Wednesday, July 28, 2010

Die Cutting Presses / Finding the Right Press

Written By Mark Batson Baril

Developing The Best Cutting Method For A High Volume Product - A Case Study

An engineering team for a large automotive subcontractor needs to develop a best method to produce parts in large quantities. Here are the details of that project and hopefully some answers that will swing them in the right direction.
  • Total yearly volume - ramping from 2 million now to 16.5 million parts per year within 2 ½ years.
  • 20 different but similar images to be cut. Possibly going to 40 within three + years.
  • Images range from 12" x 14" (305mm x 356mm) at their smallest to 17" x 20" (432mm x 508mm) at the largest. Images are rectangular with radius corners. Some of the images have 1 - 3 simple interior cutouts that must be stripped. Tolerances are ±.060" (±1.53mm).
  • Material is .015" (.381mm) Polyurethane
  • Material typically comes 60" (1,524mm) Wide X .015"(.381mm) X long rolls.
  • Raw material and cut parts have an unlimited shelf life.
  • Cutting and finishing operations will take place in Mexico.

A numbers break-down looks like this:

  • 16,500,000 parts per year / 20 images = 825,000 parts per image per year.

  • That's 68,750 parts per month of each image.

  • If the factory works 20 days per month they will need to cut a total of 68,750 parts per day.

    Some of the basic needs include:

  • The process must be fast

  • Easy change-overs / set-ups

  • Material yields must be excellent for as little waste as possible.

  • Cut quality must be good. The final assembly process, after the cut, is forgiving of some quality issues arising out of production compromises.

    And here's where the fun starts! What's the best method to cut these parts? Given the large quantity of parts, the large number of different images, and the material to cut, we should look at three different cutting methods. Flatbed diecutting, rotary diecutting and digital diecutting. The production and pricing numbers are ballpark estimates but are close enough to make a good comparison. For the sake of comparison I have also used an average sized part of 16" x 16" (406mm x 406mm) with an 18" (457mm) repeat in both directions.

    Flatbed Diecutting:
    Because of the roll goods, the very wide material width, the quick change over needs, plus the quantity of images and the easy cutting material, I automatically lean toward a steel rule die type set-up cutting against a hard plastic cutting surface. A belt drive system would allow this fairly floppy and stretchable material to feed well into the press and would allow the final parts to flow off the machine for stripping and/or packaging. Multiple layer feeding and cutting are possible, especially against a specialty belt material. By using a CNC controlled belt feed and cutting head system on some type of wide beam press, fantastic yields can be achieved through the use of the entire 60" (1,524mm) width of material. A production rate of 50 impressions per minute is a conservative enough number to use for comparison. A multiple up tool or full bed beam press may improve the numbers. At this rate one machine would need to run three shifts per day twenty days per month in order to keep up with the volume (50 per minute x 60 minutes per hour x 3 shifts or 24hrs = 72,000 parts). Some pressure could be taken off by adding a second machine or experimenting with multiple layer cutting. My best estimate is that this material could be fed and cut in layers of at least 3 deep, reducing the cutting to one machine on only one shift per day.
      CNC / Steel Rule Die / Belt Feed Diecutting: Finished part tolerances ±.015" (±.381mm) ; Typical tooling cost for a one or two on die that moves with the head $200 - $400 USD; Capital expense for press/feeds $125,000 USD; Belt maintenance/etc…. $10,000 USD yearly; Tool life 100,000 + impressions; Tooling change-over time is 20 minutes.

    Rotary Diecutting:
    Because of the large quantity of parts to cut and the possibility of larger volumes after the initial three years, we must take into consideration the fast process of rotary diecutting.

    Both Soft anvil and Hard Anvil cutting are options. In soft anvil you cut against a hard plastic blanket somewhat the same as the belt talked about in the flatbed diecutting above. In soft anvil cutting an inexpensive steel rule type rotary die is used. In hard anvil cutting you cut against a steel cutting cylinder with a solid steel machined rotary die. Both methods achieve very fast running times ranging from 75 to 150 feet per minute on a project/material like this. The major differences are trade-offs between quality and costs of tooling and machinery. Soft anvil cutting will typically produce a less accurate part than hard anvil. Soft anvil will typically be the least expensive route to take.

    Given all the parameters of this project the following information is applicable.
      Soft Anvil Diecutting: Finished part tolerances ±.060"(±1.52mm) ; Typical tooling cost for 4 - 6 on full width rotary die $1,500.00 USD; Capital expense for press/feeds $125,000 USD; Belt maintenance/etc…. $10,000 USD yearly; Tool life 100,000 + impressions; Tooling change-over time is 20 minutes.

      Hard Anvil Diecutting: Finished part tolerances ±.010"(±.254mm) ; Typical tooling cost $25,000.00 USD in 60" width - approx $3,000 - $6,000 USD in 18" width; Capital expense for press/feeds $300,000 USD in 60" width - $75,000 USD in 18" width; Tool life 1,000,000 + impressions; Tooling change-over time is 20 minutes.

    For both types of rotary cutting, a production rate of 100 feet per minute should be a conservative enough number to use. At this rate, on a 60" wide machine, one machine would need to run one shift per day fourteen days per month in order to keep up with the volume (assuming a 16" x 16" part running three across the web at 100 feet per minute x 60 minutes per hour x 1 shift or 8hrs = 96,000 parts). More pressure could be taken off the machine schedule by experimenting with multiple layer cutting.

    In the hard anvil cutting the 60" width becomes a major hang-up due to the cost of the tooling and the cost of the capital equipment. The tool handling also becomes a factor as these monster tools are heavy! The web width could be reduced in both the machinery and tooling as well as the material but some yield compromises would have to be made and the production volume would be cut proportionally. If you had 16,000,000 of the same part this method would certainly be more attractive.

    Digital Diecutting:
    For this project I am including laser cutting, waterjet cutting and knife cutting within the digital diecutting areas of production. All are quite capable of doing a nice job on this material in the material width, within tight tolerances with excellent edge cutting results. With our average part having 70" of cutting (16" x 4 plus an internal cutout), our yearly volume of 16,500,000 parts would have total cutting inches of 1,155,000,000 (yes that's billion) inches. At 200 inches per minute, a good average for digital cutting, there is 96,250 hours of cutting. One year for one shift is 2,080 hours so we would need 16 machines running 3 shifts to keep up with the volume. Cut at four times that speed with a multiple head machine , or common cut as much as possible and you still need alot of machine time to get through the year. The advantage is that there are no tooling costs and set-up time is just about zero. The disadvantage is that the process can't keep up and be cost effective. Once again we run into the large volume fact that nothing beats diecutting for speed!

    All In All:
    Given all the factors discussed, the best route to take is to pursue both the flatbed diecutting and the soft anvil rotary diecutting. There are several manufacturers that would be willing to run real tests on real machinery in order to qualify the processing speeds and the quality of the cut. The other major factor to be tested on press is stripping of the internal waste pieces. Depending on the size and location of these cutouts, the rotary press may have the advantage over the flatbed process. The total cost for this type of testing should be limited to your supplied material, the applicable tooling, and an agreed upon fair hourly rate.

  • Friday, July 23, 2010

    Membrane Switch Cutting

    Written By Mark Batson Baril

    It’s amazing how many things out there involve specialty cutting.

    Do you own a Microwave Oven? - a Treadmill for exercising? - a flat faced calculator? - a machine with a pressure sensitive operation switch? If you do, then chances are you own and use a Flat/Tactile Membrane Switch (or several) everyday! These switches can be used for everything from the simplest of on/off switches on a blood pressure reader to a complicated multilevel/multitasking switching/control system for a printing press.

    It’s hard to say when the technology came into being because some of the simpler connectors/switches have been cut with dies since the 1950’s. When was electricity invented? Membrane switches really took off in the 1980”s when consumers were demanding lower prices and manufacturers had to push for an alternative to the traditional molded/hard printed circuit boards that were so often used as the base for switches on most machines. Their main attributes are, the relatively low price, their relatively quick turn-around production time, plus they look pretty cool!

    Membrane switches are typically made up of six different layers that are all die cut (with Steel Rule Cutting Dies, or laser cut, router cut, matched metal or rotary diecut, etc... depending on the situation) separately and then assembled. The concept is nearly simple in that as with any electrical switch you are trying to create a space between two wires when the circuit is inactive and you are trying to make them touch when you are connecting or making them active.

    The basic layers are;

      # 1 - Graphic Layer - This layer of thin plastic material is what the user sees and touches. It acts as the guide to show you where to push the switch and it sets the tone for the product and it’s operational instructions via the graphics. As with the other layers of a membrane switch, the graphic layer is silk-screen printed. Sometimes the top/graphic layer is diecut as a final pass once everything has been assembled to it, other times it is cut into it’s shape separately. # 2 - Graphic Adhesive Layer - This layer acts as a two sided glue to bond the graphic layer to the top circuit layer. It’s shape can often times be the most intricate in that there can’t be any adhesive that touches the actual switching area. Each of the areas where there is a button or switch must be cut away. # 3 - Top Circuit Layer - This layer acts as the first half of the electrical connection. Silver Ink is printed on polyester to form the electrical paths. Protection from electrical interference from outside the circuit is stopped by printing conductive ink shields, or applying aluminum foil, metallized mylar or copper foil on the top surface of this top circuit. The feeling of the switch is created in this layer as well. You know that great “Popping” feeling you sometimes get when you press one of these switches? That’s when these switch guys have found your “tactile optimum” a.k.a. “feels good point.” The plastic that this layer is made of is put through a process where a heated mold forms little domes at the areas where you will push. When you push down on this dome you get the feeling that you are actually doing something. I hate those switches when you can’t tell that you have pressed anything! # 4 - Spacer Layer - This is the really ingenious layer! Some designer probably made a fortune on this! This layer creates the space between the two circuit layers. The general shape of the outline is cut as well as holes at the points where you want the switch to activate. When you push the dome/top circuit down it pushes through this spacer layer and makes the connection to the bottom circuit, thus completing the electrical circuit. Some of the feeling of the switch is created in this layer as well. When all of these layers are assembled there is air trapped in the “spaces”. The designers will install more cut-aways, called air-tracks, between the various spacer holes. The movement and resistance of this trapped air, when the dome is pushed by the user, can make it harder or easier to push down depending on how many they plan for and how wide they make them. # 5 - Bottom Circuit Layer - This is where the final electrical connection is passed to from the top circuit. The electrical leads from both the top and the bottom circuit pass through a part of the switch called the “TAIL.” This tail is just an extension of the printed plastics that extends beyond the visible part of the graphic layer and goes to the inner workings of the machine you are controlling. # 6 - Rear Adhesive - This double sided glue layer adheres the completed switch to the surface of the machine/circuit board, or whatever is planned for the tail to go into.

    And that’s it! Of course there are about a million variations of how this can go together. Different plastics, metals, rubbers, etc..., can be used to create different electrical properties, feelings for the switch, etc.... . Backlighting can be created, LED’s, resistors, capacitors, even memory chips, can all be added to a switch of this type. Every manufacturer has their own techniques for not only making the switch work but for making it feel like it should for the user, and work for just about any situation.

    The concept is fairly simple yet when you see either a set of dies, prints or even cut parts laid out in front of you, it can look like a fairly complicated puzzle.

    What’s the future in this type of market? Faster and Cheaper! What else! Many manufacturers are actually producing the cuts for membrane switches with lasers. They can produce one switch or short production runs this way with no die costs and no waiting time for the tools. Diemakers hate to hear that! Digital printers now produce the top/bottom circuit and graphic layers direct from the file without having to produce screens/plates/etc.. .

    I’ll be using the membrane switch on my printer now and then switching my computer off to wait for that next inspiring question to hit my desk. Thanks for reading!

    Friday, July 16, 2010

    Make Ready Patch-Up Techniques

    Written By Mark Batson Baril

    Starting Patch-Up at the Right Point During Make-Ready is important, let's explore the basics....

    On a cutting press, with a new job, at the beginning of a make ready - What % of cutting does a cutting pressman start to patch-up the make-ready sheet to get the most mileage from the die?

    This is a tricky question in that having a die last forever and making a profit on a job are often two very different things. Balancing die costs, press time, run length, and the likelihood of repeating the order in the future can become very complex. In most shops an operator is given a set amount of time that he or she should take in order to make the job ready to run. The shorter this given amount of time tends to get the higher the percentage you talk about in your question tends to be. If your press and make-ready system are set-up well, you will not necessarily have to sacrifice die life for a quick make-ready.

    "Spot up" (patch-up) is the process where a pressman uses tapes or other thickness building devices (paper, metals, etc….) to add thickness to areas of the press and the die that tend to be lower than the rest. On a brand new smaller platen type press or a press with a well made die-set used as the cutting surface - the surface that the die rests against and the cutting surface will often be ideal. This means that if you have a well made tool to put in this type of press you will be able to start your patch-up at about 99%, depending on the material being cut. If you have an old beat up machine that never comes down straight twice, and was made with a cutting surface that has more hills and traps than your favorite golf course, then your patch up may start down in the 30-40% range.

    It is for good reason that there has been much talk about setting up your press permanently with a sheet that levels the footprint (takes out the hills and traps), it works and will save you tons of time and add die life as well. By spending this initial set-up time just once on both old presses and new presses you should be able to bring the percentage of nice even cutting up into the 90% range before you need to start your spot up. Again your perfect press situation must now be matched with a perfect tooling situation and profits will soar! Give us a call if you want to find out more about leveling or footprinting your press.

    So to answer your question - there is no real answer. Every press and every press person will have their own intricacies that need to be dealt with. Starting the patching process when you are just starting to see the first cuts penetrate the material is ideal, you just have to work towards getting as much of the image coming through the material at the same time as you can.

    Thursday, July 8, 2010

    Score Bend Testers

    Written By Mark Batson Baril

    I have recently been given a SCORE BEND TESTER by my superior, and have been instructed to start using it. I have been in the industry for 20 years and have had no need for this device. Can someone please tell me how I go about implementing this into my daily routine, and what are the parameters for it's use. I do about 25 - 30 make-readies on Bobst Diecutters a week. Thanks in advance for any help...

    A Score Bend Tester
    Made by Thwing-Albert Instrument Company

    The Score Bend Tester is a device used to test cartons, after they have been die cut, for their strength at the scores. The main result the tester is there to calculate is how much force it takes to open the carton up, from its flat, ready to fill, condition. There are other testers out there that measure the board strength before it is converted into a carton or before a score is formed, but for this question we are focusing on the testing of the scored board only. There are several machines on the market The ones we have researched cost between $6,500.00 and $10,600.00 USD.

    The main purpose of having and using the Score Bend Tester is to control the quality of machined filled boxes. As companies that use automatic machinery in their packaging lines become more sophisticated, they are demanding equal sophistication from their carton producing vendors. Most of these machines find themselves within the Quality Control and/or testing labs of medium to large sized box shops. Typically, parameters are set-up for how much strength it should take to fold open the carton during the machine filling operation. It is then the job of the carton manufacturer to stay within those parameters. The only way to properly test and document what is actually being produced is to run tests on some type of bend tester. As in any statistical process control situation, every production runs' quality control will vary slightly from one to the other. Many companies will take test measurements at the beginning, middle, and end of the run. Each test sampling will usually have at least ten cartons and again will vary depending on the size of the run, the number up the tool is running, the parameters set-up by the final customer, etc...

    We have learned that the testers are used all the way from the sampling process for new cartons, up through the first article inspections done on press, and on to the final production runs. By using the tester as a guide from start to finish, the manufacturer can get controlled information in order to make educated decisions on everything from paper parameters to tooling specifications. To try to insure maximum speeds in their finishing operations, some companies also use the machines to test the flat diecut cartons throughout the run to insure consistency and conformance with their own gluing departments requirements.

    So, those are the basics of what the machine is typically used for. As far as putting it to use as a regular part of your day to day operation, it would seem that this will be a combined effort between you, your quality control department and your customer. The same combined effort holds true if you are using the machine for extra information for your own production improvements. Instead of including your customer in the mix, just include anyone effected by the bend strength of that scored paperboard. Sounds like you have your work "cut-out" for you.

    Many thanks to the Thwing-Albert Instrument Company, who sells 15 - 20 of these machines worldwide per year, for their pictures and candid information.

    Thursday, July 1, 2010

    Cutting Plates for Flatbed Diecutting

    Written By Mark Batson Baril

    Much attention has been placed on cutting rule height tolerances, degree of bevels, type of edge (ground or shaved) and the abilities of the make-ready artist to achieve the best cut. What should one expect from a steel cutting plate? Is the thickness tolerance of the cutting plate equivalent to today’s steel rule height tolerances? How smooth should the surface be? Under normal conditions, how often should parallelism be checked?

    Wait a minute - did someone say artist? In today’s’ world of computer generated this and computer aided that, can there possibly be room for hand/eye/brain type skills anymore?

    As we have talked about several times in other articles, the diecutting and the diemaking functions are completely interrelated in everything from information gathering to the materials and skills used to connect them. The cutting rule and the cutting plate are the very last parts to make the connection in the process. If all goes well this last connection is a smooth one and the perfect cut or "burst through" is achieved. In the perfect world, with many materials, the rule and the cutting plate actually never make contact with one another, as the final particles of material are actually pushed apart or burst through. In reality the connection can sometimes be a very aggressive strike.

    Over the years, as the industry as a whole has strived to improve upon itself; new methods of manufacture, new materials, improvements in die design, etc.... have been applied to the areas that relate to the question posed. The idea being that if we can improve upon each of the individual, yet related areas, the final process and product will get better.

    The areas that apply to this final connection are;

  • The Cutting Press and the quality of its’ parallelism, whether it be across the entire flat surfaces that the die and the cutting plate are applied or around the cylinder that the jacket is applied to. Wear or poor adjustment will effect die and plate life as well as help to produce a poor product. Often times this is where the artist (press person) is asked to turn a school kids’ finger-painting into a Picasso.

  • The Steel Rule Cutting Die and the quality of its components and design. Today’s best ground edge-hardened rule will achieve a consistent height tolerance of ± .0005* (.0127mm). This is unbelievably close to being perfect! The bevel may drift much more than this from side to side yet this will not effect the height. Designed in balancing knives or blocks that distribute the weight of the presses’ stroke across the entire cutting surface (press, die and plate) are tremendously important to maintaining control over height and pressure. Without these balancing knives the press is forced to rock across the cut, defeating any efforts to take advantage of close tolerance presses, rules or plates. Help, call in the artist!

  • The Cutting Plate - Finally getting around to the meat of the question. Given the perfect situation (press, die, press operator, etc..) you should expect the world from a good cutting plate. Billions of impressions? What are the limits? Anyone out there feel like bragging? (resurfacing doesn’t count). The thickness tolerance of today’s best quality cutting plate will be better than or equal to ± .002" (.0508mm). So to answer the second question; NO, the thickness tolerances are not as close in plates as they are in rule height. In fact, in talking with several companies that build the plates, the feeling is that they may be as close as they are going to get to being perfect.

    Many of todays best cutting plates are built per the following;
      1. The initial blank sheet is ground to a rough state that is close to its’ final thickness. 2. The plate is hardened to the desired Rockwell (usually 47-52 RC). 3. The plate is then further finished by hand to its’ close to perfect state (talk about artists!). 4. The plate then goes through its final machine grinding process to create the desired smoothness and final tolerance requirements. The smoothness of the surface is produced in a number of different ways and each manufacturer will claim that their method is best or most cost effective. Everyone’s cutting situation is different and therefore it is hard to say what is smooth enough.

    There are companies that produce the entire plate with no hand work at all. Who’s are better and why? may be a great future subject.

    Parallelism - What is this anyway? Parallelism is the relationship that the top surface of the plate has with the bottom surface of the plate. The best situation is to have a plate who’s’ surfaces have no deviation from parallel (a parallel plate). An un-parallel plate means that your plates’ overall shape is that of a wedge, or perhaps several wedges. Cutting plate parallelism with a deviation of .0015" (.0381mm) is common. Although we have not been able to develope a hard fast answer for how often parallelism should be checked, the basic principals of process control would dictate that each companys’ maintenance routines will vary and will also be constantly changing as newer and better technologies are applied.

    In many cases a soft "under the cutting plate" material help adjust differences in die/cutting plate heights and deviations and take a step towards eliminating the need for press foot printing at all. Expensive on a short term basis, yet one of the cheapest products on the market if looked at from a long term cost standpoint.

    All in all, a great subject that goes to the very core of the diecutting process!