a resistor at all, but wrap the fuse in a polyimide heated film which serves the same purpose: current through the film which for long enough, without the other signals needed to send the inflation signal, will heat up the thermal fuse enough to open it. Yet another appears to be a type of inductive coil wrapped around the thermal fuse. The intent, in all cases, is to have current-generated heat when current is flowing at the wrong time, which then heats up the thermal fuse to create an open circuit.

This Miata.net thread discusses the various types, with pics: 90-93 Airbag Module Thermal Cutoff Fuse

Just didn't want to hear about an airbag going off in your face driving down the road!

All the best,
Todd

 

[bicknell]

I appreciate the additional information. For the record, my module is alternately working and throwing a code 13, which is the RH front sensor bad. I've tested the sensors and wiring and I'm pretty sure they are fine, so I think my "repaired" module is still flakey.

I'm not sure if the Ford and Mazda operate exactly the same, but here's how I understand the Ford wiring. There is a deceleration sensor in the A-pilar, and three crash sensors across the front of the car. To set off the airbags takes the deceleration sensor closing, and then any one of the three crash sensors. The module checks all three to be sure they are open (really, 1200 ohms resistance) when the key turns on. If the car is started 3 times with a faulty sensor, it activates the resistor and blows the thermal fuse. The large capacitors down the side of the module store enough energy to set off the airbag even if the 12v system is no longer operational for a few minutes. The wiring diagram is basically capacitors -> a-pillar deceleration -> clock spring (airbag) -> module -> 3x crash sensors -> module ground.

Your warning is a good one though, having that fuse disabled would increase the chance of an errant deployment if a sensor was bad.

I'm kinda at a loss as to what to do next though. Even with the capacitors and MOV replaced there are clearly still flakey components. My board from the desert is better, but still not perfect. Both my boards show moisture under the solder mask.