The post Attained CMSE – Certified Machine Safety Expert Qualifications (TUV Nord) appeared first on Edge Robotics and Automation.

Understanding the benefits and how to factor into your operation to extract these is key to achieving a highly valuable outcome or just an expensive white elephant.

Capturing Value with Industrial Vision System

To capture returns from a vision system you need to understand how these systems add value and offer efficiency to your operation. In this series of articles covering machine visions systems we will explain what these systems can offer, and how to correctly consider for your operations.

We briefly mentioned the guide to mastering vision system in our Dec article of NZ manufacturer, and will expand on some of the deeper technical capabilities in a later article. Our technology page covering machine vision can also serve as reference.

Industrial Vision system can be used across a range of industries and purposes, in general covering such as presence detection, guidance, gauging/measurement, general data collection, and the slant for this article quality control. The benefits from industrial vision systems roughly fall into three categories.

Vision systems offer the efficient use of labour, and by removing choke points, the ability to consistently maximise equipment throughput. Manual QA is not a highly skilled task; but a role neither the less that requires training and experience to achieve consistency and speed. This learning dead-time poses a challenge when it comes to increasing staff for scaling operations, and is particularly problematic if you rely upon casual or seasonal labour. Even with experienced staff, human inspection is a common bottleneck and one that will rapidly fall apart if pushed beyond the limit.

A vision system will significantly faster perform a detailed consistent measurement and do so independent upon the product quality. The first point is easily comprehended but being independent upon product quality is normally a bigger benefit to local operations. The rate of human inspection is highly dependant upon the quality of items; bad products require additional time to check and therefore your production lines slow down.

Primary or secondary industries have time/batch variability and therefore the production rate and/or staff count will also be variable to match. To handle this variability either staff are kept in reserve wasting money; or production needs to be below full capacity to provide a safety margin. A vision system does not care about the quality and runs at constant rate. Taking that a step further, if the next bottleneck is post the vision system, then production output can be keep consistent and the only variability is the amount of rejected product. Production managers of manual lines can only dream of that capability.

Value Recovery Opportunities from Vision Systems

Labour efficiency is the leading driver for approving the investment in a vision system; at least in a company’s first step. In practice, and in all applications, I have ever encountered, the extra value recovered and reduced waste, dwarfs the labour saving.

Initial gains upon installing vision system will stem from the improved quality assurance and reducing leakage of waste throughput manufacturing process. Product recalls are minimised by 24/7 consistent electronic QA identifying faulty products prior to shipping. If recalls do happen then there is detailed traceability and records for targeted action. Mid-process faults are identified early reducing resource waste and production capacity on unrecoverable units. Catching these faults early also reduces the possibility of equipment damage and lengthy downtime.

The big transformational changes to efficiency and profitability come when the vision system moves beyond a simple labour aid. Consistent accuracy and real-time visibility provide confidence to reduce safety margins within manufacturing operations. These tolerances being in the form off oversupply and commonly known as “give away”. World-class manufactures will also feed this back to engineering teams for continuous improvement gains by becoming more aggressive at the design stage.

Another rarely considered benefit is that the value of your products can increase purely from the tighter QA control. This happens in many primary and secondary industries where wholesalers or retailers will pay a premium for product that has been subjected to state-of-the-art QA. You need to be aware on the other side of the fence is your products may suddenly be locked out of certain markets or penalised if you don’t use these systems. This has been widespread practice for many years in manufacturing and horticulture; and now some opportunities and concern for others within agri and aqua-culture related industries.

Operational Flexabilty from Vision System Integration

Vision systems offer the ideal solution for flexible low volume, high variability production. Vision system allow intelligence and guidance behaviour in a production line reducing the need for complicated mechanisms and bespoke engineering. Real-time intelligence is one of the key attributes that defines the leap between mechanised and automated systems.

Conventional production lines have numerous custom mechanisms tailored to each production part. These custom structures incur considerable design and build expense to cover bespoke machine components. In high mix and flexible manufacturing this poses a challenge since the smaller production volume of each specific does not outright pay for machinery; and therefore, machinery needs to be multi-purpose. Multi-purposes machinery becomes incredibly expensive with custom jigs and fixtures for each part, and a swath of adjustments that can be cost prohibited to automate or time-consuming and troublesome for manual change over.

Many of those involved with such machinery are indeed highly skilled and experienced; but having multiple one-off components also means issues with reliability, maintenance, and service headaches over the equipment’s lifetime.
Vision system technology will not be the correct choice in all applications, but it allows production lines to move toward generic standard equipment and more modular future proof layouts.

Instead of engineering specific fixtures to align parts, the parts can instead be identified on a simple conveyor belt and guidance sent a robot; plus you could integrate QA. Rather than designing and building new jigs, possibly reconfiguring the line for new production items; some trial parts could be quickly tested, and the vision system given an additional program. Change over of production runs becomes simpler and quicker with reduced physical changes. Manual mistakes can be avoided with automatic switching of vision programs connected to MES systems (Manufacturing execution system). When your integrator is highly skilled they can also have the vision system check for incorrect manual configuration at run change; many recalls could be avoided if this is implemented correctly.

Manual Quality Control

Before diving deeper into how vision system capture value, firstly consider human based quality control approaches and the challenges faced.

Sample Based Quality Assurance

A common method of manual quality control is using a sampled based approach were 0.1, 1 or even 10% of products are evaluated. This limits the cost of the quality control; however, it has one major drawback in that it assumes failures are random.

In practice anybody away from the accountancy office knows that is not how failures present themselves. Often there is a intermittent malfunction, temperature fluctuation, or a batch of poor materials somewhere in the production line and you get a chain of faulty products. Sampling for quality control you run into two problems. Small, medium, or intermittent faults are often never detected until delivered to the end customer. On larger faults you may get lucky and happened to find the failure quickly, but most times a significant amount of production has been wasted before the fault is identified.

100% Inspection

Automation as this article discusses has opened many opportunities for 100% inline section, however, historically two types of products have been manual inspected to this extend.

• Complete inspection is common on high value products or products that cannot afford failures such as healthcare/aerospace. The product value or competition equalling regulatory requirements in these industries therefore supports the expensive quality control process.
• The other application is relating to handling highly variable natural products in both the primary and secondary industries, both common in NZ and Australia. Having natural products means the defect rate in such industries is so high it is a necessity to inspection each and every product. This creates a challenge for developed countries since the material unit cost is often low and therefore the human labour component becomes a significant portion. In a global market that sets the product value, the added labour cost cannot be recovered, and this is a straight loss of margin.

Human Inconsistency

The other downside to human quality control is the consistency of the inspection; this is particularly troublesome in our primary and secondary industries. In these industries quality control is often subjective and makes the scope for inconsistency of human judgment even wider.

As someone who has spend over a decade on highly advanced, and many patented, vision systems the above is situation that plays out time and time again. When a vision system is installed, you often end up in a discussion with the end users operations team how the machine results are differing from the human quality control. This is actually when the lights start to fully go on for the operations team.

During trials and commissioning you constantly run blind tests, and unknown to the manual QA team give them the same parts numerous times and compare to the machine results. So part you are saying is graded incorrectly, has been through your QA various times with various gradings yet the machine has instead run consistently. The cut point of the machine may need fine tuning, but that is often the first time that operation has had extensive of repeatability, and the human variation has just be laid bare.

Vision System Opportunities

A number of levels of automated machine vision system are out there; on the simplest end of the spectrum you have barcode and colour chart readers on packaging, gauging for checking bottle tops, counting and confirming items such as blister packs, and scoop detection in infant formula or washing power etc.

It gets more complicated when you go beyond the ordinary and towards the primary and secondary type industries common to NZ and Australia. There is a vast array of opportunities in these industries that are not captured. As you go from the run of the mill application such as gauging bottle caps, the vision expertise becomes critical and it almost seems like a dark art unless you totally understand optical physics, in-depth knowledge of equipment, and computer science. Even advanced industrial automation teams struggle with technical knowledge to solve primary and secondary industry challenges, or academic and research institute tend to have the technical knowledge but not the commercial nous to deliver pragmatic solutions outside the lab.

The Secret of Machine Vision System

The most important factor that needs to acknowledge in your overall operation is there is always a grey area to vision systems or any form of measurement. This is true for both manual and automated, and it is there even you do not know, or don’t want to know about it.

This factor is that is often met with trepidation, but this is the critical element to be understood by all if you want to be world class. Sticking your head in the sand or pretending it does not exist and your project could be doomed, and at best you’re leaving a competitive edge on the table.

The value provided by a vision system is driven by that grey area of measurement. How large that grey area is, what factors can influence that grey area, and most importantly how to tailor your operations to work in with that grey area.
Understandably many want the world to be black and white; but understanding that grey region lets the system operate much more aggressively capturing more value, and the visibility to correctly focus improvements.

The post Vision System Benefits, Outcomes, and Your Hidden Opportunities appeared first on Edge Robotics and Automation.

The immediate question typically asked from end users is how they go about using robot welding systems and skills they will require. As mentioned in our prior article is how technology and usability has significantly improved since the early applications days. This article will expand on the user skillsets and programming tools for modern robot welding.

Your Skillset for Robot Welding Systems

Beyond the well-known benefit of reducing labour, the other key aspect is helping scale operations. Robot welding offers the opportunity to maximise the outcome from your existing welding expertise. Moreso than just labour saving, this is becoming particularly important as qualified welders are becoming a scarce resource.

Already having operators with experience of robot welding would be best prior before install, but its not critical. One secret it is not robot programmers but experienced welders who make the best programmers of robot welding cells. It takes years to gain the knowledge and experience to become expert welder; and required robot skills for welding can be learnt in weeks.

The complicated robot element is during the install and handled by your integrator. By the time the robot welding system is configured and setup for production, the robot skills are on the very simple end of the spectrum. Properly trained and support by your integrator, and some homework, within a few days for getting feet wet and within several weeks becoming self-sufficient.

Were one factor often not considered for manufactures is effectively productising and scaling the use of your welding expertise. Whereas manual welding requires expert hands-on involvement for each part; pair your welder with a robot and their expertise can them be applied to hundreds or thousands of parts from the initial effort

The difference between a 5 year experience and 10+ year experienced welder is not motor skills but the knowledge how to get the best quality, speed and solutions to problems. Robotic welding allows them to get away from dirty and monotonous aspects and into professional role based around this knowledge. There will always those who prefer the hands-on work but having access and learning the robot element offers greater opportunity for experienced welders.
Give the welder a robot and not only does it remove the need to spend their day in the dirty, hot, and hazardous environment; but to have that time being mentally engaged and challenged. The welder then spends their time deciding how each part should be made, different techniques, current settings, speed, stick out etc.

In the next leap for manufacturers that concentrated knowledge effort with the consistent quality control helps continuous improvement and lean manufacturing. With a consistent quality of parts your welding operations and production engineering group can then work towards gradual improvements whether that is higher throughput, less warpage, less materials, better seam quality etc. This is both beneficial to your business overall bottom line and the daily role of your welders meaning you hang on to them longer.

Programming Tools for Robot Welding Systems

The original means of programming welding robots, and in general all industrial robots involves physically driving the robot to each location and “teaching” that point, and thus the name given to a robot’s handheld control is a “teach pendant.” This was and is still regularly used in many applications and is valid approach, you are after all hands-on teaching the robot to the real part.

There are two main downsides to this approach for welding applications.

1. Firstly, this is very tedious and takes the programmer a very long time to program each part. I have met robot programmers who have spent a week teaching a single part for heavy industries.
2. The second, and far more costly, is welding production is halted during part teaching. The general rule of thumb is a robot has the rate of three manual welders, if not more when considering external axes repositing parts.

All in all, each hour of robot stoppage runs well into the order of hundreds of dollars of lost production plus the time of your robot programmer.

This manual approach is for why for years robotic welding was restricted to large mass-producing markets such as automotive. The overhead in capital cost, and overhead per job has made robotic welding into our smaller manufacturer a long time coming.

Offline Robotic Programming

Offline programming can be achieved by a range software and simulation packages that allow a robot to programmed on a computer and allow the physical unit to remain in production. These packages offer simplified programming and significantly reduce overheads for each new part.

ABB Arc Welding Package and Tools, RoboGuide

As CAM packages have greatly improved the efficiency of translating CAD to CNC programs; offline robotic programming has aided the development of robot programs.

In addition to the programmer no longer having to climb all around a robot cell, there are number of productivity enhancements. These enhancements allow the programmer to focus on the welding application rather than minute details of robot motion and detailed programming commands

Welding robot’s have been around since the 1980s, so while there may be the odd occasion a programmer needs to add something specific, most of a programmer’s time can now actually be spent picking seams, the correct torch orientation, stick out etc.

This also leads to reinforce the earlier point from Part 1, the best robotic welding programmers, are WELDERS!

Compared to traditional methods, offline programming, and especially when including Add-On welding packages, can easily reducing programming time 50% and much more for large and complicated parts. The big benefit, however, is the robot remains in production earning money while the new part is being programmed. Extrapolate out these differences even conservatively, and the part overhead costs have been reduced by 80%, likely over 90%.

Offline Robotic Programming Software

Offline robotic programming software can be split into two “types” of software; visualisers, and simulators.

• Simulation programs are from the robot manufacturers themselves, and tend to be more functional software tool that also have several background robotic and general production aids.
• Visualisation programs tend to be third party programs that are generic across various robot manufacturers and various stages of a systems total lifecycle.

The true simulation environment would seem the obvious choice except most robot manufacturers really only care about selling robot, and provide a very poor-quality solution, and thus the rise of the third-party solutions. The third-party solutions tend to heavily market towards usability, aesthetics, and “program” commonality between multiple robot brands.

The other key use for third party components, is more towards engineering companies, integrators as sales/concepting environments. These programs are great for putting early detail concepts together, considering overall product flow in large systems.

There is creditability and also hyperbole aspects to the third party marketing, so you need to consider your application.

Selecting Robotic Programming Software

Consider the following points for selecting a 3rd party offering over the native solution by from the manufacture of your robotic welding systems.

1. Have multiple robot manufacturers.
2. Have complicated parts to program
3. Have robots other than ABB/FANUC
4. Happy to accept less than ideal results.

The first two elements are tied into what a third party robotic programming solution may offer. The third party solutions from the get go are focused on the user; whereas all the manufacturer solutions were, originally at least, an technical tool to operate the robot.

Using across multiple manufacturers is commonly pushed by 3rd party offerings; but it really depends upon the scale of your operations, and how many and of which robot manufactures. In reality you will still need to learn each manufacturers system and terminology. The first thing you’ll do with a ABB or Kuka robot for example is install ABB’s Robot Studio or Kuka’s Works Visual.

The third party solutions will be more intuitive to an untrained person, but the reality is whoever operates and/or programs a robot welding cell will need at least some training; and this no different to any professional tool. There is certainly some benefit long term to the easier use, but remember there are more factors to consider than simply selecting edges in a 3D model. In most cases you’ll certainly prototype a part program faster with third party option, but by the time program has been optimised the options will start to even up.

1. Contract manufactures will offer the opportunity to achieve the most gains from a suitable 3rd party solution; these being one-off part runs that put a premium on minimising initial setup and less on complete refinement.
2. Internal manufacturers should strongly consider the native solutions for extra refinement options and long-term continuous improvement and product updates over a number of years.

ABB and FANUC have been singled out because both manufactures “get it” and provide good quality tools and documentation, other manufacturers can be diabolical. The equipment from ABB and FANUC is top notch; but our preference for these brands is because of the operational factors over a systems lifetime. Not only are these the sole manufacturers with factory support in NZ, the tools and documentation are what you care about once your system becomes a reality.

Some of the very high-end third-party options do let you develop a “program” a bit faster than ABB Robot Studio or FANUC Roboguide; but there is more to developing a program than just the motion of the welding torch. These offer tighter integration to the overall robot welding system, production tools, and less room for human error. In both ABB and FANUC’s cases both provide special welding software packages that are both very reasonable to use.

Note of Caution

The third-party solutions could be a great addition to your operation, but the major drawback you really need to be aware of, but you’ll never see in the marketing material or controlled demonstrations; real life accuracy.
You need to carefully consider what that risk could mean because crashing robots is an expensive mistake.

When using third-party tools, Before operating on the real robot I have always verified operation first with the native solution; and the native solution is the program used during commissioning. The way to think about this, third party visualisers will show what it should do, the manufacturer’s simulators will show you what the robot will do. (within reason)

The virtual robot(s) in ABB Robot Studio or FANUC Robot-guide use the same internal code as the physical robot and the simulation is an actual representation of what the real robot will do. (Not all manufacturers do this) In general, third party visualisers will program the torch positions adequately; but be questionable when detailing the full robot arm and exact path behaviour between weld seams. Experience using some of the larger well-known third-party tools has shown noticeable deviations with more complicated or optimised motion.

There is a certain amount of black magic in the brains of industrial robots. There are decades of work covering the complicated mathematics, practical approximations, and thousands of configuration settings. In one complicated project featuring multiple large 300kg payload robots with 50+ crossing locations with inter-robot speed of 80 km/h; it was only possible using ABB Robot Studio.

The post How to Use Modern Robot Welding Systems appeared first on Edge Robotics and Automation.

This is the first in a series of robotic welding articles that will demonstrate the state-of-the-art technology and how advancements in usability, and Industry 4.0 concepts, offer many more opportunities for our local industry.

State of the Art Robotic Welding

Industry figures put welding applications at about 25-30% of all global robot sales, and this significant market drives targeted R&D by the major players. Technology in all manner, including capital outlay, capability, and ease of use, has dramatically improved since the early days of robotic welding.

The days of one robot being fixed to welding the number 4 engine bracket on the jeep wrangler disappeared decades ago.

Technology advancements have brought increased competition and lowered the cost of equipment. Incredibly intelligent modern welding power sources, robotic sensors, and vision systems now allow complex welding challenges to be automated with off-the-shelf equipment. These smart features have made robotic welding operations more robust and handle wider variability, allowing the systems to work longer and have fewer stoppages with minimal operator input.

Another avenue of this development race has been the simplification and de-experting of the use of robotic welding systems. Improvements in offline robotic programming and dedicated robotic welding software have simplified the necessary technology skills. These user-ability improvements are game changers for local businesses and can offer a massive 90%+ reduction in the cost of programming parts. Combining this increased part programming efficiency with the digital factory from Industry 4.0 allows high-mix, low-volume production.

Robotic Welding Opportunities for NZ Industry

The technology convergence opens many automation options for smaller niche operations in NZ and Australia. In local welding operations, this significantly reduces the two long-standing challenges of returning a payback on a robotic system.

1. Improved programming tools reduce the long-standing problem of returning a payback when faced with programming a small volume of parts.
2. Industry 4 connected systems allow the easier, possibly fully automated, part changeover for high-mix operations.

This provides the robotic welding investment opportunity for SMEs to increase their competitiveness and manageable ways of scaling capacity. These improved user-ability factors, and others on the horizon, are critical factors for many NZ manufacturers and will be discussed in later articles.

Robotic Welding Benefits

Weighing up the case for robotic welding requires gathering a comprehensive picture of your operation and exploring potential opportunities for indirect gains. This is a simple equation for large-scale manufacturing operations with double or triple shifts. In these operations, a robot welding system can produce the production equivalent of 7-10 people. Combine that output with the starting price of a quality robot cell in the range of 120K NZD, and it makes sense to measure the payback in mere weeks. Unfortunately, such local applications are rare and therefore require careful analysis of your current and potential operation.

Manual welding is a fairly simple calculation, at least in the direct costs. You have some relatively minor capital investment and some standard overhead adjustments, but by in large, the majority is labour tied directly to each item. Automation of all types is the complete opposite, with the production cost predominantly made up of capital and order overheads; all grey areas to how exactly these will play out.

Reduction in Labour

The general rule is a robotic MIG/MAG system will lay filler at about 3x the rate of a person. Larger or more complex robotic welding systems also often have part positioners or rail-mounted robots for additional reach and maintaining optimal torch angle. In addition to welding outcomes, these also allow a robot to move around a large part much faster than a person moving around a bay and in some cases, the output could exceed 5x the manual rate.

Scalability

The increased rate from a robot allows a higher output with your existing facility and resources. Skilled welders are getting harder and harder to recruit. In the extreme case installing a robot allows you to increase your welding capacity by the same as finding another 7-10 welders; even on a smaller single-shift SME scale, you’re adding the production capacity of at least three welders.

The higher production from a robot further means an increased output from your existing floor space. A common sales pitch is you can make the cells very compact, but it does make the cell more complex to operate and program, so avoid this if possible. Keep the cell to a reasonable size, making its operation simpler yet still capturing a good production rate for the given area. If you are pressed for floor space and are manufacturing towards large production or even repeat high-mix production runs, then you lean on your integrator to safely handle the additional complexity with ease.

Improved Repeatability, Consistency and Quality

The quality and consistency of the welds from robotic automation is one of the golden benefits that many companies are already well aware. Multiple facets lead to this quality; robotic welding power sources from repeatable brands are nowadays highly advanced, with complex current and feed control features offering a very high degree of heat, penetration, and filler control. The other side is the robot motion is incredibly accurate, smooth and repeatable; in essence, the perfect set of hands 24/7.

Reduced or Complete Removal of Downstream Processes

In a number of cases, the quality difference is not just reducing tidy-up but the seam quality improving such that they can be left untouched as cosmetic features to customers. 

Reduced Wastage

Improving the quality of the welding operation leads to reduced scrap factors from welding processes or, worse, found later in final QA checks. This includes not only the scrapped parts but the associated labour, consumables, and production utilisation to make that scraped part.

Hazard Reduction

The welding environment is not a clean and safe environment for manual welders. There are immediate hazards from the heat, fumes, and arc-flash, as well as the generally just dirty and unpleasant working space. Long-term damage is known from carcinogenic fumes, reduced breathing capacity, eye damage and exposure to hazardous chemicals. Pair an experienced welder with multiple robotic welders, and you remove the person from that environment.

Robotic Welding Considerations

There are a number of points to consider if your operations are reading for robotic welding.

Is there sufficient work and growth in your market to capture the productivity gains

The production rate and the continuous operation of the system is what will provide the rewards. Manual welders cannot match the production rate of the robotic welding system, but you still need to have sufficient work for the robotic welding system to provide a return. The ROI on production numbers may look great, but if the robot finishes its days’ work in two hours, you probably need to find more work to feed it. One benefit this can, however, provide is high production capacity sitting in reserve for large production runs. So, if you can pay the robot off with single shift operation great, and if you get a spike or large growth, you get the 2^nd or 3^rd shift for free.

Incoming Quality Control

One advantage still in the hands of manual welders is the ability to adapt to tolerance stacks and poor fitment. If you’re having fitment issues prior to robotic welding, you need to fix those issues first. The general rule is you want the fitment to be within 1mm, ideally less for the robot to work flawlessly without any special functions, and certainly less for thin gauge applications.

Robotic welding systems often struggle with poor fitment and varying dimensions. While dimensional tolerance can be handled with robotic systems, it will increase equipment cost or cycle time. In addition to real-time welding guidance, adaptability tools and other smart features can be added to your cell to check incoming and ensure outgoing quality.  On the cheaper end of these are technologies such as jig-based sensing, robotic touch, or simple profile-checking lasers to verify critical features. On the advanced end of the spectrum, options exist to check correct fitment and inspection of exterior weld seams by laser scanning. If you’re in heavy transport or aerospace industries, various automated ultrasonic imaging options also exist for internally inspecting multi-pass welds.

Design for Robotic Welding and Automation in General

Design for manufacturing is a well-known concept across many successful manufacturers of all sizes; design for automated manufacturing is the next level. When parts are produced with automated systems, you need to be aware of the limitations compared to people and how to develop towards the automation strengths. In addition to tighter quality control, you also need to factor into how a robot will reach into weld seams; the order required to control warping and weld jigs should be actively included in the part design. You want to consider how to best load and unload parts into jigs, and the commonality of weld features across part variants to increase code reuse.

Upstream and Downstream Processes

Your production process likely has to change to maximise ROI from the robot system. Imagine if you took your current production, then tomorrow your welders suddenly started welding 3x faster than before? A great problem to have, but the outcome would likely be chaos in the immediate future, and this is best solved before the robot gets turned on.

There are the obvious challenges of avoiding a bottleneck somewhere in your production, limiting the true output of the new flash robotic welding system but additional factors need to be considered when transition to any form of automation.

Direct Processes

In the immediate vicinity of the robotic welding system you need to consider how to load and unload parts. The manual (un)loading of a single station is a simple and low-cost solution, however, limits the production time of the robot, and you need to consider this in the overall ROI picture. More efficient options have multiple welding stations to provide alternating welding/manual (un)loading operations to keep the robot in operation. You can also consider automated means of loading in high volume or fixed production. In the correct application, automated handling can add significant rate, stability, and safety improvements to your operation. Including automated handling can be achieved with additional robots, or in some cases, the same welding robots can pick components from additional stations and, once complete, place them elsewhere. In reality, most NZ and Australian applications will be high mix / low volume scale of operations, which is more suited to the more flexible multiple station cells. Multiple stations are often a good option to get the most from your robot, but you need to consider the part mix and cycle times to verify you get return on the additional expense. If your typical part takes 5 minutes for the robot to weld; and 5 minutes to swap out, this would be an ideal application to feature multiple stations. On the other-hand, if the part requires 10 minutes to weld and a minute to swap out then the ROI for additional stations is limited.

Indirect Processes

Further you need to consider the wider picture and look at possible internal and external disruptions. Internally you want to consider buffering between manufacturing and QA checks as soon as complete rather than prior the next step. This de-couples each process so any stoppages are localised, and this also helps human/machine interaction. Staff have breaks through-out their shifts or maybe overloaded from down members some days; you don’t want your automation to be idle while they’ve having a coffee. Those that are committed to Lean and Just-In-Time manufacturing processes will for the immediate future need to considering adding some safety margin until existing processes are verified or improvements confirmed. If your suppliers are late or you receive unacceptable parts, then you have minimal time to take corrective action before your production grinds to a halt. Your assembly and logics operations are going to be quickly overrun if they get a hold up.

The post Robotic Welding – State Of The Art Opportunities For NZ appeared first on Edge Robotics and Automation.

The post Industry 4.0 Opportunities for NZ appeared first on Edge Robotics and Automation.

Industrial Vision System Technology Types

Machine vision systems cover a very wide range of technical capabilities, real-world complexities that human take for granted, and come with integration challenges that need to be handled appropriately.

Vision systems will fall into one of below technical categories.

1. Vision Sensor
2. Smart Camera
3. Off the Shelf Vision Systems
4. Custom Vision System

Vision Sensors

Vision sensors are the simplest and basically provide a simple pass/fail/Numerical solution as per a regular sensor. Applications for these can be code readers, measurement pass fail, or presence detection.

Examples: Teledyne DASLA Boa-Spot, Cognex Insight 2000,

Smart Cameras

Smart cameras are targeted to more complex or higher throughput applications than the sensors. Smart cameras have higher performance electronic components, and the vendors provide significantly more intelligent features. These applications are when true vision skills will start to come critical to your vision system investment be discussed later in this article. Smart cameras, while often materially expensive are a great solution for many “known” application. The best benefit to smart cameras is however how easy these are integrate into lines and the modular architecture they provide.

Examples: Teledyne Dasla Boa 2, Cognex In-Sight 9000

Off the Shelf and Custom Vision Systems

Vision Systems as technical classified within the industry are when you need the flexibility to apply an expert level of knowledge. These for pushing the boundaries on known problems, or solving entirely new applications. The simpler end of these applications can be handled by off the shelf solutions that allow a system to be custom packaged with a family of components and provide simple graphical programming that is an extension of the smart camera concept. The other extreme are fully customised solutions that allow technical teams to pick best-in-class components for each element and expert code for custom programs and expert level implementation of machine learning principles.

Examples: Teledyne Dasla Vision Systems, Beckhoff Twincat Vision, MVTech Halcon

Modes of Vision System Operation

To become world class you need to use your technology investments aggressively, and to succeed with vision systems of all types means mastering the fringe areas of pass/fail results. As stressed in part one, the value from vision system is driven by the undefined grey areas of results. Getting these high level details correct will provide the best returns from your vision system investment.

There are three general modes of vision systems for your operations. How you should make this selection depends upon details of your operation, long term strategy and technical maturity of the application. Unfortunately none of this is normally well understood and correctly implemented.

Vision System Mode – Pass Known Good

The most common method employed is passing only known good products, with the questionable portion classified as rejects. This a perfectly valid approach in many mature applications but the details need to be considered in-depth rather than just acting by blindly.

The approach is quite often installed by default without properly considering the application details and user benefits. The common mistake is implementing a poor performing system, and therefore, to make sure all bad products are identified, cut points are often set so high that a sizable portion of good products are rejected. The result is you end throwing away good product and not capturing all the possible value. Compounding this factor in when not properly coached by an integrator, the poorly implemented vision system becomes considered the ground truth with no offline QA oversight. The lack of visibility means this can go on for extended months (*cough* years…) and easily reverse any efficiency gain.

Further down this article provide a example of how waste of good product occurs if not implemented properly.

Vision System Mode – Fail Known Bad

The second option is fail only known bad products. This provides the opportunity to improve the efficiency of rejecting bad products while ensuring you capturing the full potential value of good products. This is suited for production operations that have later manual processes, manufacture high value items, or products that tend to have only cosmetic defects. Filtering out absolute known bad products means that you cut your losses on those bad products and do not tie up any more manufacturing resources, yet still the opportunity to capture value from all good products.

Products that can be faulty on perception are a soft failure, and for numerous products, consumers will let the occasional borderline product go. Your company need to consider this fringe area of products value and how this relates to your maintaining your brand. Premium brands may need to accept less value capture to protect their reputation; while budget brands however can have the opportunity to increase a little more efficiency.

Vision System Modes – Manual Inspection

The last mode of operation is having the vision system split results into three groups; good, bad, and undecided, this allows targeted manual inspection of the middle ground. This offers a significant competitive advantage when applied to new applications, and even better, a then path and means to work to wards light out operation. The figure below demonstrates the labour utilisation against the manual QA methods explained Part 1. You can clearly see the saving in labour component or the more effective targeted utilisation of sample based approaches.

In a complete bleeding edge application that can only offer a concrete answer 50% of the time; it stands therefore human labour has already been reduced by 50%. Even with this early “poor” performing system, this is the single biggest productivity gain you’re ever going to get.

This then gets your foot in the door to gradually automation to cover 75,90,95% of cases until the commercial factors mean the mode operational to the other options. This approach is taken in many industries that were previously seen as to complex. Getting on with the problem means you start seeing gains as soon as possible, and these gain overtime then fund the improvement towards lights out inspection. Mentioned within Part 1 was high labour component for some primary and secondary industries, and this is typically the best approach to introducing automation of all types were it entirely manual.

Selection of Operational Mode of Industrial Vision Systems

The decision of which operational mode for an industrial vision system is based on a range of factors. It is important to get this decision correct and be realistic about the outcomes. The critical part with this mode selection is that it will affect the design of your overall production process. As mentioned, a industrial vision system is often easy-ish to upgrade, fully changing the overall production process is rarely possible in an existing line.

If you have committed to “pass known good”, then find you’re rejecting 3% of good products and want to divert questionable to a side line to check; that is going to be much harder to do once an entire line has been installed or upgraded.

Overall there is considering the scale of vision system investment to determine the size of that grey area. Rarely is that a 1:1 investment to performance ratio, and an integrator with a range of vision system experience will be able consider the options for your budget or the best ROI. Intertwined in the same process is how to go about using the confidence limits of the system. What is fundamentally being questioned is the effort to considering the grey area outweigh the possible value that could be captured.

At a minimum the following factors should be considered:

1. Available Budget
2. Realistic confidence of measurement
3. Product Value
4. Fault Cost – wasted material, wasted production capacity, equipment damage
5. Repercussion cost – recall and fines; soft factors such as customer perception
6. Production rate
7. Labour cost
8. Labour availability
9. Accuracy of manual inspection

Practical Example – Pass Known Good and Level of Investment

In the simple applications such as verifying bottle caps etc, the technology is so mature that the mode of operation is almost always pass only good. In these cases the unknown grey area is small and the products unit cost minimal, the default behaviour is to treat the grey area as failure since the cost of manually inspecting the unknown results is more than potential recapture of value.

What could be considered in this application is the performance of the by the degree of industrial vision system investment. A better performing system will have a narrow grey area and provide opportunities to capture more value by correctly grading more good products as good. This is one of the factors that is often missed, and is demonstrated in the below figure. While overall scale might not be that significant between a basic and advanced system; difference in throw-away loss can be several magnitudes.

Practical Examples – Pass Known Good & Fail Known Bad

An example of pass only known good, and fail only known bad can be examined by considering the presence of scoops within washing powder. The pass only good mode should be chosen to remove all cartons that do not have a scoop since customer could be quite annoyed. Fail only confirmed bad should instead apply to identifying multiple scoops. Multiple scoops are not ideal since this wasting material, but the customer will be more tolerant. In this example detecting a missing scoop probably needs to occur in 99 of 100 such faults, whereas detecting multiple scoops is more a bonus, and detecting 4 out of 5 cases could be considered acceptable. Pragmatically this is important because verifying at least one scoop is much simpler and cheaper than verifying only one scoop (Particularly on transparent scoops, and multiple scoops tend to sit inside each other)

The post How to gain all opportunities with industrial vision systems appeared first on Edge Robotics and Automation.

Industry media is full of cobots; there are many examples of real wins and opportunities, but some big hyperboles and myths are routinely pushed. This article will cut through the hype to offer the proper balance between cobots, robots, and many inter-playing factors.

Cobots are a significant breakthrough for industrial automation, but the biggest advancements come not from the technology but from wider industry changes:

1. Made automation less daunting for small and medium enterprises.
2. Pushed the industry players to place emphasis on the end user and simplification.

For growing SMEs, the biggest obstacle to automation is often confidence, which is where cobots can help. In many cases, the investment in industrial automation makes sense economically, but fear and lack of confidence limit companies from making that leap. Stakeholders can now be introduced to a much less daunting cobot system as the first step of automation.

Beyond the cobot technology, all manufacturers have placed a large emphasis on training and demystifying the skills into bite-size chunks. For example, technical details prior represented as electrical drawings are initially shown as pictographs. Graphical programming reduces the otherwise overwhelming learning curve for beginners.

To experienced robot engineers, cobots are actually no simpler to integrate; arguably even more difficult due to limited features, but it takes experience to reach that level. In the robot field, cobots are also a great teaching aid for beginners because all the same concepts are present.

Cobot and Robot Reality

While cobots have opened opportunities, the marketing narrative has also pushed historic connotations on “conventional” industrial automation and robotics. The overwhelming narrative from dedicated cobot manufacturers pushes the 1980s picture of industrial robots and is not at all a true sense of state-of-the-art. You will clearly see a difference if you compare the marketing story of dedicated cobot manufacturers, and major players who produce both cobots and robots.

If you’re going to take one thing away from this article, it is to understand Cobots do not replace Industrial Robotics; but Cobots and Robots complement each other.

In the hype that surrounds Cobots, it is often missed how industrial robotics have also greatly advanced; and offer just as wide opportunities that are yet to be dreamed. What is often overlooked is how smart the modern units have become and all the available features that can be included for minimal cost. Even from the mid-2000s till today is a quantum leap in your robot package. Slaving away on a factory floor with my industrial-focused PhD work around 2008, robot technology was frustrating limited, whereas today, you’re only limited by your imagination.

The below table shows a high-level state-of-the-art comparison between cobot and robot systems. The technical details will be expanded in a later article to give a more in-depth comparison and explain the subtle differences. Removing the spin, the user and safety differences between cobot and robots is much closer than most realise.

The spider figure below is very good at demonstrating the complementing relationship between the two technologies. The figure has been arranged with opposing attributes opposite; such as speed vs safety; or simplicity vs intelligence. Achieving one attribute is easy; succeeding at both requires much more experience and innovation. What is very clear is how cobots have offered attributes not previously given emphasise by key players.

The big change yet to be seen is the impact of established robotic manufacturers fully integrating cobot technology. The recently released FANUC CRX series and some rumoured new ABB Cobots will be another big change in the cobot/robot world.

The big robot manufacturers have been producing robots for six decades and have all the got all classical elements to a very fine art; but they did get caught off guard by cobots entrants. These companies have the billion-dollar R&D budgets and technical knowledge to leapfrog early entrants. If you take FANUC CRX or new ABB units these are actually more towards a full convergence of cobots and robots.

When to Consider a Cobot for Automation

The big opportunity cobots provide is the economics of retrofitting automation into manual operations and the flexibility provided to SMEs. This distinction for brownfield sites is important for cobot applications; it can lower the initial barriers and is a pragmatic first automation step.

The big returns from cobot automation come from the following use-cases

1. True collaborative applications
2. Integrating into existing production lines
3. Mobility around the factory floor

True Collaborative Applications

The technical definition for “collaborative” applications is the “robot” and person working in the same space, on the same product/task, at the same time. True collaborative applications are very rare; think of a person assembling electronics and then guiding a cobot to fasten the screws while assembling the next unit. Most “collaborative” systems demonstrated in marketing actually come under the technical definitions of coexistence or cooperation. These applications have much more flexibility, and many more options could better suit your operation. This will be covered later in a technical slanted article.

Retrofitting Cobot Automation in Manual Production

Cobots provide the opportunity to automate existing production processes without widescale rejigging of a production facility. Sometimes an existing facility has one key bottleneck that is transformational when solved by fixed or robotic automation. What is more common with a brownfield site is a range of bottlenecks that are inherit to the facility that won’t allow a significant throughput increase.

Designing a production space from scratch allows a process flow and layout to suit automation. Unless you talk about a massive corporate integrator, there will be a negligible cost difference between a cobot or robot solution on a clean sheet install. Then head to head, rarely can a cobot remotely compete with a Robot on pure ROI.

On a clean site, a robotic system will attain a much higher ROI from the increased production rate, production continuation from higher system intelligence, plus longer lifetime and extended service intervals. Designed with modern safety in approaches in mind, a robot solution will allow a similar footprint to a cobot option.

The equation is different in a handcuffed brownfield site; the flexibility to arrange the facility to suit safety systems and remove inherited bottlenecks doesn’t always exist.

Modern safety systems are highly advanced; these feature numerous electronic means of detecting human encroachment and providing protection. However, due to the incredible speeds and forces from robots, effective electronic safeguards require minimum response times and clearance distances; these can be problematic to crowbar into existing spaces. A robot solution may be impossible to retrofit into a production facility, or it can, but the robots are slowed down towards a cobot to ensure safety.

Cobots can offer a good fallback option even when not an actual collaborative application.

• The collaborative safety features can be a backup option when the more efficient external safeguards don’t fit a brownfield site.
• The slow cobot speed will not negatively impact the production rate since other bottlenecks still exist.

The flexibility of Cobot Automation

The ability to move a cobot around a facility provides good flexibility to suit flexible manufacturing. This can be very beneficial to small companies where the automation cost for one cobot can be used across multiple tasks.

There are two factors to this flexibility

• The ease of repurposing a cobot
• The difficulty of moving a robot

Moving a robot around a facility is entirely possible; however, each location will probably require some form of permanent nest. This is not a question of cobot or robot safety technology but the physics involved with a robot. Robots provide much higher speeds and accelerations, and compounding that, robots are built much more rigid and, therefore, heavier. The forces involved require some significant engineering to keep them attached to the ground unless a robot is limited to a cobot speed. You’ll not mount a robot on a self-contained moveable trolley unless it operates slowly.

The second issue with moving a robot is that the safety system should be completely revalidated each time the kit is reassembled. This validation needs to be conducted by a “competent” person. A cobot should also be revalidated for each move, but if you’re not disconnecting safety hardware nor using external safeguards, then checks can be minor in comparison.

Cobot Automation Mistakes

Believe Cobots are Intrinsically Safe

This is the big one and rarely do you see the record set clear by cobot manufacturers unless pushed by a technical audience.

Cobots have numerous safety features and attributes that make them possibly safe; but the overall system defines the safety outcome.

Take the following far-fetched examples to picture how the application, not the cobot itself, defines the safety outcome.

1. A cobot with a knife attached.
2. A person standing between a cobot and a furnace

Beginners often blindly follow the marketing narrative without paying much attention to the fine print nor fundamental machine safety concepts. Visiting tradeshows, even most cobot demonstrations, would not meet the safety standards and technical specifications. Time and time again, you’ll see the common mistakes of not factoring in related workspace hazards and crush points, region of human contact, and factors such as sharp edges on grippers or held products. It commonly takes the two ridiculous examples above for the inexperienced “click” and start fully understanding their responsibilities.

Developing an application’s holistic risk and hazard view is one of the first tasks that should be done with any industrial automation, cobot automation included. Viewing safety as an afterthought to consider just prior to production is a common mistake, even with many by machinery companies.

Not following machine safety concepts from the initial steps can lead to the following outcomes:

• Cobot installations unknowingly pose an unacceptable risk to operators.
  • Unfortunately, operators will also often not question the risk
• Remedial work to fix safety is often destructive to the original purpose and ROI of the system.
  • What typically happens is extra electric safeguards and/or guarding must be added, and the outcome is basically installing a cobot as a robot with a greater final cost and less performance.

Not Factoring ROI

Every application is different but a robot will normally have a production output of 2-5 people; a cobot on the other hand will often struggle to match a sole person. A cobot is simply not going to match a tuned robot or mechanised system for production rate.

Cobots are really best suited to support roles rather than direct labour replacement. Cobots can be great for increasing relative production from a person who completes multiple tasks. Supportive functions can be achieved such avoiding machine wait times, or left to plod along at simple tasks while human has higher priorities.

Sanity checking is required to ensure a cobot is the correct solution; take the following CNC example between a cobot and robot implementation. A robot will be faster, but it depends upon the CNC cycle time if the overall impact is significant. An important note with machine tending applications is you shouldn’t need a new program per part,

Its Not About Coding

To the beginner dropped into the world of robotics the sight of programming is a daunting experience. The actual critical skills for both industrial robots and cobots, understanding the motion concepts, figuring out an automated process, and problem-solving all the little intricate details that humans solve subconsciously.

A graduate engineer can program a robot easily, but it takes years of experience to become proficient with all the real-life problems above.

Programming is really only converting that problem into a language a machine understands. Even the highly featured languages from ABB or Kuka are on the very simple end of programming languages; languages from FANUC, Kawasaki, Yasakawa are completely prehistoric. Even with cobots, once the concepts are understood, novices will often move to text programming relatively quickly.

Cobots will sort all you problems

One key lesson for anybody starting with automation is there are numerous drawbacks compared to a person. Typical high SKU counts should not be a significant problem to any professional engineer in the Industry 4.0 world. Inconsistency and variation of incoming materials still is the archilles-heel to automation.

Human labour has the ability to subconsciously adapt to all sorts of variation due to sight, touch; and to problem solve on case by case basis. Automation system can handle variation but it will come at the expense of cost, complexity or rate;

Guarding is Not Evil

A company heavily focused around the idea of not wanting to use fixed guarding is the sure sign that all their knowledge has come from cobot marketing. There are good intangible benefits to avoiding fixed guarding such as ease of access, flexibility, and asthetics etc. The reality is; the simplest, best performing and cost effective automation solution, will generally use fixed guarding were possible.

Automation can handle variability, but it will come at the expenses of rate, complexity or cost. The number one source of variability in a production line will be human labour. That is not meaning malicious intent; but all the subtle things a person could unconsciously do differently, and different operator habits.

The best way to avoid variation is to physically control how humans can interact with the automation system. In many automated systems, safety will be linked to the doors being closed; but the electronic lock on the doors will controlled not to provide safety, but to control when an operator is allowed access.

The post Cobot Automation the Opportunities, Reality, and Myths appeared first on Edge Robotics and Automation.