It is a little difficult piecing together all the information. And much of the detail is not even released.
Takata stated that the recent airbag problem is caused by:
1) excess humidity was allowed into the propellant wafers. I am not sure if this occurred during manufacture or in the warehouse.
2) less dense propellant wafers. I am not sure if the low density is only caused by poor machine capability, but now Takata states that there was an auto reject AR that was able to detect poor compaction, and therefore low density, but the AR could be turned off by the operator.
The result of both problems is that the when the airbag is triggered the propellant can burn too fast. If the burn is too fast, the pressure gets too high too fast. And if the pressure is too high then pieces of the airbag assembly can blow off. Here is Takata's letter to NHTSA.
This passenger side airbag problem occurs with propellant wafers made from around April 2000 to October 2002.
The odd thing is, if you look at prior recalls of driver side airbags from Takata, they include propellant wafers from round the same time and for the same reasons. I am still linking it all together.
The lessons so far:
- Track your material. In the case of Honda, the location of specific lots of wafers could not be tracked which caused wider ranging searches. In another specific additional recall, Honda lost track of a few thousand service parts resulting in recall notices being sent to hundreds of thousands of customers for serial number inspections of previously repaired cars.
- Do a good investigation of problems. When the 1st incident occurred there should have been an effective problem solving exercise. However, it seems that there were some gaps in the process. As a result the recalls were slow to occur and covered too few suspect vehicles. This resulted in higher costs and risk to customers. Did I mention at least one person died from this issue?
- If a failure occurs on part A due to a component X, and a very similar component Y exists on part B, shouldn't component Y be examined to confirm whether both parts A and B need to be contained? In fact in this case I am not sure how different the wafers are between driver and passenger side. They may be common.
- One of the problems noted is that an Automatic Rejection system was turned off by the operator. Wow, so many issues here. For the PFMEA team, now we have to be sure that expected controls are now locked out and will actually be functioning. For the manufacturing team, they have to be involved in the PFMEA and know why the process is as it is, what the risks are, and have effective training. For the management team, what was the reason the AR was shut off? Too high pressure for every piece of production? Why couldn't the people on the floor get the problem fixed? I doubt they simply opted to shut off the AR without some other frustrated efforts.
- Also, the AR was not part of the propellant wafer manufacturing system originally. But, it was added sometime in Sept 2001. Reject control is a common fight we have. THe reject system has to be well designed and maintained.
