Develop new molds suitable for production in the h

How can a set of injection molding tools realize the transformation from design to production? Most companies produce corresponding molds from CAD models. The mold is installed on the injection molding machine, some kind of plastic is injected into the mold, and people eagerly gather to see the produced injection molded parts. The parts look good, but obviously there is still much room for improvement. The injection molding engineer intervenes and uses his/her professional skills to seek a compromise between the best molding parameters. After the mold trial - including the basic test, you will get some pretty good parts - at least it looks good

exploration - in the laboratory for a few weeks

maybe you are counting your efforts to produce components of drain valves made of plastics like this, and using the experimental design method (DOE) to focus on the setting of the machine. To this end, you organize your common think tank, pay attention to the key variables, create a DOE and implement it by your injection molding engineer. As a result, many parts will be produced - each produced by different parameters, so each part has a certain degree of change compared to other parts. The question is how much and where they have changed

for most of you, the necessary tools to discover these changes are installed in your materials and quality laboratory. You may use contact and non-contact coordinate measuring machine (CMM) programs to check the features you can see and touch. You can conduct some destructive experiments through casting, slicing and final polishing, so that you can observe the "hidden" characteristics inside the material with microscope, optical comparator, CMM and manual tools. You may also use X-ray inspection to measure the number of cavities (hollows) in plastic parts and observe the change of wall thickness

the common complaint about destructive testing (casting/polishing, etc.) is that the root problem of the laboratory is that the process of processing components is usually "destructive"

depending on the complexity of parts/tools, the discovery stage may take 3 weeks ④ the pointer is loose; Time. If a set of expensive injection molding equipment has to wait for the experimental results, or worse, it is being used in the production of components before the results come out, people can't help asking whether there is a better way

a possible solution

you may have heard the term cat (computer-assisted tomography) and experienced scanning in a hospital. This technology uses X-rays to pass through your body and generate a three-dimensional image of the part you want to observe. The control accuracy and quality are greatly improved. On the contrary, there is a similar technology in the industry called CT (computerized X-ray tomography). The difference between the two lies in the energy of X-ray, machine parameters and reconstruction methods/algorithms

for those who try to scan plastic parts with CT machines, you may have seen their eye-catching pictures (especially the scanning with the newer generation of CT machines). Compared with traditional measuring instruments, these systems are faster and do not require complex fixtures, so the deformation of components caused by clamping force is eliminated

metroom system

received information from the Zeiss development laboratory that a new generation of CT measurement technology devices have been in mass production. It is quite innovative, so the company created a new category called "metrotomography" to describe this new technology. Metroom system is designed by the best business areas of Zeiss Engineering Department, including technologies from our high-performance CMM, such as guide grooves, drive systems, controllers, rotating meters and Calypso measurement software, which is the leading in the industrial field, which gives the system:

◆ most importantly, it has good accuracy - for typical plastic parts, its measurement accuracy can reach 9 microns

◆ the measurement results have excellent repeatability - Metro system will pass the metrological repeatability & reproducibility (gr&r) test

◆ the software used can evaluate ANSI y14.5/iso gd&t graphic standards

◆ it can scan the parts within 60 minutes - if the parts are small enough (medical plastic parts), it can scan more than one part of China's 5 mines with rich business experience in the industry each time

◆ the scanning is non-destructive, and you can retrieve your parts after scanning

typical applications

using x-rays, the internal and external structures of components can be easily visualized. By scanning the whole workpiece, because China's extruder products are closely connected with strategic new industries, all the information about your workpiece evaluation can be obtained. For example, this information can be used for:

◆ virtual real comparison between your part and CAD model

◆ defect analysis: check porosity and purity

◆ assembly inspection: calibration of assembly parts in your workpiece

◆ reverse engineering: create STL and CAD data successively

◆ mold correction with virtual and real data

◆ dimensional measurement of components

process - an example

part is placed on a rotating table and generates complete three-dimensional data after a full rotation (minutes). As shown in Figure 1, CT datasets (right) and CAD models (left) initially have different origins and may have different positioning

figure 1

different from a typical industrial CT system, there is no data processing and scaling in this system. Anyway - Metro system generates a 9+l/50 micron mpee (measurement error) data set, which includes every place inside and outside the component. Data mpee is the acronym of "maximum allowable deviation" in ISO metrology industry, and is used to describe the traceable system measurement error (accuracy)

the next step is for Calypso software to compare CAD model data sets using a specially drawn data reference frame (DRF) (Figure 2). Once this unique visual indication of CAD model overlap is corrected, the CT dataset is the "boundary" of the CAD surface, which reveals the basic geometry of the component. This realistic software tool can help people quickly determine the characteristics of small plastic parts without molding, and these plastic parts may often be ignored because of their size

figure 2

calypso software does not use stereolithography (STL) format files for measurement - the general STL format has been developed for visualization applications, but it cannot accurately describe the geometric dimensions of components. On the contrary, Calypso software directly measures components from the original CT gray model

once the data set is compared with the CAD model, the analysis can begin. The very practical and fast first step is to start with visualizing (through color changes) how all internal and external surfaces are relative to the CAD model (Figure 3). It can even observe the more in-depth position inside the component, and software tools can study the component from any direction. The cutting tool can also be used to reveal the voids contained in the material

figure 3

now remember that the scanning and analysis of components can be completed in a few hours. We say that the color map is helpful for zeroing in injection molding, but what you really need is a complete design drawing report. Using Calypso software, which provides tools to drill down more deeply to develop a measurement scheme (offline), for example, plastic parts of such fluid export valves. With this powerful geometric dimension and tolerance (gd&t) calculation engine, Calypso software can process all your illustration numbers and report the corresponding data in an infinite format. Figure 5 shows an output report in Excel format, while Figure 6 shows a balloon chart with arrows pointing to the features under consideration. It is not the color that is used as the fundamental method to show the condition - within or outside the tolerance, but it is just easy to be used to show the percentage of the tolerance used, etc

Figure 4

Figure 5

Figure 6

metrography: several points to consider

accuracy: resolution and accuracy are not the same thing - it is very important, but as an example, high resolution can never compensate for low-precision electromechanical design. The bottom line is that the system is intended to be used where the measurement results must meet known reference standards. You can ask the manufacturer to explain to you the accuracy of their X-ray tomography system

reproducibility: if you are evaluating a new measurement system like metro, take some time to confirm that the results given by the measurement system are reproducible and not affected by the operator. One of the gold standards for judging system performance is the test of measurement repeatability and reproducibility (gr&r)

a good empirical method is that your measurement results must be repeated to 3% of the feature tolerance and reach 10% of gr&r. Performance testing is undoubtedly time-consuming - sometimes it takes 3 to 4 days to complete a complete gr&r. However, it is an excellent technology to find measurement problems and ultimately prevent critical errors

speed: if you want to measure, you must do it quickly - remember that your goal is to develop new molds in one day - not a week or more. Metro system can complete the scanning of plastic parts (including data reorganization) within one hour

software: without software tools that can provide useful results, even a fast, accurate and repeatable system is useless for measuring targets. The surveying software must be able to output any CAD file, have a strong gd&t calculation engine, provide a fast way to quickly create and measure schemes, and most importantly, measure the original CT data set

final idea: when it reaches the digital image, don't just believe what your eyes see, you need to use other information - especially if you want to use data to make important decisions. Although powerful image processing requires powerful processing algorithms, it only needs to load a two-dimensional image into any image software to perceive the ability of image processing data. By changing the contrast only, you can generate very different pictures and get very different results. The ability of software to display clear images is a good starting point, but if you want to measure or reverse engineer from data, this is only the beginning. (end)