The customer insisted the process was sanitary and the company at which I was working bid on that basis. However, the process actually wasn’t sanitary; we wasted lots of money and suffered loads of aggravation proving that.
In this case, the process had several telltale signs that it wasn’t sanitary. The clean-in-place (CIP) procedure involved little more than a hot water rinse, and the reactor only had a mill finish. The process used an explosive powder and flammable liquids. In addition, the product was thixotropic and required pressures in excess of 150 psi; almost all sanitary systems operate below 150 psi — sanitary valves normally aren’t designed to operate above that pressure.
A sanitary process typically demands: 1) quick-disconnect (e.g., tri-clover) clamps; 2) sanitary butterfly or diaphragm valves — surprisingly, ball valves aren’t allowed and, not surprisingly, gate and globe valves aren’t sanitary; 3) tubing made of type-316L stainless steel with 25–32 RA (roughness average) finish on the inner diameter and 30 RA (No. 7 finish ~150 grit) on the outer diameter; 4) a multiple-segment CIP process generally involving a hot water rinse, caustic wash, biocide treatment and a final hot water rinse or sometimes an injection of low pressure steam; 5) lobe pumps — some types of pumps are banned because their castings don’t meet finish requirements; 6) elimination of dead-legs, for example, by using brewers tees; 7) polishing even of utility pipe/tube to meet exterior finish requirements to avoid crevices for “bugs” — this also applies to valve actuators, which must be covered; 8) pipe hangers and supports that are rounded and polished smooth to a 30 RA exterior finish — ideally, with surfaces sloped 1–5% to prevent water and dust accumulation; 9) tubing that is sloped and designed for disassembly for clean-out-of-place (COP) procedures; and 10) a CIP/COP verification procedure known as the riboflavin test. These requirements apply for human and pet food, cosmetics and some consumer products. Pharmaceuticals call for more elaborate standards; materials that come into contact with blood must comply with even more stringent mandates.
Always keep in mind that sanitary tubing is much weaker than conventional pipe and isn’t covered by ASME/ANSI standards. Vendors establish their own maximum operating limits for tri-clamps — there’s no uniform standard like that for a 150-psi ANSI flange. In a tri-clamp, a clamp holds an O-ring in place against a groove in a ferule butt-welded to the tubing or pipe; socket welds are discouraged in sanitary work. One or two wing-nut bolts tighten the clamp over the ferrule. Tri-clamps come in three types: light duty (<150 psi); heavy duty; and high pressure (>500 psi). Only the high pressure clamps use two bolts.
Because tri-clamps generally are made of type-304 or type-316 stainless steel, their mechanical strength steeply declines with temperature. One vendor’s 3-in. clamp was rated at 350 psi at 70°F and 150 psi at 250°F, or about 218 psi at 150°F, the operating temperature of our process. Fortunately, hydrotesting typically is conducted at about 70°F. However, you must consider mechanical strength at design pressure, which usually is at least 30 psi higher than operating pressure. For high temperature operations, increase the design pressure on which the hydrotest is based to ensure the system is mechanically sound.
Sanitary tubing has a smaller inner diameter than a comparable pipe. If you’re pumping a pseudo-plastic or thixotropic fluid, the pressure spike may be significant. That’s why sanitary services commonly rely on lobe pumps, often when centrifugal pumps would make more sense; lobe pumps have casing pressures up to 450 psig.
These high pressures pose a significant challenge during hydrotesting of sanitary systems. In one case, I was told that a vendor’s butterfly valve was “mechanically unsound.” I had considered adding air to a vent to keep the valve closed during hydrotesting. Typically, 2 psi of air can provide an additional 10 psi of backing pressure for a valve spring. As a result of this information I replaced the sanitary butterfly valve with a “sanitary” ball valve rated at 1,100 psig. As I said, the process wasn’t really sanitary.
Blindly assuming that a process is sanitary may result in unnecessary expense and frustration. Always question, gently, the design basis.
DIRK WILLARD is a Chemical Processing contributing editor. He recently won recognition for his Field Notes column from the ASBPE. Chemical Processing is proud to have him on board. You can e-mail him at email@example.com