Punch Grinding...Thriving in Chaos

Many shops wince at soaring customer demands for faster-than just-in-time support.  Many wistfully recall days of more leisurely and “reasonable” leadtimes.   However, according to one shop, focusing on what was is illusory — and, perhaps, strategically dangerous CAMBRIDGE, ONTARIO — Stefan Lorbach began his career as a tool and die maker in 1959.  Between then and now, he’s seen a great deal of change.  It’s like going from “the earth to the moon,” he says.  During the 1970s, he notes, it might take three to four months to make a die.  Today the same die is done in a few weeks. “When I opened Stema Punch & Die in 1986,” he says, “it was a one-man operation supplying tool and die shops and metal stampings.  Deliveries were three to four weeks.  Now if I tell customers that I need seven days for a job, they want it quicker.” But this is precisely why he started the business.  And even though much has changed since 1986 — jobs have gotten tougher, more challenging, and technology has constantly advanced (some of the equipment Lorbach uses today, like a Studer S32cnc punch grinder, wasn’t even available a year or two ago) — what hasn’t changed is the demand and expectation by customers for ever-faster job turnaround.  “In some respects,” Lorbach says, “we were ahead of the game in servicing the customer.  We’ve done JIT since day one, before it was called JIT.  Now everyone expects that level of service — and more.  But that’s always been our approach to customers.  That’s been our strength.  That’s what we do.  Customers call us early in the morning and need a part by late the same day.  That’s the norm.  Change for us is routine.

A Diversity of Technology

Stema employs 25 and operates a wide variety of some 60 machine tools — everything from CNC and manual wire and sinker EDM machines, to CNC and manual surface, centerless, cylindrical and punch grinders, to CNC turning and vertical machining centers, to CMMs, optical comparators and more. And largely due to the nature of the business, Lorbach must forego the luxury of building to inventory.  Every punch produced is unique — diameter, length, points, profiles, surface finishes.  The only thing approaching routine in the chaotic schedule is the prepping of punch blanks.  These are made up in advance, but beyond this “routine” operation, everything else about Stema is custom and on the fly. “You might think we’d be able to run some jobs as a batch, a family,” Lorbach says.  “But that’s rarely the case.  The mix is just too high.  We may do two ovals, followed by six round, followed by four square or rectangular punches.  And if it’s not the mix, then it’s the delivery requirements that make running batches next to impossible.”  Even when a volume opportunity, say 50 to 100 pieces, does occur, it still must be produced just in time.  In a 50-piece order, Lorbach says, there may be ten of one type punch, eight of another, four of another — all of differing lengths, diameters and profiles.  So, the idea of running production, as appealing as it may be to most shops, just does not translate well into the Stema operation. Machine-to-operator ratio is 2:1, often 3:1 or more, which Lorbach says is critical to flexibility and cost containment.  “Our employees often run two or three machines at once,” he says, “and this affords us the flexibility to work with a schedule which is almost entirely unpredictable.  An operator will use one machine for roughing, another for finishing and a third for something different  — putting a flat on a part, for example.  This way we’re not constantly changing machines over to meet different requirements.  And if a customer needs one or two pieces in an emergency, we don’t have to interrupt a process to dress a wheel or do a radius.  Which is where we could potentially lose a lot of time and money — in the constant starting and stopping to accommodate a rush job.  This is why we have so many machines and so many different types of machines.” Eye on HorizonBeyond keeping his focus on what is admittedly a “crazy” production rhythm, Lorbach is constantly engaged in monitoring technology changes that may allow him to do something faster, better, simpler, more efficiently and without, of course, any sacrifice in quality. A good example: A new Studer S32cnc punch grinder from United Grinding Technologies, Inc. (UGT, Miamisburg, OH).  “We already had two CNC punch grinders,” Lorbach says, “and then business got to the point that we needed a third.   I knew Studer made good machines, but I didn’t think they offered a punch grinder in North America.  As fortune would have it, just when we were looking, UGT introduced the S32cnc.  We went and looked at the machine, spent two days grinding punches, and bought it, the first S32cnc in North America.” The S32 provides Stema with another degree of flexibility.  Lorbach says that he uses the machine exclusively on very hard material, M2 tool steel.  Punch blanks, ranging from 0.125" to 3" in diameter and 2" to 6" in length, are roughed and finished on the Studer. “In some cases,” Lorbach says, “we’ll have to do secondary operations, but these are cases where we have to EDM additional profiles or threads, or produce keyholes.  But about 80 percent to 90 percent of the punches are roughed and finished, ready to go, on the Studer.” Other than flexibility, another reason Lorbach settled on the Studer is its ability to hit and to hold close tolerances.  He reports that most of his punches require ±0.0005" on the OD.  However, some very special punches require an OD tolerance of ±0.0003". “We can get ±0.0001" on the Studer,” Lorbach says, “and we not only can handle 1" ground profiles, which is pretty standard, but we can easily tackle 2.50" to 3" profiles for custom punches which, again, gives us an additional competitive advantage.” Programming the Studer has also added to flexibility and increased throughput.   Not only does the machine control store the process parameters for a large number of routine jobs, making changeover from job to job as easy as calling up a part number on the Fanuc 16i touchscreen and letting the machine adjust itself, but “tricky” parts or new, complex ones can be readily addressed through built-in routines that can be adapted to the specific need, and then stored for the next time the part comes along. 

Saving Crucial Time

It should be no surprise that automation is not something that fits easily into Stema.  Which is not to say Lorbach doesn’t carefully think about various approaches to time saving and process enhancement, including forms of automation.  There are some regular part processes, he says, that lend themselves to unattended operation.”  Producing threads on punches by EDM, for example.  In this case, an operator can load up a significant number of parts on the table and run them at the end of a shift, overnight.  Automating the punch grinding process, however, is a different matter. “Because of the short cycle times,” he says, “in order to grind punches through the night we’d need to install a robotic loader to handle the blanks.  And then we’d probably work on only six parts or so at a time, which would take just 35 to 40 minutes.  So, rather than automate the punch grinding process, we focus on making fixtures that are easy to changeover and allow us to save time that way.” Sometimes considerable time can be saved by simply looking closely at an established process.  He notes, for example, that some of the punches coming off the Studer, in particular square or oblong punches, will require matching die buttons produced by wire EDM.  In these cases, programming the part often required entering as many as 86 different instructions, a process which took considerable time — twice as much time, in fact, as that required to make the part. “We were producing the part in six minutes,” Lorbach says, “but it was taking us 12 minutes to program the EDM.  So we called in our software guy, gave him some profiles, and asked him to rethink the process.  What he came up with was not only fewer key strokes and instructions, but also a significant reduction in time.  We now can program the same operations in one and a-half minutes — from 12 minutes to 90 seconds.  This is what we try to look for, refining our processes to meet our needs, and as our needs change, we always have to consider the processes — even established processes.” The same sort of process evaluation is ongoing between Stema and UGT.  Programming technicians at UGT have been working on software routines that will reduce setup times and allow Stema to produce parts quicker.  “From the manufacturing end,” Lorbach says, “we know what we need to make, how to make it and about how long it should take and what it should cost.  The key to reducing time and cost is in the steps required to make parts.  And this is where UGT has been working very closely with us.  They understand the unique nature of our business and are trying to tailor our processes to make us more efficient.” 

The Next Step — and the Next . . .

Lorbach is guarded when it comes to discussing what might be next — the next machine he might acquire, new technology he’s evaluating.  What he will suggest is that he’s always looking at ways to eliminate or simplify steps and operations.  Which means looking at change not only as the routine, but also as an opportunity. He explains that customers today are increasingly unwilling to hold inventory; they want to invest in only what they need, when they need it and expect suppliers to deliver according to their schedules.  But suppliers like Stema can ill afford to build to inventory in anticipation of orders that might be coming.  So they have to build on the fly and commit to just-in-time service.  If not, customers will find suppliers that will (or at least try to) meet their demands. “This is not an easy business and it’s constantly changing,” Lorbach says.  “People don’t want to spend money.  Two or three years ago a customer might order 12 pieces at a time.  Now that’s down to three or four, and they order three or four times more frequently.  Which puts a burden on us to constantly find ways to be more flexible, fast and efficient.  And this drives us to find new ways of doing things.” An interesting aspect to the concept of JIT in its present derivation is that while customers are demanding faster turnaround on smaller orders and immediate attention when they have an emergency, by and large they seem unwilling to pay a premium for what many suppliers deem premium service.  “If you think you’re going to be compensated for JIT service,” Lorbach says, “think again.  Customers expect you to hold the line on price.  They look at what you charged them for the last order and expect to pay that amount, even though that order may have had a week’s leadtime and the one you’ve just completed had a leadtime of six hours.  Charge a premium on rush job?  That’s a nice theory, and it might have worked in the past.  The reality today, however, is that nearly every job is a rush job.  What once was a rush job is now normal; what’s now a rush job once was impossible.”  Which just furthers the argument for investing in new technology and diligently maintaining existing equipment.  “The only way I can attempt to hold the line on price is to find a way to take costs out of my processes,” Lorbach says. “This is why finding better tools and techniques and eliminating steps and simplifying operations