Está diseñando un nuevo sistema de control. ¿Cómo te aseguras de que el hardware que elijas sea fiable?

A few thoughts: 1- Select components that adhere to recognized standards like ISO 26262 or IEC 61508, which provide guidelines for functional safety and reliability in various industries. 2- Implement a robust preventive maintenance strategy, incorporating utilization-based, time-based, predictive, and diagnostic approaches. This strategy should include regular inspections, simulation and field testing, and maintenance of system components to address potential issues preemptively. 3- Employ certified service engineers with rigorous training and access to a comprehensive knowledge base. 4- Adopt a holistic lifecycle approach, from design through maintenance, ensuring each phase meets stringent reliability criteria.

The typical Control systems has 2 types of process/plant demand. One is Base layer process control (BPCS) that is served by Distributed Control system (DCS) architectures. The second part of the control system addresses plant and process safety and is served by Emergency Shutdown (ESD) & Fire & Gas systems. Most DCS system architecture allows for hardware redundancy configurations for CPU< Memory, Communication and even Input/Output (I/O) modules. Most ESD systems uses SIL3 certified hardware and follows IEC 61508 standards. Most F&G systems uses SIL 2 again high reliability component hardware. The Supervisory Control & Data acquisition systems (SCADA) uses Remote Telemetry Unit (RTU) with redundant hardware modules options in architecture

I go by track record… been using #AllenBradley #rockwellautomation PLC’s and related components for over 30 years, and only have had a handful of component failures in this time We Just de-commissioned a PLC5 system we put in 25 years ago. It ran all that time without a hitch.

Para asegurar la confiabilidad del hardware en un nuevo sistema de control, sigue estos pasos: Define claramente los requisitos del sistema. Especifica componentes de calidad de proveedores confiables. Realiza pruebas exhaustivas de los componentes antes de la integración. Considera la redundancia para componentes críticos. Asegúrate de un diseño térmico adecuado y una fuente de alimentación estable. Implementa protecciones contra sobretensiones y sobrecorrientes. Cumple con estándares de la industria relevantes. Realiza pruebas integrales del sistema y monitorea continuamente. Documenta de manera exhaustiva el diseño y las pruebas realizadas.

Understand your process,its parameters,operating conditions,adherence to standards for process design and equipment. How many I/Os you need to accommodate,spares I/Os for expansion later in the process lifecycle, redundancy if you going to implement it. You also need to understand how much safety you need to implement,that is are you going to need just a normal BPCS(basic process control system) or you need to implement ESD(Emergency Shutdowns ) with Safety Trip Amplifiers. You also need to understand how much latency you can allow during data transmission,this leads to understanding the correct communication protocols to implement and the communication architectures to design. These are just a few points.

Choose products from reputable manufacturers and those that have good support. I've found that the support you receive during and after the sale is as if not more important than what products you specify. Component cost is also important but secondary to product quality and support.

Aqui estão alguns passos gerais que eu sigo para especificar um hardware. Requisitos do Sistema; Ambiente de Operação; Tipo de Aplicação; Entradas e Saídas; Capacidade; Manutenção e Suporte; Custo; Tendo um orçamento disponível para o projeto. Busco o melhor equilíbrio entre desempenho e custo.

True adage Steve. The past few years has highlighted the importance of how your partner with suppliers. Companies who excelled through the chaos of pandemic and then supply chain strains were those who invested in likeminded, long view strategies. Top tier suppliers merge quality products with world class service, and enduring supply chains.

Inexplicably, this article is completely omitting design. There is a lot more to hardware than specifications & evaluation. Before you can establish spec’s you need to *fully* understand the use case, and the huge variety of challenges/requirements that the hardware will need to overcome/adhere to in the field. Understanding what the hardware *actually* needs to do and designing backwards from that is ~75% of the work. If you have done this correctly, and feel confident that you understand the problem set that you are solving *better* than the customer, then everything else will fall into place with simple diligence. If you fail to do this, then no matter how well you execute, the solution will not be appropriate.

Performance requirements Compatibility with existing system Compliance with standards Vendor support Mean time to Fail/repair Redundancy Cost Analysis Are few amongst many other factors to consider.

This cannot be stressed enough: test, test, TEST! When thinking through the traditional FAT, SAT, and unstructured testing, there are a lot of considerations and documentation elements involved. Some times it can be overwhelming. Whether it's a large project or a small fix in the field, the key element to testing is to ask yourself: what can go wrong? If you think through worst case scenarios and "bad day" analysis, you'll be well on your way to creating testing requirements and building a more robust response as a result.

Test every aspect of the control, including alarms and warning generation. It gets tedious, but the more time that is spent in thinking about the process and effects, the more the system matures.

We test every panel, every device, every I/O point before shipping to the customer. It's 10x easier to fix in the panel shop rather than in the field.

Beyond FAT onsite inspection and testing is very important. Bench testing includes testing for Power supply, precision, check for different inputs and disturbances to verify performance. Using proper test instrument to measure and record parameters.

Las pruebas de hardware son cruciales para garantizar la confiabilidad en un sistema de control industrial. Aquí están los puntos clave: Verificación de Funcionalidad: Antes de la instalación, verifica que el hardware cumpla con las especificaciones y funcionalidades necesarias mediante simulaciones y pruebas. Manejo de Perturbaciones e Incertidumbres: Evalúa la capacidad del hardware para manejar condiciones adversas y escenarios no ideales durante las pruebas. Métodos de Prueba: Utiliza simulación, pruebas de banco, pruebas de campo e inyección de fallos para abordar diferentes aspectos del rendimiento del hardware.

Concur with Jason on this. The "what can go wrong?" point is critical. We can't test everything, so we must have a sound methodolody to determine those events that: 1) could actually happen (i.e., there's a vulnerability that can be exploited), and 2) could generate consequences that meet your threshold (i.e., cause a really bad day). That's where your limited test resources belong...where those two conditions intersect. --IMHO

Perks of being a preferred systems integrator is that you have the ability to request samples. Typically these requests are fulfilled quickly because the vendors know what’s to come!

I totally agree with the "TEST, TEST, TEST" philosophy. But, unfortunately, many people, especially "young engineers/technicians" just starting their careers and "some" customers not totally appreciating the "Real World", do not fully appreciate the limitations inherent with each phase of testing. Yes, FAT does have great value. But, after FAT is "transportation". "Things" often get shaken loose. I watched a junior engineer spend three days debugging an issue only to "discover" a loose screw. When challenged, he commented "but was checked at FAT". I reminded "FAT was BEFORE transportation." Similar to customers who "thought I was just wasting time". Likewise, FAT does NOT test for field installed devices and connections.

Todas as funcionalidades e possíveis ocorrências devem ser exaustivamente testado de modo a garantir que o desempenho não seja comprometido decido e eventos inesperados.

When testing hardware for a new control system, key parameters to consider include performance, reliability, compatibility, scalability, environmental suitability, power requirements, certifications, and cost. These factors ensure the hardware meets system needs efficiently and reliably.

When developing a new control system, I always start from the principle of equipment redundancy, in which there are two or three controllers, which in turn use watchdog timers to exchange with each other and switch to the backup controller in milliseconds, without affecting the operation of the controlled one. equipment. Another very important quality is backup communication channels, which allows you to protect against data loss on systems with an upper level of control.

99% of the time, redundancy is not needed. There are certain mission critical applications where redundancy should be and is required. This is more things like control systems in airplanes, nuclear reactors, etc. It doesn't apply to most of manufacturing.

Redundancy provides the ability to minimize control system downtime. With the cost implications of these solutions I typically expect only mission critical units apply hardware redundancy. Consider a risk based approach based on hardware reliability data and process criticality.

Is no one aware of "Safety PLCs and PACs" with their own level of inherent redundancy? I implemented my first "Safety system" in 2005. Is still going strong today.

Implementation of hardware Redundancy in a short run incurs additional cost and in a long run earned revenue. The revenue which would be lost in a period of a production stop is more than the revenue used to implement hardware Redundancy.

Existem situações que mesmo com equipamentos de excelente qualidade ainda não se elimina a possibilidade de uma eventual falha cause prejuízo materiais, produtivos ou pessoais, por isso, usar a redundância é um recurso extremo, mas que oferece maior confiabilidade.

When considering hardware redundancy for a control system, prioritize criticality and redundancy configuration (hot, warm, cold). Perform failure analysis to identify potential points of failure and select reliable components with proven reliability. Ensure compatibility and automatic failover mechanisms for seamless switchover. Thorough testing and validation are essential, balancing costs against potential downtime.

IPD - intrusion protection devices (physical and software) need to be installed especially for exposed field ports (e.g., ethernet) to introduce (not prevent as nothing will prevent malicious dedicated intent) barriers of access.

Hardware protection involves safeguarding components against environmental and electrical hazards. This includes using enclosures, cooling systems, surge protectors, and selecting high-quality, reputable components. Implementing redundancy for critical parts ensures system resilience, enhancing long-term reliability against physical and electrical stress.

Properly sizing wires, circuit breakers, overloads, fuses is extremely important. Know the environment your control panel is going into also. If it is extremely dusty, do not put a fan in your control panel.

Hardware devices can be protection from cyber threat by implementing * Intrusion detection system (IPS) *intrusion Prevention System (IDS) * Firewalls/NGFW

When selecting hardware protection for a control system, consider factors like environmental conditions (temperature, humidity), electrical protection (overvoltage, overcurrent), and electromagnetic interference shielding. Evaluate the reliability, durability, and lifespan of components. Implement surge protection devices, fuses, and circuit breakers. Ensure proper grounding and isolation techniques. Perform thorough testing and validation to guarantee effectiveness in real-world conditions.

Electrical drawings are very important. We also include network layouts, device settings, IP addresses in our drawing packages. This will come in very useful for the end user.

I agree with complete drawings and documentation. I had one "VERY INTERESTING" customer who specifically excluded such from our "scope of work" in effort to reduce his cost, but "thought" he could get for free since he "knew" we would generate for our records. Anticipating such starting on FIRST service call I made a point of ALWAYS handing his maintenance manger copy of the contract at start of EACH service call. After the third service call he paid "FULL PRICE" for the documentation he had originally removed from the contract. Each service call cost him more than the amount he was initially trying to save. Interestingly, there were no further calls after.

Comprehensive hardware documentation, including technical specifications, maintenance manuals, and certifications, ensures correct usage and upkeep. It provides vital information for system integration, routine maintenance, and confirms compliance with safety standards. Proper documentation supports system reliability by guiding operations and maintenance throughout the hardware's lifecycle.

Documentação clara e bem. Orientada, colabora para que o produto seja usado e empregado de forma adequada. A apresentação, organização e esclarecimentos precisos permitem a outros profissionais maior confiança e certeza no emprego do equipamento.

For comprehensive hardware documentation in a control system, prioritize detailed specifications of each component, including model numbers, manufacturer details, and technical datasheets. Provide wiring diagrams, connection schematics, and layout plans for clear installation instructions. Document operational parameters, such as voltage, current, and temperature ratings, along with maintenance procedures and troubleshooting guides. Ensure compliance with industry standards and regulatory requirements, including safety certifications and environmental directives. Regularly update documentation to reflect any changes or revisions in the system configuration.

Evaluating hardware through stress, life cycle, compatibility, and field testing ensures performance under expected loads, longevity, seamless integration, and real-world operability. This process identifies potential failures and compatibility issues, ensuring the hardware meets reliability and performance criteria, crucial for minimizing system failures and extending operational life.

TUV certification. Why have the expense of additional evaluation when the manufacturer/supplier has already had "certified" by qualified 3rd party.

Reliability is a relative term and for any system can be summarized in a nutshell... # the hardware is compatible and strong enough to withstand process or environmental challenges. # Redundancy is available in case of any hardware failure. # Communication is robust and has also redundancy. # Software is up to date and handles the process conditions well.

1. Spare parts needs to be considered. 2.Operation & maintenance Staff to be trained well and updated to the system requirements. 3. Find a different way to check the system performance.

I know this article is about the reliability of the control system hardware; however, in addition to hardware, several research papers are associated with the structure and type of PLC code and the influence of the designs in simplex and redundant systems. “New strategies for reliability analysis of Programmable Logic Controllers” and “Bayesian-Network-Based Reliability Analysis of PLC Systems” are examples of some of the research where the point is that the execution of the logic has a direct relationship to the field devices the PLC is controlling and the overarching system reliability. The research looks at simulations and alternate calculation methods for probabilities, as there is much uncertainty.

Based on my experience as System Integrator of advanced control system using PLCs and DCS, I recommend the following 1. Based on Process Control Philosophy classify control needs into two broad categories viz, BPCS and SIS 2. Follow the best practice of selecting DCS for BPCS and PLCs for SIS with approved SIL rating 3. Ensure reliability by analysing Failure data of modules selected for configuration 4. Implement Fail Safe design. 5. In the SIS, implement unit or plant shutdown as per the severity of Emergency situation 6. Insist for structured development of control software for DCS and PLC

Curated resources, currently, code on hardware cannot write itself therefore people with the experience need to code and challenge the solution. Couple this with the understanding process needs (e.g., execution speed, SCADA click to device response) to provide a robust hardware control solution.