Pilot Plants: Strategies to Enhance Design-Build Quality
May 29, 2017
This project for a major oil/chemicals company, used objective-oriented project specs, plus codes and standards to control quality.
Caution should be used in the application of corporate engineering standards for components, materials, and methods of construction to pilot-plant projects. The application of inappropriate standards or the misapplication of standards can result in unnecessary cost and schedule delays without improving the quality of the pilot plant in a meaningful way. In fact, the inappropriate use of commercial plant specifications may limit the options available to the plant designer to the point where key pilot- plant project goals may themselves be compromised (Dukhedin-Lalla, 2003).
Pilot and demonstration-scale plants typically have a design life considerably shorter than that of a commercial facility. There is also typically a risk component to the project, meaning that some or even many items may need to be removed and replaced after initial start-up to accommodate the needs of the pilot testing program. A shorter design life and the enhanced need for flexibility toward future change should both significantly affect the decisions about what methods of quality control and inspection are used on the project. It should also affect the methods and materials selected for the pilot plant itself.
I’ve given a few examples of decisions which could be entirely out of place for a commercial plant, but which make a lot of sense for pilot plants.
Avoid Painting Carbon Steel Pipe
If exterior corrosion is an issue, consider using non-metallics, galvanized steel, or 304 stainless steel with 150# fittings. Done properly with a blast and then multiple coats of properly selected industrial coatings, painting pipe is both expensive and extremely disruptive to the production of small plants. The labour and schedule savings from eliminating it will almost certainly pay for the additional cost of stainless steel pipe and fittings for line sizes below 3” NPS. Where sch10S pipe and fittings may be used, stainless steels may be a cost effective substitution for painted carbon steel pipe even up to 6” NPS. Of course this is not a viable solution where exterior chloride exposure to hot lines is an issue, such as in some facilities located outdoors immediately near a seacoast. Under those installation conditions, the risk of chloride stress corrosion cracking of hot lines exists and may dominate the decision.
Use Tubing for Process Lines
When you do appropriate line sizing on a pilot plant, you will often find that the required line size is smaller than ½” NPS pipe. While small pipe might seem like an acceptable alternative, when lines on a pilot plant are made larger than necessary, the dead volume in piping can become significant, affecting both the quality of results and the speed with which meaningful results may be obtained.
We use stainless steel tubing and compression fittings up to ¾ inch OD, which has an ID approximately equal to ½” pipe. Tubing and compression fittings have numerous advantages over pipe for pilot projects:
- bends replace most elbows, reducing the number of joints and leakage points, or welds and associated NDE if welding is required
- every joint is a union, permitting easy disassembly and reconfiguration
- leakage integrity is higher than that of threaded joints, particularly when temperatures are above 175 C
- lines can be efficiently field run, rending isometrics unnecessary
- speed and labour productivity of fabrication is dramatically higher than that of pipe
Note that many of the benefits of tubing can evaporate above ¾” OD, and for linesizes even smaller than ¾” OD for exotic materials. We frequently switch to piping valves at even the ¾” OD size, using male connectors to adapt these valves for use in tubing lines. In many cases, piping valves at ½” NPS size are superior in terms of both performance and price to those available with integral compression fitting ends. Compression fittings- particularly tees and crosses- and compression fitting valves in exotic materials increase dramatically in cost as the size of the fittings increase.
We’ve also found more than sufficient evidence that the product of most of the major two-ferrule compression fitting manufacturers is not only more or less equivalent in terms of meaningful metrics of quality, but that these components are also interchangeable and intermixable without meaningful degradation in the quality of the assembled joint. This opinion is based on our review of significant 3rd- party testing. The one exception is the Gyrolok fitting made by Hoke, which has a different ferrule geometry rendering it incompatible with the industry standard 2-ferrule design offered by such companies as Swagelok, Parker, Ham-Let and others.
We obviously buy a lot of stainless steel tubing, and have done for decades. We have found that welded seam stainless steel tubing can be obtained with suitable quality for use with compression fittings, at a small fraction of the cost of seamless tubing. We do stock both based on customer preferences, but when given a choice between the two I personally will select welded seam tubing nearly 100% of the time. When selecting seamless tubing, there are very few mills we will buy from. We have encountered problems due to poor cleaning and annealing practice at the mill which become obvious only after cutting full random lengths in half and inspecting the bore. We recommend this inspection be done on a representative basis with every batch of seamless tubing.
Don’t Fear Threaded Pipe
For line sizes 2” NPS and below, and for service temperatures 175 C (350 F) or below, we make extensive use of NPT threaded pipe and fittings. If the right pipe thread sealant system is selected, threaded pipe offers significant benefit in terms of labour productivity and ease of re-work for future modification, the latter point being absolutely key for pilot operations. We use threaded pipe and threaded joints in this range for both nonhazardous and moderate hazard duty services, including flammable and moderately toxic services, and have done so with good success for over twenty years. Above 175 C, we tend to limit our use of threads because the thread sealant system options diminish. Note that the most effective thread sealant system uses both a bulk gap-filling and high pressure lubricant material (i.e. teflon tape with a specific gravity of at least 1.3), and an anaerobic pipethread sealant such as the many excellent sealants offered by Henkel under the Loctite brand. We have found that using either sealant on its own results in inferior performance to the use of the combination. In particular, the anaerobic pipethread sealant provides thermal cycling resistance which is extremely important to the long-term leak tightness of the resulting NPT joints.
Above 2” NPS, for nonhazardous services such as air, water, low pressure nitrogen, drain and vent, we use galvanized pipe and roll-grooved fittings such as those offered by Victaulic, Shur-Joint etc. This provides external corrosion resistance without the need for welding, blasting and multi-coat paint systems.
Heat Tracing and Insulation
On most pilot plants, line sizes are small and heat loss can be critical to successful operation. The main challenge with heat tracing and insulation isn’t installing it in a neat and professional manner in the factory. Rather, the issue is with ensuring that whatever system is used can be easily and completely removed, modified and replaced correctly by the operations and maintenance crew. We have over the years come across a number of insulating methods and materials which make removal and replacement easier and more likely to be complete. However, on a project where it is important to keep small lines hot, the best piece of insulating advice we can give is to install the pilot plant in a building so that the key problem- the ingress of rain and snow- can be controlled without requiring extensive and labourious re-work every time a hot line needs to be removed to clear a blockage, or altered to provide a new feature.
Electrical Area Classification
This topic may serve as the subject for one or more future articles. We have frequently seen a pattern of mis-application of electrical hazardous area classification with respect to lab- and pilot-scale projects over the years. On many pilot-scale projects, the money spent on matters related to electrical hazardous area classification generates extremely limited benefit in terms of meaningful safety improvement for operators. Alternative methods may offer superior, effective protection for a much lower cost. In some cases, a misunderstanding of the requirements and scope of protection offered by electrical area classification has resulted in an increase in real operational hazards. A thorough review and evaluation by people knowledgeable in detail about the requirements and scopes of NFPA 496, NFPA497, NFPA 70, API500 etc. and their application to pilot plants is recommended any time a pilot plant processing flammable materials is being considered. Zeton definitely has the skills and experience to help with that evaluation.
Zeton designs and builds custom projects for our customers, and is flexible to using the methods, materials, suppliers and processes that our clients value most. That said, in our experience, even the largest oil and chemicals companies do not have appropriate and fully developed specifications optimized for pilot plant projects. Instead, what many large companies have is corporate standards intended primarily for equipment and piping with a very long design life. These specifications can sometimes contain extremely valuable advice based on hard lessons from the company’s past. However, they can also tend to be excessively complex, and self-referential in a web which ultimately requires access to the entire set of corporate specifications and design guidance documents to be truly useful. A proper review of the entirety of a set of corporate standards may take many man-months of labour, which frankly you should expect to pay for if you expect them to be followed in detail. Corporate specifications of this sort are almost never appropriate to the size, scale and design life of a pilot project, and frequently include requirements which scan substantially drive up the cost and schedule of a pilot project which will seldom if ever bear fruit over its design life. They should be used in an extremely judicious manner.
Using the Pilot Plant Designer’s Expertise to Best Effect
If you’ve selected a specialist company like Zeton to design and build your pilot plant, you’ve made a good choice. You’ve selected the benefit of working with a company with 750 projects of a similar scale in their experience list, and a host of repeat customers. That scale-specific experience is invaluable to a successful pilot plant project. But to take maximum advantage of that benefit, you have to be careful about how much you tie the hands of the people who design and build such systems for a living. Be extremely careful what you put into hard specifications on a project. Instead of focusing on telling us how you want a project done in detail, it is better to spend that effort to transfer the learnings from the lab work to the designers, to set appropriate objectives for the resulting pilot plant, and reviewing the results of basic engineering and the scope of work for the detailed design and fabrication of the project. By focusing on objectives rather than itemizing and specifying the details, the pilot plant designer can make best us of their experience help you make scale-appropriate selections of materials and methods to best match those objectives.
Conclusions to the Series
A pilot-plant project has a different design data source and different objectives, scale, lifespan, operational conditions, and product than a commercial plant project for the same technology and therefore should be given a separate, distinct design and project execution approach.
Design for operational flexibility and rangeability is key to pilot-plant success. Multiple operating points and parametric design established through the use of lab experiments and simulations are the rule rather than the exception.
Natural break points for temperature and pressure arising from materials selection should be taken into consideration for design, and these breakpoints should not be crossed without due consideration. Particular care must be given toward the selection of operating temperatures for titanium equipment, as minor changes in temperature can have significant impact on required titanium wall thickness.
Notably, some design factors change with scale while others do not. Accordingly, some unit operations of the commercial plant, plus some methods of fabrication and materials of construction may be limited to minimum physical size or throughput scale and require alternative approaches to be successful at the pilot scale.
In particular, the small size, significant complexity, and tight schedule of pressure hydrometallurgical pilot-plant projects make them ideally suited to a modular, skid-based design-build approach. We at Zeton have found tremendous success using this method, and found it to be an invaluable way to decrease costs and increase efficiency over the course of a project.
1. Dukhedin-Lalla, L. (2003). Pros and cons of applying plant specifications to pilot plants. In D. Edwards and C. Teich (Eds.), Pilot Plants in Process Development Topical Conference Proceedings (pp. 213–21). San Francisco: American Institute of Chemical Engineers.