Earth Consultants

Applying Lean Six Sigma to the Environment

E093: Lean and Green Project to Reduce Wastewater in State of Washington

16 min read

In today’s episode, I share a presentation from the IISE Seattle conference on May 23, 2022.

Joshua Fowler from Romac Industries starts off the presentation by giving some background on the project that I was involved with as the lean consultant, and Michael Johnson from Lean Environment Inc. discusses the technical solutions applied to reduce wastewater by 50%, with cost savings of $50,000 in 2021.

You can also watch the video of this presentation at https://www.youtube.com/watch?v=HIG_RtrcZ4I

Links

If you enjoy this podcast, please follow us on your favorite podcast app. Any ratings you could give us, or shares across social media would be greatly appreciated!

Promotions

Have you ordered the new book, “Lean Six Sigma for Good: Lessons from the Gemba (Volume 1)?” The book is made up of 8 chapters written about experiences from Lean and Six Sigma practitioners, to give you tips and tricks to help you work with nonprofits in your area. All proceeds donated to charity. We are also close to releasing Volume 2, so check back for the latest news.

Transcript

Joshua: So first and foremost, apologies. I’m not like a huge public speaker, so this is probably not going to be very formal. I’m just going to talk to you like you’re a regular group of people. But you are, but you get my drift. All right, so my name’s Joshua Fowler. I’m an environmental engineer for Romac Industries. I’ve been with the company for about three years, and this is an offshoot of a grant program that we were working with the Department of Ecology, which got me involved with Brion, and then eventually involved with Michael. And so this is kind of a waste minimization project that we’ve been working on that’s had some offshoots with some other production benefits and stuff. And then there’s my name, but then those two names next to mine, Frank, and Kim Howard, those are our two maintenance guys. Between them, we probably have about 35 years of passivation experience going on there, so I just want to mention them because they’re a very key component of this project.

So what is passivation? I guess, first, I’m going to start with why Romac passivates. So we passivate because we need corrosion resistance out in the field for our products to withstand the elements. Basically, what happens is the oxide layer of our stainless steel components is damaged during the fabrication process from welding, grinding, cutting, those sorts of things. What that does is it reduces the component’s resistance to the corrosion. Through passivation, we can remove contaminants that affect that corrosion resistance, i.e. Cause rust, that sort of thing, and leave behind other elements of the alloy. For stainless steel, that’s going to be mostly chromium and nickel. And then when those are exposed to air, it restores that oxide layer on the stainless steel components and helps get that corrosion resistance back in there so those components last in the field.

What we have pictured here is the view of our passivation process before we began the project. This is actually a mirrored image of the room itself, so we have this exact same setup on both sides of the room. We have two production acid tanks, and then a static rinse tank, and then a table where we also have a pressure washer hooked up for our operators to rinse off everything as it comes through the process. What happens is a component will come in. It’ll start and end up in one of these two production tanks, and then it’ll passive for about 15 minutes on average depending on the component. And then once it passivates, it’ll be lifted up, brought into the static rinse tank, and submerged in there for a few minutes, brought back up, and then brought to the pressure washer table, where it’s then sprayed off. Once that’s completed, it’s out the door for the rest of the production process.

A little bit more about the production tanks. Each one of the production tanks is basically a hydrofluoric bath, which we make with a combination of nitric acid, ammonium bifluoride, and water. Each one of the tanks is heated to help with the reaction process. The rinse tank is just normal city water, and then the pressure washer is also supplied by the city water. What would previously happen is, essentially, since the two production tanks are heated, they have evaporation, and so we would refill the evaporation, what’s come out from evaporation with city water, which eventually would dilute the tanks bad enough to where the batch times would be unacceptable for production levels, which would result in us having an acid spike every about six to eight weeks.

What that would be comprised of is we would pump both of these tanks out and send them off as hazardous waste, we’d pump this tank out and send it out as nonhazardous waste, and then we would take a whole new set of ammonium bifluoride, nitric acid, and we would re-spike the tanks. At the time, we were using a 67% concentration nitric acid, so it was a very dangerous process. We had to stop production. It was not a fun thing to do. The production guys that I mentioned earlier in full Level B hazmat suits with SCBAs. It’s not something that was very desirable for us to do. It’s like my house.

Now, this is the 2.0 version that we currently have within the building. It looks very much the same, but there’s some key things here that we’ve changed up big time. Probably first and foremost is, with the help of Michael, we realized that our static rinse tank was almost probably closer to being a third acid tank than it was a static rinse tank. So what we started doing was using the static rinse water to make up the evaporation from the production tanks, which significantly reduced our hazardous waste, helped with the bath variation. I’ll let Michael explain a little bit more about this 6 gallons that were taken out for the tanks in his portion of the discussion.

But now that we’re doing that, the recirculation, making up that evaporation, it removes the need for those large change-outs, and it’s also allowed us to go from 67% nitric acid to 20%. We have a manifold where we actually do mini spikes, and at the push of a button, we’re able to spike each one of the tanks. Part of the reason we went to the 20% is, obviously, if plumb in 67% and we have a leak, we have a very big issue on our hands, and now we have an inherently safer process and it’s helped out with our waste minimization.

I believe, according to my notes, and I was discussing this with Frank the other day, I mentioned that we haven’t been doing the large acid change-outs. We actually used to do those every six to eight weeks when we would change these out. We haven’t done that since September of 2021, so we’re already seeing some significant changes from that.

This is hard to decipher, so I’m just going to explain it, but this is just some information I got from our business analytics guy, and it shows some of the benefits of our change that we’ve already experienced. Here, you can see where our consumables for the room have gone down. In 2021, we spent 39% on consumables like the nitric acid, ammonium bifluoride, some of the caustic soda we use in the treatment process on the other side of the room for our discharge permit. And then also, probably the biggest one, and the part that Ecology loves the best, is that we’ve had a 50,000-pound reduction in hazardous waste just like that. We’re not having to get rid of those acid tanks anymore, so that significantly reduced the amount of hazardous waste we’re getting rid of, which, as it’s trending for 2022, to have a 50% reduction in overall cost of consumables for the room, especially waste costs, but then also a 50% reduction of hazardous waste generated from our Bothell facility overall, so we’ve really seen a lot of improvement in a short amount of time.

Just to segue into what Michael’s going to discuss, this is the step-wise improvement strategy that we’re working towards in the future, and this is what Michael’s going to discuss here in just a second. That’s my portion of it.

Michael: My name is Michael Johnson. I am the hired help for Josh and his team. I’m a chemical engineer and chemist by training, and I did my first Six Sigma work for IBM in the Motorola program in 1986. So I’ve been doing this for a while in terms of practicing this stuff, and I am still very much a student, so we’re kind of learning together.

There’s a few things I’d like to offer you as takeaways from this process. The first is eliminating variability in a wet process or any industrial process. For example, when you’re talking about your rinsing process in your PET, all of the things we’re about to discuss relevant to his wastewater process is perfectly relevant to how you’re using water in the PET recycling. So this is something you can take back to them and say, “Have you thought about counter flow rinsing perhaps, or conductivity control as a means of controlling the rinse process?” Anytime you see a wet process or an industrial process, from washing your dishes to washing your car to passivating large what are essentially plumbing parts, we could cover that, but Romac makes large plumbing parts, so that is something you’ll be able to take away with you.

These are the step-wise things from the Romac process. This is what we’re talking about. We talked about this big sawtooth pattern. That is normal in almost every industry in the world. That’s almost the starting point for every place you’ll walk it. As an engineer, you’re walking in and you want to help somebody establish process control. You see they’re using their acid bath; they’re dumping their acid bath every six weeks. That’s a business opportunity every time you encounter that. What we want to do, and we want to encourage you to do, is reduce the size of the dumps until you can eventually get to a steady-state process.

In Josh’s case, they’ve been dumping these large, very nasty tanks every six weeks, and the alternative to that is taking a small quantity of the acid out once every few days. At the current rate, they’re taking only 6 gallons out per week for each one of these tanks, and replacing that with fresh acid. Now, the normal person will say, “Hey, wait, I’m throwing away perfectly good acid.” Yes, you are, but you’re establishing equilibrium in the process and ultimately establishing a steady-state process, which will eliminate variability and improve the process.

The reason it improves your process is because when you have a sawtooth process like this, when the bath is brand-new, you only need 2 minutes. When the back is old, you need 7 minutes or 10 minutes or 15 minutes, and you don’t know the difference, so you always have to design your process for the worst case. Where normally they could get away with a 5-minute process, originally, they may have to do a 15-minute process, and they always have to do a 15-minute process because they don’t know the composition in the bath. We don’t need fancy math to do that. You can do all of this with a spreadsheet. You can get into CSTR models and very fancy mathematics around all of this, but all of this can always be modeled in a spreadsheet. A little bit in, a little bit out. How much came out?

The other thing to think about, and a concept I encourage you to think about when you’re doing industrial processes, wet processes, is there are two kinds of processes in the universe. There’s a subtractive process and an additive process. In a subtractive process, you’re taking something from a part. In passivation, you’re taking iron from the surface so it doesn’t oxidize when you place these parts in the ground. Or if you’re washing a car, you’re washing the dirt off of the car. That’s a subtractive process. When you’re painting a car, you’re putting something on the part.

There’s two fundamental differences and your technical goals are different for an additive versus a subtractive process. In an additive process, your goal is zero waste. It is achievable. It is almost always achievable to have zero waste in an additive process if you think about it. If you had no overspray, you’d have no waste. In a subtractive process, the goal is a little different. You want to take off only the thing you’re taking off and no more. So if you have a rinse process, yes, you’ve successfully rinsed off the hydrofluoric acid, but now you have rinse water, and we have a strategy for dealing with some of these things. So philosophically, elimination of variability, additive versus subtractive process, and those are two takeaways I’d encourage you to think about, things you can use in your career, or things you can take back to your manufacturing facility, or help all the way down to washing your dishes.

As Josh indicated, we started with these two basically big pots of water, and the pots of water, when you take a dish out of a dish bucket, the dish is wet, correct? Now, the difference between the dry dish and the wet dish is what we call drag-out. The mass of material taken from one bath to the next is what we call drag-out. That’s, mathematically, a very important function for industrial processes. Semiconductors, PET, washing cars, all of those things have drag-out.

The difference between these baths, in this case, we’ve separated the two baths into what we call a counter flow system. In a counter flow system, we are moving our parts from dirtiest to cleaner, to cleaner, to cleaner, finally into the wash table. But the good news here is we use the same water in every case. So we put a certain amount of water here; we reuse that water here, and here, and here. So in this case, we’ve created a counter flow system using the exact same water that is needed in every process tank. The advantage of that, obviously, is you’re using far less water.

You can mathematically model this process. We’re not going to do that today, but the relationship between the number of rinse tanks and the counter flow system is the relationship between the amount of drag-out divided by the amount of rinse water, and that’s referred to as the rinse ratio. That’s a little out of scope, but as long as you have a little extra time. The challenge there is making sure that you get the water used efficiently, and then arrangement of a counter flow system.

We’ve reduced our cycle time, by the way, from 8 minutes or 10 minutes down to 21 in the total cycle. What we’ve done on the rinse tank was we used you’re probably familiar with ion exchange or water softeners, the water softener system. We’re simply taking water from the rinse tank and we’re running it through cation and anion exchange systems, and using that water as process water. This can be regenerated with hydrofluoric acid and sodium hydroxide and you create sodium chloride, saltwater. So it’s a very straightforward process once you figure out you have the tools at your disposal.

In this case, again, the subtractive process, you only want to take out the dirt. You don’t want to take out all the good stuff. When you throw away that 6 gallons a week of hydrofluoric acid, we want that acid back. We take out the dirt by doing what we call acid distillation. Nitric acid is volatile, so we take that out, and we put the nitric acid back in the process. We have to add back the hydrofluoric. The hydrofluoric acid, we haven’t figured out how to recover that yet. Next. We need the atomic tweezer for that. But the nitric acid is recovered, and goes back into the process baths. Again, with a subtractive process, the goal is to remove only the thing you want to take off, and in that case, it’s the metal oxides.

Now, somewhere over the rainbow, when we get this all said and done, we want to completely automate the entire process. So here, we’ve incorporated passivation, 5 minutes per. Instead of 15 minutes in one tank, we’re going to put 5 minutes in three tanks. The advantage of that is now you can automate the system so that each transfer is 5 minutes apart. We don’t care how long the rinse is [inaudible 0:15:33]. The rinse is not time-dependent, but what we do care about is the amount of water used in the system. Now again, you’ve got a certain amount of drag-out coming in each part divided by the number of counter flow systems there, and again, that number is geometric. So if you take half of the pollutant off in one, and half of the pollution off in the next, you’ve reduced the pollutant by 75%, then an eighth, then a sixteenth, and it goes on forever. So just by adding process tanks and adding number of water, you can have as clean apart as you want.

And you could always define your part cleanliness in advance prior to getting to your flow designs. So if you decided I want apart with 10 ppm total dissolved salt, you can define that. All you need to do is then define the amount of drag-out you have, and then figure out how many rinse tanks you need to achieve that. In almost every case, and I’ve been doing this for a long time, almost every case four rinses is about as far as you need to go. Beyond that, your water is so clean it’s almost as clean as the incoming water. That’s been my experience. Five has been used a few times in my career, so four counter flow rinses. With the passivation system, now instead of having a single source and adding chemicals to every bath, we’re going to use the same strategy, and the mathematics are a little more complicated, but using acid distillation and then counter flowing the acid back out. Taking a little tiny bit of acid out all the time, and putting that acid back in the last system, where it’s the cleanest, maintains the baths in a system that you’re also going to see the best technical results.

The cool thing about this is now we’ve gone from a process where we’re doing 21 minutes of load, and then the operator’s running around and hurting himself and bending over and doing all the things that we don’t want them to do, the dirty, dumb, and dangerous, the three Ds, now they can simply push a button and, every five minutes, they get a new clean bath. That’s the of the end goal. By the way, 5 minutes and 30 seconds is a vast increase in capacity with [inaudible 0:17:42]. We’ll be up to that in the next nine months or two years or however long it takes, but that’s the goal that we’re trying to accomplish.

Net result now is 50% reduction of waste and 50,000 in savings. Eventually, we’ll have eliminated hazardous waste completely, we’ll have recovered all of the acid, we’ll have a 200% increase in capacity, and the dollar figure is astounding. And again, this grossly understates the real value when you start thinking about the labor costs and improved quality and quality of life for the operators, so it’s a very cool outcome.

Male Speaker: We do have one minute for questions, then we’ll stay. So if you have more questions, you can have more questions a bit later. Any questions for now?

Male Speaker: How will the process do change from hazardous waste to wastewater because it’s still the same acid going out into the wastewater? is there due to a regulation that the amount of acid going to the wastewater?

Michael: In the United States, there’s a rule called permit by rule. So it’s anything where the wastewater is hard plumbed to a permitted industrial wastewater treatment facility is not designated hazardous waste. It’s a regulatory nuance that we’re taking advantage of now, and we’ll be recovering that material once we have acid distillation. A temporary strategy.

Copyright © BPI. All rights reserved. | Newsphere by AF themes.