Automation, built around your floor.

When we talk about industrial automation, we mean the layer of equipment that does what an operator's hands and eyes used to do: collaborative robots tending machines, vision systems verifying parts, motion controllers running indexers, safety scanners watching for someone walking into a cell. We're not a control-system house — we don't replace your DCS — and we're not a turnkey-line OEM. We're the integrator who builds the cell that fits between your existing equipment, validates it, and walks it through your customer's audit.

Adams added the automation pillar formally in 2009. We made our bones in fluid power, and we still do. But manufacturing changed: a lot of customers who used to need a third pneumatic cylinder now need a cobot tending the machine that the cylinder used to feed. Rather than send those customers down the road, we hired the engineers, brought in the brands, and built the practice. Today about 35% of our project revenue is automation.

Universal Robots is our anchor cobot brand. We're a UR Authorized Systems Integrator with seven engineers UR-certified at the application level. We build cells around UR3, UR5, UR10, UR16, and UR20 cobots, with end-effectors from Robotiq, Schmalz, and OnRobot depending on the application. Cobots are the right answer when you need flexibility, low ceiling-height, fenceless deployment, or fast SKU changeover. They're not the right answer for very high-payload or very high-cycle applications, and we'll tell you that.

For vision, we standardize on Cognex In-Sight (smart cameras) and DataMan (barcode/DPM readers). Cognex has the deepest library of off-the-shelf inspection logic in the industry and the best deep-learning toolkit for the messy inspection problems that traditional rule-based vision doesn't solve. We've deployed Cognex cells for fastener-thread inspection, label verification, presence-absence checks, vial cap-color sortation, and assembly-line traceability with 100% capture.

Motion and safety: we work with Allen-Bradley CompactLogix and ControlLogix for cell-level PLC integration, Banner safety scanners and light curtains, and IFM/Sick proximity sensors for in-cell logic. We're not religious about brand on the controls layer — most of our customers already have a brand standard, and we work to it.

Most of our automation projects are scoped from a customer pain — "we can't find operators for line 3," "changeover takes 90 minutes, we need 30," "audits flagged sampling-only inspection." The first conversation is on the floor, not in the conference room.

From that, we scope a cell: cobot, end-effector, vision (if applicable), guarding, safety devices, controls integration. We do a $4,000 to $8,000 paid feasibility study for cells over $80k — it gets the engineering decisions made before either side commits. The deliverables are a CAD model, a controls architecture, a pre-validation HAZOP, and a fixed-price quote. About 80% of the studies we do convert to a build.

From there: 6 to 14 weeks of build and FAT in our Tampa shop. We mock up the upstream and downstream interfaces and run real cycles before the cell ships. On-site commissioning is typically 3 to 7 days. We train your operators in person, leave a runbook, and stay on call. For projects in regulated industries, we walk the cell through IQ/OQ/PQ with your quality team and provide the documentation package.

We get this question on every kickoff call. The honest answer:

— Cobot when payload is under 16 kg, cycle time is over 4 seconds per part, you need flexibility for SKU changes, you don't have ceiling height for hard guarding, or you need to deploy in under 12 weeks.

— Industrial robot when payload is over 25 kg, cycle time is under 2 seconds per part, the application runs one product for a decade, and you have the floor space for full hard guarding.

In between: it depends on the application. We've built cells where the cobot was the wrong answer and we said so before the quote went out. We've also built cells where the customer was sure they needed a $400k industrial robot and we proved out a $90k cobot cell that does the job. The diagnostic is always: what does the application actually need, not what's fashionable.

An automation cell that the operators don't trust gets shut off. We've seen it. Our deliverable isn't a cell that works on day one — it's a cell that works on day 365. Every project includes hands-on training for two shifts of operators, a one-page laminated runbook, recorded video walkthroughs, and a 30-day warranty period during which we field every operator question that comes in. We also run a quarterly UR programming course at our TechFest training facility for customers who want their own people fluent.