Seeking tooling longevity

Figure 1The untreated staking punch on the right was worn and unusable after three weeks of use, while the treated tool produced differential carriers for more than 26 weeks. The left punch is an unused tool treated with the thermal diffusion process.

BorgWarner Inc. is a drive-train developer and manufacturer that specializes in blanking and finishing of shafts, gear and spline rolling, grinding operations, and assembly.

The Muncie, Ind., manufacturer of four-wheel-drive and traction control parts for the automotive industry was having a problem with the punches it used to deform the ends of shafts (axles) in the reduction carrier and differential carrier.

The punches, used to produce Ford Motor Co. and General Motors SUVs and pickup truck parts, were wearing out almost as fast as they could be manufactured. Toolmaker time, maintenance, and press downtime were costing the company a bundle.

After doing some research, BorgWarner's production team decided to have their punches coated using the thermal diffusion (TD) process from the TD Center (see Figure 1). The punches were coated and heat-treated to harden their steel surfaces, helping to extend punch life.

Identifying Costly Punch Problems

Malcolm C. Lyon, senior mechanical maintenance engineer at the Muncie plant, alerted Brian Bunner, senior assembly manufacturing engineer, that they were consuming staking punches at a faster rate than they had in the past. In fact, it was the company's No. 1 tooling expense.

The manufacturer requires that every maintenance operation has a written ticket requesting the work. "We get a summary report regularly that indicates the top five problem areas within a particular operation," said Lyon. "We monitor the tickets to determine problem areas in every department, and the staking press always made the list.

"Many tickets were issued requesting new staking punches," said Lyon. "It was clear that punch life was a problem."

BorgWarner's toolroom was spending most of its time making and changing punches, which are not simple to make. All punches are blanked from D2 material, hardened, and finished. The finishing process requires a lot of grinding to tight tolerances.

"To determine the punches' cost, I went back several years and found a considerable change in the number of punches we were using," said Lyon. "Punch costs were in excess of $77,000 annually. Downtime and unscheduled maintenance were disrupting production."

Breaking Down the Process

"This was a new assembly operation for a differential carrier using a staking operation patterned after the reduction carrier staking operations that we ran for years," said Bunner. "We had to change out the punches used in the staking presses every two to three weeks. Because multiple fixtures were involved, punch costs and changeover time were expensive."

Figure 2Carriers completed through the staking process show the six axles per assembly that are deformed to retain the pinion gear components.

The carrier consists of axles fixed between two carrier plates. Each axle is the rotational point for a pinion gear of a planetary gear set. The axle is hit on the ends to deform it so it doesn't back out of the carrier housing (see Figure 2). During the process, punches were wearing away at the contact points and eroding. As a result, the carrier began sticking to the punches in the upper tooling.

"A stripper plate also is part of the operation, and if it doesn't work right, the part hangs onto the punches and the dial table indexes with the part still on the tooling's upper portion," said Lyon.

"Now the parts become stacked two-high and when the press comes down during the next cycle, a jam occurs. The press is stopped and manually raised to clear the jam, which is costly in terms of part loss and increased downtime, and it raises safety issues," Lyon said.

Press operators had to bump the carrier assembly to get it to release from the lower tooling fixture. "When it got that bad, the press was stopped and a toolmaker was called to change the punches," Lyon said. "There were times when we didn't have any new punches to install. We would pick and choose the best used punches we had to keep production going. Otherwise we had to stop the assembly line."

Five different staking operations are used for this particular punch configuration (see Figure 3). "We have approximately 130 of these punches in use every day on multifixture dial index machines," said Bunner. "Each machine consumes 20 to 50 punches during a tool change. If several machines require punches to be changed in one week, we can easily deplete our stock.

"This is a fairly uncomplicated process," said Bunner. "We're deforming axle ends so they can't fall out. This is controlled by visually inspecting each part. Once per shift, or after a tool change, we take a part over to a press- out machine and push the axles out from each direction. A minimum amount of force is required. As long as we are above that level, it's a good part."

"When they press the axles out," Lyon said, "they're determining if sufficient material has been deformed during the staking operation. Normally, there's ample material resisting the press-out and the force is well above the minimum limit."

Solving the Problem

"We were already using a snap ring installation tool treated with the TD process on the assembly line," Lyon said. "We knew the coating increased tool life, and we researched what it could do for the punches."

"We had quite a bit of experience with the TD [vanadium carbide] coating on the snap ring tool," Bunner added, "and over the years we learned from that application."

Once Lyon and Bunner had defined the problem, they redesigned the punches to make sure that they would fit after a coating. "After we manufactured new punches and had them coated, they were installed in the staking presses," Bunner said. "We discovered that instead of lasting two to three weeks, they lasted longer than six months."

Figure 3Various reduction and differential carrier styles are positioned in a dial indexing staking system.

After switching to coated punches, the company noticed that tooling costs dropped to less than $7,000 per year.

"Currently we're grinding the punches to the right diameter to compensate for the coating," Bunner said. "We usually process 100 punches at a time, and we have approximately 130 punches cycling in production."

"Before switching to coated punches, we had an inventory of 180 punches, which would last for two to three weeks," said Lyon. "Now we can support production with half that amount, and actually schedule planned maintenance for punch changes."

BorgWarner also has the ability to stagger punch changes on different machines. This has reduced the manufacturer's punch inventory and cut down on unscheduled maintenance.

"When a differential staking press is down, we may have only a couple hundred pieces ahead of the assembly line," said Bunner. "This means we have the time it takes to run 200 pieces to get it fixed—75 to 100 minutes. It takes longer than that to change the punches."

Improving Part Quality

"Before using the TD process, we judged part quality by the press-out, staked part appearance, and whether the parts stuck on the tooling," Lyon said. "When the punches would wear and erode, they left a rough material surface on the staked axle.

"We have better-looking stakes over a longer period of time now," Lyon said. "This adds to the quality of our product."

"Improving punch life didn't allow us to operate the staking press any faster in terms of cycles per minute," Bunner said. "However, we do have a big increase in productivity by reducing downtime for changing punches. That's what affects productivity more than a machine's cycle time."