Engineering Design Solutions

Recently, Fastco was able to provide an innovative re-design of a carbon steel fastener into a stainless steel Phillips drive screw. Anyone who works with forming stainless steel knows that it is a lot more difficult to form that carbon steel as it is a harder material.

One of our process engineers, Mike White, spearheaded the process of redesigning the part. It went through three designs before turning into a functioning part that we could run smoothly on our equipment. Here’s the story from Mike’s perspective:

“This part started out as a carbon steel fastener. We already knew when quoting it that the upset ratio on the head was very high. Another concern was the amount of force being applied to the Phillips drive on the tool while forming such a large, thin head. So when the customer expressed that they wanted this part to be switched to stainless steel, we weren’t positive we could do it. Still, we told them we would experiment and see what happens.

Being stainless we switched the part from a 2 die, 3 blow to a multi station machine. This allowed us more forming stations as well as access to an induction heater. Upsetting stainless steel this much, we knew we’d need the heater. We went through 3 different progression designs on this part before we found a solution.

Design #1

We wanted to form a long cone upset into a bubble because we knew a bubble would be ideal for putting a Phillips drive in the next station. The closer the upset is to the finished shape of the head, the less material you have moving across the Phillips insert and trying to snap it off. While the wire heater has many advantages, it can also cause problems. In this case the wire was so hot (soft) that the part would buckle on itself instead of forming into a bubble like we wanted even though we were well within the forming parameters in regards to our upset ratio. We could not lower the temperature on the heater because doing so would cause the extrude insert to score up quickly and our Philips insert to break sooner.

Design #2

Since we couldn’t make the bubble upset we wanted, we decided to take the long coned upset and flatten it a little bit in the next station to make a short coned upset while forming the shoulder at the same time. This would lessen the amount of force required to form the head. This design broke the Phillips punches very quickly. Having a cylinder shape hitting any drive is not ideal when working with stainless steel. So we went back to the drawing board.

Design #3

We knew that design number one was the most optimal, but it didn’t work. So we turned our thoughts to, “What do we have to do to the machine to make this work?” In order to stop the buckling of the upset, we would need to contain the material inside a tool so there would be nowhere for it to go. This can be done with a telescoping upset configuration. This is something we use all the time on various machines. However, in this case, it wouldn’t work because we wouldn’t be able to load the part in the die before the fingers dropped the part. To correct this issue, we designed a new upset configuration that worked perfectly.

We came up with a solution for the loading issue, which gave us the upset bubble shape. Even with this new tooling working how we wanted, we still weren’t making the amount of parts we wanted before the Phillips punch broke. It was better than previous attempts, but still not enough to justify running this job.

As a last ditch effort, I wanted to try a custom punch configuration before we pulled the job and called it quits.

This seemed to do the trick. We ran substantially more parts before we ran out of material. The Phillips punch never failed. In the end, the custom punch and upset configuration worked for us.

All three designs worked flawlessly during simulation. In the simulation world, every variable is right where you want it to be. The part is on center, tooling is made in the middle of the spec and the material is perfect. So while we were able to see the material flow, tonnage, and tooling pressure, we were skeptical of these results and had to just use our experience to determine what would happen in the real world. On the majority of parts, a lot of these variables wouldn’t cause any issue, but when you’re dealing with a difficult part that is riding on the edge of forming capabilities and parameters, any slight change can affect the outcome.”