I was the engineering manager at Charmilles Technology Manufacturing Corp. (CTMC) in Owosso, Mich., where we produced a wide range of electrical discharge machining machine tools that included both wire-cutting and die-sinking models.
The Roboform 40 and 41 series of die-sinking machines had a long-standing problem meeting one of the final quality control checks during machine runoff at the plant. All linear axes of the machine were required to meet pitch, yaw, and roll angular specifications of 8 arc-seconds. The only problem was with the Y-axis pitch, which often required reworking (or “tweaking”) to get within the 8 arc-second specification.
The machine was comprised of four basic grey cast-iron machine elements. The L-shaped base incorporated a large work table and mounting surfaces for the X-axis linear rails on the top. The saddle was guided on the X-axis rails, and also mounted linear bearings perpendicular to the X-axis for the Y-axis guide rails.
The Y-axis had the linear rails mounted on the bottom of the box section casting, and the linear bearings for the Z-axis on the front of the casting. The Z-axis was another heavy box section casting with the linear rails mounted to precision ground surfaces on the rear of the axis. At the full 400mm extent of Y-axis travel, a large overhung load was created on the machine base and X-axis saddle.
Preliminary calculations on section modulus and linear bearing stiffness convinced me that we were experiencing a combination of structural bending and bearing deflection, which produced the angular deflection. It didn’t appear that redesigning the structure would be cost-effective, and we were already using the stiffest roller-bearing linear guides commercially available. Other ideas to solve the problem, such as counterbalancing the overhung load, would likely also add unwanted cost.
Given the bearing spacing, I calculated that a vertical linear correction of 13 microns at the bearing supports would compensate for the 8 arc-seconds of angular deflection at the full 400mm travel. However, I could not find a practical way of producing this motion, particularly since the full 13 micron correction was only needed at the full extent of Y-axis travel, and the amount of correction would need to be proportional to the extent of axis travel.
Jeff, I totally agree with you. A one time permanent fix as opposed to constant tweaking is always preferable. By the way tolerances are given for a reason. Once you have solved the problem to consistantly be, not only within tolerance, but exceeding by a factor of 4, any further messing around is a waste of time and money. Well done kyoshi.
Thank you for the compliment. It did work out as almost the ideal engineering solution, solving a long-standing product issue with no additional cost other than the initial programming of the Blohm grinder. The sensor based approach would be a fascinating project to try for an order of magnitude improvement over current best practice for "standard" machine tools.
As a control engineer, I am fascinated by the measurement approaches suggested. As a mechanical/project engineer, I am on the side that says if you can grind in the compensation on each part and you get what you want in terms of accuracy, you should. Any kind of sensor can eventually fail, connections fail, and sensors always needs to communicate with the controller. If you don't have those components in your design, you won't have the failure mode, you won't have to put them on the BOM, inventory or buy them. I think this is a great solution.
I appreciate your input. With the wide range of sensor technology available today in conjunction with inexpensive computational power a self-correction machine tool could obtain incredible levels of accuracy.
If you were using an angular sensor with a reference to the work surface you could make linear corrections to the position to compensate for the positioning error introduced by the angular motion of the structure. The linear correction of 5 ppm per arc second would be dependant on the the distance from the linear measurement reference for the Y-axis. Angular motions would still be introduced at the tool, possibly creating inaccuracy when large die sinking tools are used.
The machine in question is a relatively rigid structure, the angular deflections were extremely repeatable under a range of operation conditions. Real time measurements would be more expensive (and less effective) than mechanically compensating for the underlying structural deflections.
I agree with you, but inorder to make corrections and ajustments you need to make measurements. This is a real time sensor, so while you see what are the errors in position, you can make correction.
The workpiece is attached to the table, the diesinking form tool is attached to the bottom of the Z-axis. Angular errors, positioning errors, straightness and squareness errors will all produce unwanted deviations from "perfect" in the finished workpiece. Improving the machine accuracy will improve the accuracy of the workpiece, measuring the errors will only tell you how much error you are introducing into the workpiece.
If I understand it correctly, you have one reference surface and then the other that changes its orientation. You can use a unit called EZ-TILT-5000, that reads two sensors and can provide you with individual angular orientation of each sensor and also the differencial. This way you can always see the actual deviation of the required surface in reference to the standard.
It looks like a great sensor, but I am not sure how you would compensate for the angular error in this example. The angular machine deflection is referenced to the machine table, and a mechanical compensation is necessary to maintali Z-axis squareness to the machine table (and workpiece).
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 3
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
A quick look into the merger of two powerhouse 3D printing OEMs and the new leader in rapid prototyping solutions, Stratasys. The industrial revolution is now led by 3D printing and engineers are given the opportunity to fully maximize their design capabilities, reduce their time-to-market and functionally test prototypes cheaper, faster and easier. Bruce Bradshaw, Director of Marketing in North America, will explore the large product offering and variety of materials that will help CAD designers articulate their product design with actual, physical prototypes. This broadcast will dive deep into technical information including application specific stories from real world customers and their experiences with 3D printing. 3D Printing is
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
12 
