An investment in a pilot plant often is an important step for successfully scaling up a product. The pilot plant can provide critical technical knowledge of how to build a large-scale plant. Indeed, a properly engineered pilot plant delivers important insights. Discovering faulty assumptions and errors on a small scale can avoid financial, safety and technical risks in subsequent larger units.
A pilot plant essentially is an evolution of your product to increased scale. A pilot plant’s primary objective is to demonstrate chemical and physical stability of a new production technique. Modern modeling software offers a quick and iterative way to find optimal design parameters — but models contain many assumptions that often have vast unknown effects on chemical processes and carry inherent risk.
How can you ensure your pilot plant is optimally designed to deliver the correct insights? This article describes proven pilot-plant design and optimization techniques that increase your opportunities for successful scale-up.
Laboratory-scale data are a small piece of the information puzzle required to design a successful pilot plant. You’ll need a variety of other essential details (some perhaps not fully known):
Commercial information. This often is an afterthought but is extremely important to a project’s success. This category includes goals that establish budget, timeline, expected lifecycle operating cost and desired return-on-investment for the project. The spend and timeline targets will affect everything from the operating ranges you can use to the suppliers available.
Additional commercial aspects of a project include environmental permitting requirements and site selection.
It’s important to identify required environmental permits and apply for them early in the project. Obtaining the correct permit approval often takes longer than building the pilot unit.
Early site selection also is critical because a pilot plant must be designed for its destination. Are all the utilities (steam, water, nitrogen, etc.) required by your process available on site? What are the dimensions of the available space your unit must fit into? Do needed heating/ventilation/air-conditioning and electrical services already exist? Do you know seismic zoning or wind loads?
When possible, select a site where you can use existing infrastructure and with available chemical plant operators and maintenance personnel. This will save you money during design and operation.
Technical information. Establish a clearly defined basis of design first. It’s important to ensure all stakeholders agree on the production goals before you begin designing the details. Disagreement on the pilot plant’s goals, desired rate of production, finished product specifications or raw materials to use can delay or even derail the project.
Once stakeholders agree on those basic inputs and outputs, key information to gather includes:
• desired operating ranges (pressure and temperature) for each phase (e.g., heating, separation, cooling, etc.);
• order of component addition;
• all process steps with technical details including any required separating or blending of products for final use;
• mixing, heating and cooling times; and
• distillation or drying rates (if required).
A lab-scale unit can provide certain key information, particularly data on chemical composition at different points in the production process. These details are important to the pilot plant design and define the product evaluation and validation steps.
Figure 1. Such an arrangement simplifies initial hookups and, if required, later moving the skid to another location.
Additional information. Usually scale-up isn’t performed in-house. Most often, you must enlist an outside engineering firm. Consider the areas that your in-house team lacks sufficient knowledge and fill in the gaps with outside resources.
One common in-house deficiency is a thorough grasp of the Process Safety Management of Highly Hazardous Chemicals Standard (29 CFR 1910.119). Understanding the requirement isn’t easy; you need experts with field experience in this area to avoid expensive mistakes.
Your outside engineering firm will require a variety of information:
• What automation and controls are needed?
• Will you require any “add-ons” like clean-in-place (CIP) or effluent neutralization systems, etc.?
• How do you plan to handle waste byproducts?
• What method of packaging and shipping do you want?
• Where will you get seasoned operators to run the pilot plant?
The earlier you can start planning for these operational considerations, the more accurate your design will be.
Tackling Design And Costs
Even though you’re building a physical pilot plant, the process steps begin with modeling. Modeling software is the perfect zero-to-infinity knob for rapidly dialing into target specifications. Process modeling of your desired product inputs and outputs quickly will determine if scaling to the selected size is feasible — and doable at a reasonable expense.
After determining the optimal mass and energy balances via simulation, selecting equipment is the next major step. Think one move ahead. Strategize on the next-size pilot plant and the data necessary to create a confident design for that scale. The goal is to use large-enough equipment for key processing steps to get valid data without over-engineering that results in high costs.
Follow a 10:1 rule when sizing your system. If you’re making 1,000 gal/d, you should aim to select equipment that allows for 10,000 gallons. Going beyond that scale doesn’t give you reliable data on heat transfer and reaction kinetics.
Once you have a working design and model, the next step is to value-engineer your pilot unit to address factors you can control. While you can’t change the corrosive nature of a required chemical, you likely can achieve cost savings by taking steps in other areas, e.g.:
• Re-evaluate output requirements. Throughput goals are a balancing act between expense and data validity. Select a production output that provides the right flow. Create a model for the next level that isn’t excessive. A 1,000:1 difference is too big to offer much confidence whereas a 2:1 ratio won’t give you the proper scientific data.
• Cut non-critical instrumentation and sampling points. Not all data will lead to design decisions. Re-evaluate which instrumentation and sampling points are providing important data and which are just “nice-to-have.” A single flow meter can cost several thousand dollars and every sampling point requires extra design, equipment, etc.
• Reduce access points. An initial pilot plant design usually provides access to too many areas. For example, do you require stairs or a permanent platform on the side of a reactor for an instrument you check once per month? The cost of a ladder is much lower than that of a platform.
• Exclude known processes. A pilot plant is engineered and fabricated to test unknowns, not to model a complete process. If a portion of your production facility involves well-understood technology, then leave it out. This can include having bulk tanks of CO2 instead of scrubbing flue gas, not purifying your final product or testing packaging methods. Similarly, you likely can cut a downstream process that doesn’t feed another unit operation in the main process technology sequence.
• Take advantage of modular construction. The iron triangle rule of project management states that scope, schedule and budget drive a project. Off-site modular construction of a pilot plant is a great method to reduce your project timeline without compromising scope or schedule, especially on a project that needs on-site improvements or permitting. Your pilot plant can be designed, built and fully tested off-site. This parallel-path project schedule can shave months off a project timeline.
• Re-examine materials of construction (MOC) requirements. Highly corrosive products and other materials may require expensive MOCs to run safely. While you must ensure your skid is safe, you might be able to reduce costs by reviewing exactly what MOCs you need where. If, for example, part of the process just delivers benign ingredients to a tank, maybe you can downgrade materials for that section of piping. The best way to confidently make this call is to coupon test at lab scale.
Figure 2. Running piping in streamlined racks, installing equipment on the skid’s outer edge, and providing a center aisle enhance accessibility for maintenance.
In addition, logistics is important to project success. You must optimize constructability and equipment installation sequencing to minimize interference for later equipment and instrumentation additions. Two-dimension and three-dimension skid models can identify how equipment is layered and simulate additions to the skid to form a sequence plan.
Lowering Operating Costs
Pilot plant costs don’t end when the system is installed at your facility. Changes to the utilities and raw materials required for operation and daily pilot plant procedures can significantly influence overall expenditures.
Even though a pilot plant often is a temporary installation, it still can markedly affect your overall manufacturing site. To optimize your operational expenses, consider the following:
• Design with installation and operation in mind. A pilot plant skid’s physical layout can substantially impact installation, maintenance and repair time. For example, placing all process connections at one location on the skid can simplify utility hookup on site arrival or when moving the skid to a different location (Figure 1). Ensuring easy access to all equipment and installing piping into streamlined racks (that run along the tops or sides of skids) aids accessibility for maintenance personnel. You can improve this process by arranging equipment on the outer edges of the skid and leaving room for a center aisle when possible (Figure 2).
• Select an appropriate site. Locate your pilot plant near utility hookups whenever possible. It’s beneficial if required draw materials for your process already exist. Check your permits and draw capacity. The pilot plant might alter an existing permit or require draw beyond the capacity of existing utilities and raw material supplies.
• Resist the automation temptation. For a temporary process setup like a pilot plant, manual operation usually is most cost effective. The expense of automating functions within the pilot plant might outweigh the benefits. Control valves are the perfect example of excessive automation that often bloats the cost in a pilot system. If specific valves on the pilot plant are open most of the time, you don’t need a programmable logic controller to turn them. A plant operator with a valve checklist is an inexpensive alternative.
• Manually add ingredients. An operator with a shovel often is a cost-effective alternative to expensive automated solids-handling equipment. A pilot plant usually requires a much lower volume of ingredient additions than a full-size plant. You may include automated equipment if your process highly depends on sequencing precise ingredient additions or you want to test the automation for larger-scale production.
• Rethink packaging. Can you eliminate bulk product storage and manual packaging? Switching chemical totes with a forklift every few hours is significantly cheaper than packaging automation. Packaging usually isn’t a critical process to test at pilot scale because most packaging technologies (and costs) are well known.
When considering ways to lower operating costs, avoid sacrificing the robustness of your overall system. People tend to be less consistent than automated equipment and potentially could introduce a new source of process upsets, increased transient states or required startups and shutdowns.
The Secret Sauce
The last major area to consider is project management. Poor project management rather than a design flaw often causes a project’s failure. From scheduling mishaps to sub-contractor miscommunication, many steps can go wrong once your pilot plant leaves the design desk.
There’s a considerable advantage to having an outside pilot plant design/ build expert manage your project from concept to completion. Risk management plans, scheduling tools and equipment specifications are a few examples of highly refined tools your design/build partner will feature.
Look for a partner with knowledge of both plant operations and design/build/engineering experience in your size project. The process systems company should have a deep understanding of the fundamental relationships and dependencies that exist between the major phases of projects and the design/build/equipment tradeoffs for your type of project.
When you find that great partner, get out of its way! Trying to split management between your team and the outside project manager takes the liability of the project outcome away from the firm. Many operating companies think it’s more cost effective to order their own equipment. However, this is a great way to unwittingly hold back an engineering design team, order the wrong equipment, weaken technical efforts or delay fabrication. These problems cost more time and money in the long run and can add frustration for both you and the firm.
Find a partner you trust and lean on it to avoid mistakes, prove your process and deliver a successful pilot plant.
Lining Up A Good Partner
Working with an experienced engineering and fabrication firm allows you to avoid costly pitfalls during many stages of your pilot plant project. But how do you find the right firm? Any company can look good in its sales brochures or on its website. Three strategies can help you sift through the marketing and sales noise to find an outfit you can count on:
1. Request case studies, photos, testimonials and references.
To get more than a surface-level impression of the company you may engage, ask for proof of a firm’s work with pilot plants and the industries it has served. Request a reference list or reach out to others who have used the outfit before; dig deeper than the typical “Would you use this process equipment design firm again?” Focus on questions that highlight potential strengths and weaknesses, such as:
• What went right (and wrong) with its project? How did the company handle unforeseen challenges?
• How would you rate the quality and value of the final system?
• How did the project manager and craftspeople act during the project? What were the firm’s greatest area of expertise and least experienced area?
• What differentiates the company from others you’ve worked with?
As you go through the case study materials, consider some key internal questions:
• Can you see similar traits to your teams in the clients the firm serves?
• Does the firm’s expertise and problem-solving capabilities translate to the sorts of challenges you experience in your plant environment?
• Can you see a consistent project management approach demonstrated in its work?
Photos and videos provided by the firm can give you valuable information. Has it worked in plant environments similar to your own? Do the images give you a sense of finished system quality? Do they emphasize the company’s safety practices and job approach? Diving into past work may require a phone call, not just reviewing photos and written materials.
2. Bring in technical and commercial stakeholders.
Involve both your engineering team and your commercial stakeholders to evaluate quotes and capabilities. A successful project for you means satisfying many groups when it comes to finance, production and engineering. Having these stakeholders participate in the vendor selection process is critical for building early buy-in and ensuring you select a vendor that can meet everyone’s needs. Forwarding a presentation isn’t the recommended method for this. In-person meetings allow for dynamic interaction where stakeholders can ask questions and dig deeper for clarifying information, enabling them to better gauge true capabilities.
3. Visit the contenders.
If possible, travel to the sites of your top candidates. You want to see each’s facility but you also should get to know the company’s leadership and learn how the firm operates. Are you just another project or are you important to them? A company’s values, culture and internal processes can be just as relevant as its technical capabilities. A clean shop filled with empowered, passionate employees often can provide a much better service experience than a top-down autocratic outfit.
While this vetting process might seem expensive, it represents a small investment of time and money compared to what you’ll spend on the pilot plant. The right design/build partner makes the difference in whether your project is on time, on budget and ultimately successful. Select a partner that can genuinely tell you the stories behind how it solves challenges with pilot plants — the kinds of challenges relevant to your operation.
MATT BENZ is principal project manager for EPIC Systems Group, St. Louis, Mo. Email him at [email protected].