My brewery anaerobic digestion experience; part II

By February 17, 2015Anaerobic, Mechanical, Treatment

This is part II of a 3 part story.  The first part can be found here.

The Newbio system was a downflow system; put water in at the top and remove it at the bottom- opposite of how most anaerobic systems work. In order to not lose biomass they had a bed of sand at the bottom of the tank acting as a filter. Kind of like the grain bed in a lauter tun. Beneath the sand was a network of slotted PVC pipes that the filtered water would enter and then head to the sewer. Like all filters, the sand bed would plug. To clean it there was a ‘fluidization bar’ in the sand. Simply put, a 30 HP pump was used to pump water into the bar and out jets spaced along the bar. The idea is the fluidization bar would move through the sand like the minute hand on a clock, fluffing up the sand as it went. OK, now think about this. Sand. Moving parts. You know where I’m heading.

Obviously, erosion of various components was a huge issue with this system. And it was just one design flaw with this system. The entire system was stainless steel. After a month of so of operation, many of the cast stainless fittings on the fluidization bar were eroded smooth. Several times fittings failed due to corrosion. Of course, the slotted PVC pipes just beneath the fluidization bar were prime suspects for erosion as well. And they did. Once a hole was worn through the PVC. There were several other close calls where there was significant erosion but caught before it failed.

Aside from erosion, there were many other reasons for the failure of this system. Here they are in no particular order.

  • This Newbio’s first full sized installation. They had a 6 ft tall trailer mounted pilot unit, the tanks at Deschutes was 24 ft tall. When the sand bed started to plug on the pilot unit, the fluidization bar would have no trouble moving through the sand. However on the full sized tank, when the sand bed plugged all of that water was sitting on the sand. The pressure differential caused the sand bed to collapse into very hard sand. The pump/fluidization bar had to run more than designed. This lead to excessive wear on everything. Which of course lead to many, many failures.
  • Sand choices. The bottom of this tank was 2 flights of stairs below grade. There was 7 tons of sand in the bottom of the tank, and I think we changed sand size 8 or 9 times. Sometimes Newbio would think the sand is too fine, or too coarse. Once they even put in crushed sandstone, which when it got wet broke in to very fine pieces and actually floated. Emptying this 30,000 gal tank and replacing this sand was a nightmare that we got to relive a few times a year.
  • Poor electronics. Basically the control system was a hodge podge of ‘Radio Shack’ grade electronics. Yes it had a PLC, datalogger, and autodialer- but they were junk.
  • This anaerobic digester was installed in a building, and a very nice building at that. However this system generates a lot of methane so everything in the building was designed explosion proof. A big expense and recipe for disaster..
  • Performance contract. A lot of you are probably thinking about this. I can’t talk a lot about it because I don’t remember the details. But I do remember the contract was weak. I know some of the games they played were putting clauses in the contract about ‘treated flows as small as 2000 gallons per day’ instead of ‘treated flows of at least 20,000 gallons per day’.
  • The jets on the fluidization bar were very low tech. It was all stainless steel, the jets were 90 degree elbows and a hose barb screwed in to it. If the jets weren’t pointing in just the right direction (they never were), the sand would be blown off course. Eventually the sand made a pile in the middle of the tank, and a ring around the outside of the tank. The slotted pipes in between these piles were exposed. Of course we had no way of knowing, and to inspect it meant removing 30,000 gallons of biomass and 14,000 lbs (dry weight) of sand. And the bottom of the tank was in a basement down 2 flights of stairs!
  • The basic concept behind what to treat was flawed from the start. Deschutes had 2 waste streams, high strength and low strength. High strength was hauled off site as fertilizer. Low strength was sent to the City sewer and as such was regulated for pH, loading, and received surcharges. The system was designed to treat the low strength wastewater. The cost of trucking the high stength material was significantly higher than the surcharges for the low strength material. Poor engineering by a high dollar engineering firm (Brown and Caldwell)
  • Here’s another thing that even I have a hard time believing, but it’s true. This was 1998-2000, before the days of fast internet and email. The Newbio guys were often not in Bend, but programming changes needed to be made to the Radio Shack PLC that was used to run this thing. I didn’t know C code then, and I still don’t now. They couldn’t email a new program, so I had to sit on the phone for hours while the Newbio guy on the other end said stuff like “C : / / * pump run *” I had to type that in to the computer letter by letter. 1 typo and the program wouldn’t work. Nightmare…
  • Of course there are 2 sides to every story. Newbio knew I wasn’t happy with them, and knew I would tell anyone who listened at Deschutes what was going on out there (owner, senior management, etc.). Of course Deschutes just paid a lot of money to install this thing, and it’s not working. It never worked. It’s not solving their problem, it’s creating many more new ones. Anyway, in their desparate clinging for anything to hold on to this job, Newbio blamed the failure of the system on me and tried to get me fired.
  • As this blame game was happening, Deschutes told Newbio to bring in their own operator to make it work since John can’t. They brought in a college student on summer break to operate the system, that was their big fix. Since Newbio brought in an operator, this freed up my time to focus on what to do when it’s proven that it doesn’t work. And of course it didn’t work. I came up with a very simple design using mostly existing equipment that we could retrofit.
    • As it was the wastewater tanks at Deschtues were somewhat small based on the discharge limits we had. So basically wastewater had to go out the day it was sent down the drain since we didn’t have storage capacity. But we had this 30,000 gal tank (digester) that would be mighty handy for storage. In short, Deschutes went from discharging 5 days a week to storing wastewater and discharging over 7 days. That little change helped immensely. Deschutes is still operating on this system I designed way back in 2000. They’ve added more storage capacity and increased pipes and pumps, but the concept is the same and it’s still working.

My anaerobic experience

What is the point of all of this? Venting? Therapy? Just kidding.

This is the reason I am here doing what I am doing. Don’t repeat the same mistakes, and there were more, thankfully much smaller, mistakes made along the way. Don’t reinvent the wheel on something as complicated as this.

Going with new technology certainly has a ‘cool’ factor. New technologies can work, but the question is who wants to pay to be a guinea pig and what are the risks if it fails? Contracts aren’t written for the marriage, contracts are written for the divorce and make sure your contract with any supplier reflects that. One of my biggest pieces of advice is when looking at equipment for your brewery, ask for brewery references and actually call the brewery for feedback on that piece of equipment. I know in the case of Newbio, they actually installed another one of these digesters at a brewery- I think in Florida somewhere. That brewery never called Deschutes to ask for feedback on the system, which blows my mind. Did they not ask, or would Newbio not truthful and not provide a reference? Either way that was a big mistake on that breweries part.

Click here for the grand finale.

 

Share This