Another Strike

Posted by stephens on May 22, 2010 in Engineering Decisions, Rants | 5 comments

The discussion on the validity of restrike dies continues with a comment from Eric Kam that I would like to post here instead of comments due to length:

It seems to me that the reason that the restrike is more appealing and rational than the trim line correction…

by creating an adjustment point closer to the final destination (the conveyor belts and the racks) they are able to affect the “illusion” of control over some of the variation. To address the initial form or flange die may require the definition of a different tip angle, or inclusion of a CAM that was not originally part of that planned operation. Additionally, fudging a “calibration” hit or tweaking the restrike may have far less negative impact on the overall behavior of the panel.

I am not defending it, but I can see the safety-blanket-like appeal of using the purpose-built restrike die as a control point. A tool that if the attempt at re-cutting or shimming or welding up or just plain hard hitting the tool away from original design intent will not have as broad or potentially negative impact on larger portions of the product. severely mess with the restrike and you are not messing with the product.

Similar to the concept of double draw beads used in large panel draws. We know that square lock beads can effectively lock out all material flow, and that grinding on those square beads can affect all gradients of restraint until the bead is gone. However, we still see double draw beads and the significantly larger blanks that they require in production. Why? Because if I have to make adjustments to the material flow, it feels like we can safely grind and weld on one of the two drawbeads without drastically altering the die. While grinding on the ONLY drawbead feels more risky.

Rational yes, but still an illusion of control.

What folks need to be able to do more effectively during their engineering of the PROCESS and all the tools in the entire process, is to meaningfully evaluate not just what the die does in the best/worst/nominal cases that they might be simulating as their standard. But they need to consider what happens when the world changes slightly around them-as it will. In springback there is no Best or Worst case.

Different is Different. And when it comes to springback any difference can end up being wrong. The restrike die just gives us one more bite at the apple to fix what we could not stabilize in the initial hit and when it appears to fix the problem at buy-off via tweaking, it actually becomes a bit of a self-fulfilling prophecy. We will prove by having it that we needed it. Vicious cycle, the only good news is that we might get paid to make the restrike die, so I don’t foresee the world doing away with them.

Good points, Eric. But with Obama now wanting more aggressive CAFE standards in the next 20 years, the only profitable way to stamp automotive vehicles is to get away from the unnecessary bells and whistles like restrike dies and just-in-case cam operations.