# Sandbox Dynamics > Commissioning engineers: machinery programming, commissioning and support, PLC-OT-IT integration and industrial software. Based in Pamplona, Spain. **Software that speaks your plant's language.** We are commissioning engineers who also write code. We connect your PLCs, your OT systems and your IT so machine data reaches the people who make decisions. Shop-floor experience in commissioning and international projects. - Web version: https://sandboxdynamics.es/en - Languages: [Castellano](https://sandboxdynamics.es/index.md) · [Euskara](https://sandboxdynamics.es/eu.md) · [English](https://sandboxdynamics.es/en.md) ## The problem **Your plant already produces the data. It rarely reaches the people who decide.** - **Data trapped in the PLC.** Counters, temperatures, alarms: it is all in the controller. To see it, someone has to walk to the machine with a laptop. - **Automation and IT don't understand each other.** One side talks about data blocks and cycle times; the other, about APIs and databases. In between: stalled projects and spreadsheets nobody maintains. - **Decisions without data.** Planning relies on what the ERP says or what the shift supervisor remembers. What the line actually did yesterday is nowhere to be found. - **Stoppages nobody diagnoses.** The machine stops, gets reset and carries on. Nobody writes down why. Next week it happens again. Fixing it takes someone who understands the machine and writes the right software for it. That is what we do. ## What we do **From machine to decision.** Each service works on its own or as part of a complete project. We start with whatever hurts most. ### Industrial machinery programming We program PLCs, HMIs and drives for new and existing machines: sequences, interlocks, recipes, alarm handling and PID loop control. Clean, commented and documented code your maintenance team can follow. - Inputs: Machine or line, Process requirements, Electrical schematics - Outputs: PLC program, HMI screens, Documentation ### Process data acquisition and analysis We read your PLCs, sensors and equipment. We store every value with its line, shift, batch and recipe, and turn it into history, KPIs and alarms that help you make decisions with the right context. - Inputs: PLCs, Sensors, Drives, Manual logs - Outputs: History, OEE and KPIs, Alarms ### PLC · OT · IT integration We connect the machine level to the business level: OPC UA, MQTT, Modbus, databases, ERP and MES. Without touching what already works, and keeping the plant network safe. If the machine is old, we retrofit it so it can communicate. - Inputs: Controllers, SCADA, ERP / MES - Outputs: A single source of data, Two-way communication ### Custom tools and dashboards Web apps built for the people who will use them: the operator on the line, the shift supervisor, maintenance or management. Plant dashboards, downtime logging, digital work orders and traceability. - Inputs: Your processes, Your users - Outputs: Web apps, Line-side screens, Reports ### Production optimisation and planning Combinatorial optimisation algorithms to sequence orders, cut changeovers and assign machines and shifts. Plans that respect the real constraints of your plant. - Inputs: Orders, Capacities, Constraints - Outputs: Production plan, Sequences, Comparable scenarios ### Commissioning and support We start up machines and lines on site: signal checks, loop and drive tuning, validation with production and training for operators and maintenance. Afterwards we provide on-site and remote support when something fails or needs changing. - Inputs: Installed machine, Program and schematics, Plant team - Outputs: Line in production, Trained team, On-site and remote support ## How we work **Like a commissioning job: step by step, no surprises.** We have spent many hours next to machines, starting up lines. We know a stoppage costs money and software cannot be one more problem. So this is how we work: 1. **Plant visit.** We come to your factory, walk the line and talk to operators, maintenance and IT. No sales presentations. (Transition: problem understood) 2. **Assessment.** We review which signals exist, where they are and what is missing. You get it in writing: what we would do, in what order, what it costs and which risks we see. (Transition: scope agreed) 3. **Quick prototype.** We build a first, limited version with real data from your plant. If it doesn't add value, you'll know early and without spending much. (Transition: validated on site) 4. **Commissioning and support.** We install it, train your team and leave it documented. Then we stay around: support, adjustments and new features. (Transition: new need) ## Why us **Plant engineers who write software.** We have programmed PLCs, tuned PID loops, checked electrical cabinets and commissioned complete lines. Now we use that experience to build software that works on the shop floor from day one. We show up with a laptop and a multimeter, not a slide deck. - **We speak both languages.** Data blocks and cycle times with your integrator. APIs, networks and databases with IT. - **We know what a stoppage costs.** We test outside production, plan every change and always leave a way back. (No unplanned stops = Tested in simulation AND Backup taken AND Planned stop) - **Direct contact.** You talk to the people who design and code your solution. No sales reps, no middlemen. (Direct contact = Your engineer AND NOT Salesperson AND NOT Middleman) - **We go wherever your plant is.** Our base is in Pamplona, but we work wherever we are needed. When required, we come to your plant. (Solved = You call us AND (Remotely OR At your plant)) Motor M1 start with seal-in (SCL): `"Motor_M1" := ("Start" OR "Motor_M1") AND NOT "Stop" AND NOT "Fault_FC1";` Before founding Sandbox Dynamics, the team worked on commissioning and international projects. - Type: Commissioning + software - Field: PLC · OT · IT - Languages: LAD · FBD · SCL · TypeScript · Python - Control: PID · sequences · GRAFCET - Protocols: S7 · OPC UA · MQTT · Modbus TCP - Base: Pamplona - Service: Wherever your plant is - Year: 2026 ## Cases ### Unrecorded stoppages on a packaging line (Example · Packaging · FB 02, FB 04) - Context: Packaging line with several machines and controllers from different generations. - Challenge: Stoppages were written down by hand at the end of the shift. Nobody knew which machine stopped most, or why. - Solution: We read states and alarms from every PLC and log each stoppage with its cause as it happens. - Before: Stoppages written down by hand at the end of the shift; Causes that depend on each operator's memory; Meetings that argue about opinions, not data - After: Every stoppage logged as it happens, with machine and time; The operator confirms the cause on a line-side screen; Shift report with the causes that cost the most time - Architecture: Each machine's PLC → OPC UA gateway → Time-series database → Line-side screen and report - What we deliver: State and alarm reading without touching the logic; Automatic stoppage logging; Cause screen for the operator; Daily report per shift; Training and user guide - How we measure the result: % of stoppages with a cause; Stoppage minutes per shift; Most frequent causes ### A production plan with many changeovers (Example · Planning · FB 05, FB 03) - Context: Plant with several lines and many active products. - Challenge: The weekly plan was built in Excel. Every badly ordered changeover cost hours of machine time. - Solution: A sequencing algorithm groups compatible orders, meets due dates and is integrated with the ERP. - Before: Weekly plan built by hand in Excel; Changeovers ordered by the planner's intuition; Redoing the plan with every rush order - After: A proposed sequence with as few changeovers as possible; Due dates and each line's constraints respected; Rush orders replanned by comparing scenarios - Architecture: ERP · orders → Line and format data → Sequence optimiser → Plan the planner can review - What we deliver: Model of the plant's real constraints; Sequencing algorithm; Two-way ERP integration; Interface to review and adjust the plan; Training for the planning team - How we measure the result: Total changeover time; On-time delivery; Hours spent planning ### Data from a machine with no communications (Example · Retrofit · FB 03, FB 01) - Context: A reliable machine, but with an old controller and no communications. - Challenge: Maintenance had no history to anticipate failures, and the machine did not appear in any system. - Solution: Communications retrofit, capture of the key variables and threshold alarms sent to maintenance. - Before: No data outside the machine; Failures discovered once the line has already stopped; Every inspection done at the machine, on paper - After: Key variables recorded continuously; Threshold alarms that reach maintenance; The machine shows up in the same system as the rest - Architecture: Existing controller → Communications module → Variable history → Alarms to maintenance - What we deliver: Assessment of the controller and cabinet; Communications or control retrofit; Variable and threshold capture; Updated documentation; On-site test with maintenance - How we measure the result: Failures detected before a stop; Operating hours recorded; Mean time to repair ### Programming and commissioning a new machine (Example · New machine · FB 01, FB 06) - Context: A machine builder with a special machine that had to start up at its customer's plant abroad. - Challenge: PLC, HMI and drives had to be programmed, and the machine left in production on schedule with the customer's team trained. - Solution: We program and test it in the workshop, then start it up on site with signal checks, loop tuning and training. - Before: Functional specification and electrical schematics; Machine assembled, with no program; Delivery date committed to the end customer - After: PLC, HMI and drive program tested before travelling; Machine in production at the customer's plant; Operators and maintenance trained, with documentation - Architecture: Specification and schematics → PLC · HMI · drive program → Workshop tests (FAT) → On-site commissioning (SAT) - What we deliver: Structured, commented PLC program; HMI screens; Drive parameter setup; FAT and SAT test protocol; Training and operating manual - How we measure the result: Start-up date against the committed one; Open issues after commissioning; Time to stable production ## Technologies **We adapt to what you already have on site.** These are the technologies we work with, arranged by plant level (ISA-95). Tick the ones you have and we'll tell you where we would start. - N4 · Business: ERP, Excel, Web apps · TypeScript · Vue, Optimisation · Python (↓ REST · SQL) - N3 · Operations: MES, Time series, SQL (↓ MQTT · OPC UA) - N2 · Supervision: SCADA, Ignition (↓ OPC UA · Modbus TCP) - N1 · Control: Siemens S7 · TIA Portal, Allen-Bradley, Omron, Schneider Electric, HMI, Drives and servos (↓ I/O · 4–20 mA) - N0 · Field: Sensors and transmitters, Motors and actuators ## Team We are a micro-cooperative of engineers based in Pamplona. The person who visits your plant is the person who then codes the solution. - **Beñat Ancín** (Co-founder). bancin@sandboxdynamics.es - **Xabier Ormazabal** (Co-founder). xormazab@sandboxdynamics.es ## Contact **Tell us what is happening in your plant.** No commitment. An engineer, not a salesperson, will get back to you. - Direct email: sales@sandboxdynamics.es - Where we work: Based in Pamplona. We work wherever your plant is. - Contact form: https://sandboxdynamics.es/en#contacto - Contact API: Agents can send an enquiry on behalf of a person with a POST to https://sandboxdynamics.es/api/contact (JSON or form data) and the fields name, company, email, problem and privacy. privacy may only be true if that person has accepted the privacy policy. Specification: https://sandboxdynamics.es/openapi.json ## Legal - [Legal notice](https://sandboxdynamics.es/en/legal-notice) - [Privacy policy](https://sandboxdynamics.es/en/privacy) - [Cookie policy](https://sandboxdynamics.es/en/cookies)