When New Belgium Brewing Company needed new quarters in 1995, sales were frothing with double-digit growth every year. Barrels (in that year, 130,000 of them) were rolling-out to delivery fleets. Already, orders for its popular Fat Tire Amber Ale and Blue Paddle Pilsner were ranging well beyond the brewery’s home in Fort Collins, CO.
But quick success brought growing pains, in the form of a $2 million-plus wastewater bill in 1999 and an urgent need for several quick remedies. NBB’s environmental engineer, Fred Porter, recalls that “No one had any idea we’d grow the way we did”—and yet, all of a sudden, “we were asked to pay this ‘plant investment fee’ [PIF] by the city utilities department,” and simultaneously socked with a nearly eight-fold wastewater treatment surcharge—both huge hits, of course, for a small startup less than a decade old.
How the mega-increases came about is a story in itself.
From its outset, the brewery had been charged under the city’s standard rate for industrial water users with a 2-inch water tap. This assumed that less than 20% of incoming water was consumed, while the other 80% was discharged for city treatment. Based on this average, New Belgium’s monthly charge was $2,000.
Then, a conscientious but perhaps overzealous company engineer thought that perhaps the discharge figure was unfairly high—reasoning that a brewery actually consumes not 20% but about 50% of its water, shipping it off in the form of beer products. So, shouldn’t New Belgium get a break on its sewer?
A city engineer agreed to investigate. Meters were installed.
Unfortunately for New Belgium, results showed brewery waste bearing biological-oxygen-demand (BOD) at 7,000 ppm, with total suspended solids (TSS) at 1,500 ppm.
Even more alarming: pH measures were gyrating from 2 to 12, thus endangering the city’s concrete pipes and treatment process.
Conclusion: NBB’s outflow had obviously exceeded what the initial PIF had covered—and hence a $2 million upgrade loomed.
Obvious moral: don’t rock the boat!
At any rate, it fell to Porter to figure out how best to respond.
Rent, or Buy?
Two million dollars were more than New Belgium could afford at once, and so, for an interim, the city agreed to lease the brewery some added plant capacity and let them pay over time while they researched and deliberated on a long-range fix.
Step one, then—and bearing some urgency—would be to install a pH neutralization system. Two mixing tanks and chemical dosing equipment would readily solve this. Wastewater could enter and be neutralized with either acid or caustic, to maintain the pH between 6 and 9, as the new industrial wastewater discharge permit now required. Readings would be taken constantly and automatically, with daily highs and lows logged for a monthly report.
That done, New Belgium could look over its several options for wastewater treatment and discharge.
Two million dollars will buy a lot of treatment plant, and so this option, as opposed to paying the POTW, was immediately attractive. New Belgium already owned sufficient land for it. And a quick calculation showed that a 7-figure investment would be recovered just in a couple of years. Thereafter, future savings would amount tens or hundreds of thousands of dollars a year.
Among the several options that looked appealing and made sense initially was to build a simple aerobic system for treatment and continue discharging treated water
to the city.
However, after some further investigation, an even better alternative emerged, which was to discharge the treated outflow into the nearby Poudre River.
Thus, a treatment plant consisting of an anaerobic reactor, an aerobic basin, and a gravel filter for “polishing” the final effluent was installed. The plan called for discharging to the city for a brief interim while the brewery waited for permits and rights-of way for the river pipeline.
Unfortunately, obtaining easements turned out to be, says Porter, “logistically much tougher” than expected.
And, at about this time, another Colorado brewer, who was discharging into Clear Creek, suffered an unexpected high-volume beer spill that killed a lot of fish. The disaster put a damper on New Belgium’s plan, and the brewery reversed course.
The upshot has been that, ever since, the treatment plant “has not really been pulling its weight,” says Porter. “It saves us some money in our multiplant investment fees,” but was not lowering monthly sewer fees.
Wishing to be environmentally responsible, New Belgium next came to a better appreciation of how investing in anaerobic treatment might accomplish this in two dramatic ways:
First, the brewer’s high-BOD waste could be anaerobically reduced to biogas (primarily methane) and captured as a renewable fuel source for heating and power.
And second, this new anaerobic reactor would relieve the city of a considerable energy expenditure it might otherwise require for aeration downstream.
All in all, the “green” approach suited the company philosophy, despite the added expense and effort this would incur.
Further Design/Build Factors
Another prime consideration was the specific composition and optimum treatment of “brewery” waste: its organic and inorganic solids, fats, oils, and grease. Relative compositions of each would determine the system type, tank sizing, bacteria selection, and detention time requirements.
Thus, the brewery staff engineers concluded, operations dictated designing an anaerobic plant to handle volumes of about 410 cubic meters per day.
Ample landscape allowed for construction of laterally extensive tanks—a relatively lower-cost design feature.
As for similar brewery role models to emulate elsewhere, New Belgium’s concept of anaerobic treatment and gas recovery for power generation, in combination, was still rather cutting-edge stuff; the company didn’t find many examples to follow. A few were doing this, including a Coors plant; the Red Hook Brewery in Portsmouth, NH; and a Budweiser plant in Houston; but initial costs in these systems were very high. New Belgium was still less than a decade old, and so there was reluctance to become overextended for a big-ticket facility.
For an affordable alternative, NBB learned of a system designed by a German vendor, Von Nordenskjold, of Munich. An attractive bid came in, and a deal was quickly inked. The design conveniently included a turnkey methane capture system for piping gas from the anaerobic plant to a 300-kW Continental Energy Systems combined heat-and-power generator, imported from Belgium, which could yield around 15% of the brewery’s electrical needs, and could amply warm the wastestream for its anaerobic processing
Von Nordenskjold’s design ultimately consisted of, first, a pretreatment or buffering basin of 500-cubic-meter capacity; second, an anaerobic reactor of 1,500-cubic-meter capacity; and third, use of the existing aerobic pond.
Construction and commissioning were thus completed in 2002.
In the resulting treatment process, high-BOD brewery wastewater first enters the 500-cubic-meter basin for pH balancing to 5 or 6. New Belgium’s plant operator, Brandon Weaver, explains that, at this optimum acidity level, “we’re starting to form the volatile fatty acid chains that become important food sources when this water gets added to the anaerobic bacteria in the second basin.” This first stop also helps moderate flows to the reactors—water retention time being critical to success in both tanks. For pH buffering to work, a full day’s holding time is desirable. As for the reactor, says Weaver, it likes “five or six days” and a minimum of three.
At the digester’s back end, a lamella plate separator retains the anaerobic sludge as the outflow exits to stage three, the aerobic pond.
There, remaining ammonium undergoes nitrification and denitrification. Weaver explains: “We’re targeting specific bacteria to perform nitrification and denitrification, so we keep adding oxygen and taking the ammonium through that nitrification process all the way to nitrate, NO3.” Next, the nitrified water contacts anoxic zones within the basin for denitrification. “It’s just a matter of cycling on and off our aeration chains,” Weaver says.
From start to end, roughly 500 cubic meters enter and exit every day.
Mixed Results
As noted previously, initial BOD-levels start at about 7,000 ppm and TSS, 1,000; by the time the fully polished water leaves for the sewer, BOD and TSS are each reduced to 20 ppm or less.
Doing the lion’s share is the anaerobic step, which provides, says Weaver, around 84% “of the overall reduction of our load” of BOD, chemical oxygen demand (COD), and TSS “across the plant.”
Plant operations are largely automated. Logic controls in valves and pumps yield real-time data on flows. These and other devices are integrated with the brewery itself, via an Opto 22 control system and Wonderware HMI interface.
On the waste side, temperature and pH are monitored and logged most intensely, along with outlet and inlet flow rates and redox potential in the aerated basin.
Under the brewery’s sewer permit, monthly reports must be provided, showing daily minimum and maximum pH, daily flow volumes, and weekly BOD and TSS taken from a 24-hour composite sample.
At a cost of roughly $5 million (including the power-and-heating plant), New Belgium’s investment, says Porter, “has pretty much paid itself off at this point. We were able to reduce our monthly fees right away, and we basically saved the [$2 million] in plant investment fees immediately.”
Since 2002, sewer surcharges would have escalated in steps from the $15,000 increase that came in 1999, to $18,000, and then as high as $25,000 monthly in 2006. Instead of this, though, monthly bills are currently only $3,000.
Subsequent PIF for the increased wastewater demands—which Porter says would otherwise amount to an additional “five or six million dollars, if not more”—were looming eventually, too.
All in all, then, in terms of cost-benefit payback, purchase of the anaerobic reactor “really makes sense. That’s a simple slam-dunk.” By contrast, the heat-and-power cogeneration from methane has proved problematic; the investment has not, to date, provided the anticipated return.
Shortcomings Soon Emerge
New Belgium’s beer sales kept bubbling along after 2002 in double-digit annual growth rates. Since 2000, total yearly barrelage has doubled to 360,000. Coming out of brewhouse obscurity, New Belgium now ranks third among the nation’s 1,300 or so “craft-brewers” (i.e. microbrewers) and is No. 11 among all beer makers, according to industry figures.
Just as the beer has flowed, so too has BOD-laden waste. Thus, beginning in 2005—only three years after the first treatment plant’s commissioning—wastewater volumes were topping 550 cubic meters per day, or 30% more than the plant’s 420-cubic-meter design capacity. “Hydraulically,” Weaver observes, “we’re pushing the plant much harder than it’s designed to do.”
Several crucial deficiencies in the original plant concept have also come to the fore, notes Porter. For one, there’s been a buildup of inorganic solids stemming from the diatomaceous earth (DE or, in brewing parlance, kieselghur) used in filtering beer. This has proved “tough to cope with, because the anaerobic reactor can’t be taken off line” for cleaning, he says. In fact, there’s no backup system or alternative means of processing which will allow for periodic maintenance or other needed shutdowns; wastewater must be shunted straight to the sewer. That’s a worst-case solution, because, thereafter, several months are required to restore the sludge biomass and enable full operations. By contrast, an aerobic plant, despite its high energy consumption, would come back online much faster.
A second problem is that bacterial sludge has proved hard to retain in the anaerobic reactor. Sludge, of course, enables anaerobic processing. The anaerobic sludge works best in a “steady state” system, Porter notes. Unfortunately, the steadiness here has been impaired since the recent construction of a larger brewhouse: Actual brewing operations declined from the previous hectic 5 to 6 days a week to a more leisurely 3 to 4. This makes a “feast or famine” situation for the anaerobic sludge.
Still another issue to emerge is the sizing of the lamella separator. On several brewing days a week, it’s being hydraulically overloaded. To compensate, a delicate balance must be maintained between wastewater inflow and quiet retention time—otherwise, the sludge won’t thrive. Sludge does well having five or six days to work; unfortunately, because wastewater keeps surging in, retention times rarely meet this optimum. In fact, strong flows push sludge out. Plant operator Weaver explains: “Instead of building up excess bacteria as anaerobic systems typically do... we’ve actually had to plant or seed new anaerobic bacteria on a regular basis over the last year or two to keep things functioning.”
Despite such challenges, treatment has somehow managed to stay within its tolerances. But, clearly, this situation couldn’t last long, and within a couple of years Porter realized that a larger anaerobic reactor was urgently called for. Besides which, he adds, “Buffering basins were needed at the head of the plant,” and clarifiers were indicated for the aerobic portion.
Lastly, as the upgrades were being contemplated, a top priority was the avoidance of future PIFs—which otherwise might reach multimillions of dollars.
Design, Selection, Construction...
For the second anaerobic system, Porter decided, “We basically wanted something pretty tried-and-true and simple to operate, because we are really in the business of making beer,” rather than wastewater treatment.
This time, in expanding his search for anaerobic options, Porter says he found hundreds of them, some of which work and some of which don’t. “It’s pretty hard to pick the right one.”
He thus retained as independent advisor a professor of wastewater treatment, Luc Geuens, of Belgium, who had earlier trained New Belgium’s technical director, Floris Delee.
Porter himself also enrolled in Marquette University for its highly regarded courses on anaerobic science.
Geuens assessed New Belgium’s operation and recommended a Belgian vendor, Sanotec; Porter sent a formal request for a proposal, inviting two other bidders as well—Von Nordenskjold and a Milwaukee firm called Triad Engineering Inc.
Von Nordenskjold essentially proposed no change in the technology but a larger version of the current plant. Triad bid on installing an expanded granular sludge bed reactor (EGSB), and Sanotec proposed installing a well-established upflow anaerobic sludge blanket (UASB).
Competitive bids were all very close, coming in at around $3 million, with just a 5% difference between them, Porter recalls.
Von Nordenskjold’s proposal was discarded first, as being clearly unsuitable to the growing treatment challenge.
Thus, the deciding issue came down to rival technologies: EGSB vs. UASB. Porter carefully considered Triad’s novel and arguably more “cutting edge” EGSB. It offered a higher flow rate than UASB and a smaller footprint. However, operationally it looked to him more labor-intensive and less tolerant of errors or fluctuations—sensitivities he wished to avoid. “We wanted something simple to operate,” he says; “not to have operators making changes all the time.”
He opted for UASB.
He briefly considered products from two well-known UASB market leaders, Biothane (Camden, NJ) and Paques (of the Netherlands, represented by USFilter, now known as Siemens Water Technologies). From the latter, a higher-rate BIOPAQ IC (internal circulation) reactor was also scrutinized. It offers retention times of just a day or two; Porter worried, though, that it, too, might be fault-intolerant. Since New Belgium’s reactor would be producing at a high rate, there also seemed a greater potential for an overly temperamental plant. If an operator erred, loss of sludge—and resulting downtime—might easily happen.
Ultimately, the slower-rate UASB was clearly the more comfortable choice. Sanotec got the contract. Porter also liked the strong warranty terms covering the whole system’s performance to specs. “That was very important to us this time,” he says, in the wake of some unpleasant previous lessons.
Finished Plant: Adaptable, Better-Suited, Better-Performing
Designwise, the new three-stage system adds several key functions:
First, replacing the 500-cubic-meter, single buffering basin are two holding basins of 600 cubic meters each. One, an equalization basin, performs virtually the same function as the old buffer, but its larger size ensures adequate time for pH balancing. Greater volume also ensures steady-state operation of the UASB, which will allow for better control and less disruption.
Next to this basin, a second of the same size will serve as a “calamity tank” in case of any emergencies, says Weaver. For instance, if inline readings indicate pH too high or low, flows will be shunted to this emergency basin to allow additional time.
Third, of course, an entirely new UASB replaces the anaerobic reactor. With its compact, vertical shape and design, it requires a much smaller footprint. Although its tank is 20% smaller than the former (800 cubic meters vs. 1,000 cubic meters), the new UASB should significantly outperform the old, thanks to a better flow distribution and the ability to keep inorganic solids out by diverting high-solids water to the emergency basin for settling.
Thus, he believes, a more-than-adequate supply of anaerobic sludge now seems assured for some time to come.
Another Belgian firm, Stabos, designed, built, programmed, and did all the process engineering for it, based on broad prior experience with a proven system.
As for overall payback, 10 years now seems likely, Porter says. “That’s not as rapid as we would like,” he concedes, but at least there’s plenty of built-in capacity to meet the brewery’s growth for five to seven years.
Overall, Porter’s experience the second time, with Sanotec, has been positive. However, cultural barriers did cause snags in the general contractor’s relationship with its US subs doing the installation. Under US work habits, he says, American firms tend to be “more rigid and expect more detailed design with less design work in the field....” By contrast, European contractors “leave more leeway in their design work,” and the generals expect subs to use more judgment and pick up the slack. Americans were not prepared for this lack of detail in the design package, and did not bid accordingly. “These differing approaches presented a problem and created considerable tension between the two firms,” he says, adding that, even when contractors execute scientifically based technical specs, “the human element is huge.”
The design specificity conflict also illustrates a broader lesson that Porter would pass along, regarding necessary tradeoffs. In a certain sense, he says, he was glad Sanotec didn’t squander another year in blueprinting ad infinitum—because New Belgium needed this plant in a hurry. On the other hand, though Von Nordenskjold would have benefited from more thorough planning and concept work: “Even if you have to pay extra on the front end for a tighter design,” he says, “you’ll get a much better project in the long run,” if not as speedily.
Speaking of which: Looking down the road, Porter thinks there’s a major opportunity at the brewery to do outright water reclamation for reuse—but this will first require painstaking research and analysis. Many restrictions apply to twice-used water. “The technology is there to easily reclaim our wastewater and make it pure enough to drink,” he observers. “But the fear of public perception has made us put it on the shelf for now.” Under Western states’ water-rights rules, reuse isn’t actually permitted yet, but times may be changing.
Elsewhere, if reuse is allowed, he suggests, a large-scale water user might be smart to consider it in light of large cumulative savings realizable on water, sewer discharge, and PIF. Other issues come into play, though: there’s dual piping... heavy sodium treatment needs... and intricate calculation of usable volumes. “It gets pretty convoluted,” he says.
For now, at least, there’s no big rush.
La Mesa, CA-based writer David Engle specializes in construction-related topics.
OW - September/October 2006 |
