Episode 1: Why Moisture Balances Suck
Zachary Cartwright (00:00):
This is Zachary Cartwright. I'm the lead food scientist at METER Group and you're about to hear an episode of my brand new podcast called water and food.
Speaker 2 (00:08):
What moisture content has developed into is kind of a Schlep Blindness and it means a tedious or unpleasant task. They have product that needs to ship, and product can't ship if the readings aren't done, and so these quality departments are under tremendous pressure. Labs are not checking to see if their instruments are actually functioning properly. You know, so part of this Schlep Blindness thing really got us thinking: how could we address this?
Water has been called the luck of the planet by Daniel Boorstin and its impact and significance is evident everywhere in the foods that we eat. Every year, billions of dollars are spent by food manufacturers to move water in and out of food products. As a food scientist, I'm on a mission to understand how this can be done better. Today's guest is my colleague, Scott Campbell, who is the CEO of METER group METER, which was formerly Decagon Devices, has been engineering sensors to determine water content, water activity and water potential for almost 40 years. And its products are used all over the world from universities to vineyards and even on Mars.
Zachary Cartwright (01:10):
Scott is going to talk about something that is known to get him real riled up. And that topic is Why Moisture Balances Suck. So welcome to my first podcast, Scott. How are you today? Doing great, thanks Zach. Yeah, you're welcome. How are you holding up with everything going on?
Scott Campbell (01:27):
Well, it's been an interesting time to say the least. All of us at METER, all of us that can, are working from home and it's really taught me a lot about productivity and the fulfillment that I get out of work. Some of this stuff has worked well and some, some hasn't. But for the most part, being a test and measure company we've been fortunate that we're involved in an industry where people still need food and still need to produce food and supporting the businesses that do that. So we've certainly learned a lot since this has started and so far it's going okay for us.
Zachary Cartwright (02:08):
Yeah, someone told me they saw you at the office this week and your hair is getting long and you're sporting kind of a 2009 Justin Bieber look. But that's OK. Someone told me similarly that I am starting to look like Taylor Lautner going through his werewolf stage from Twilight. So
Scott Campbell (02:27):
To be clear, I would definitely take Taylor Lautner over Justin Bieber. So I think you came out on top on that one. But yes. It's funny that you mentioned it because I just asked my wife, I said, you need to shave all of my hair off, just go complete buzz cut. And she wouldn't do it. So, yeah, I guess I'm just about Justin Bieber here.
Zachary Cartwright (02:52):
Well, here we go. Justin Bieber and Taylor Lautner. Today we'll be talking about why moisture balances suck. So, to start off with Scott, can you just describe the current landscape of moisture content balances in food manufacturing? And a little bit about why they're used and what challenges they are helping to address?
Scott Campbell (03:13):
Sure. And you know, as I do this, I want to keep it pretty informal and just say to start with, we know that when we talk about this stuff that we at METER, that we are addressing it from our own perspective and with our own biases and we have a particular view on it and we understand that our view relates to selling our products. We don't make apologies for saying that we think that they work well. But certainly we have a particular opinion on this and that's what we're going to be talking about today. The reason why moisture balances are used at food companies, in my opinion, it comes down to a couple of things. As the food industry began to grow and to become a prominent producer of growth and output in the economy.
Scott Campbell (04:07):
And so mostly I'm talking here about 20th century United States. Then, the quality control groups, and initially were called quality control groups, got up and going because they saw they were making a lot of the same product and they wanted to make the same product every single time so that when somebody saw a package on a shelf and it had a particular name on it, they know that it would be the same as the last time that they bought it. So the existence of these quality control groups makes perfect sense. I'm going to start saying, well, what could we measure about a product? Moisture content just makes sense. And that's really the first reason why these approaches for measuring moisture content. They make sense. If you take a product and you dry it out and you know how much it used to weigh and how much it weighs, now you can calculate how much water was evaporated off of the sample.
Scott Campbell (04:59):
And it's just something that's like the moisture content of the sample is 23% that makes sense to people. And then once it makes sense to people, I can start talking to each other in terms of, Hey, this batch has 25% it's a little high, especially 21% it's a little bit low. And it just became a currency, a language that people used at food processors to talk about their product. The other reason why it was so prevalent and you know, why they're using the food industry is that balances exists. The capability to measure mass accurately has been around for a long time. You know, those standards exist around, you know, pounds or kilograms or whatever it might be. And so they're helping customers to address the basic challenge of what is going on with our product, specifically what is happening with our product.
Scott Campbell (05:50):
Something we can observe with our eyes that might affect how a customer sees it, how experience it, what it tastes like, what its texture is. And in that sense it is totally logical that moisture content would be used for that purpose. So principally they help people address the challenge of, number one is what I'm making today, the same as I made it last time? And number two, the problem would be profitability. So how do I make sure that my product has the right amount of moisture in it? too much? And you have food safety problems. Too little? And your yields go down. And this is something that has been known in the food industry for years.
Zachary Cartwright (06:31):
So it sounds like some of the challenges that these are helping to address are consistency and then understanding how moisture or how much your content relates to profitability. But what are some of the problems that you see with these instruments? Are they introducing any types of problems for food manufacturers?
Scott Campbell (06:50):
Yeah, I mean that the issue really is around this concept of what's called Schlep Blindness in case. So this was a term that was coined by Paul Graham a long time ago. I mean, well it was like 2012 when he started saying it. So schlep is this Yiddish word that we use in general, you know in the general vernacular in the U S, and it means a tedious or unpleasant task. And so what most your content has developed into is kind of a Schlep Blindness. So it's a difficult measurement to make. You have to do sample prep, you have to do sample prep the same way every time. When you put the product on the balance, it dries water off, but it does it at a different rate for each product and the temperature which you try to all, you always give you a different answer in terms of moisture content.
Scott Campbell (07:45):
So for example, if have some brownies and you put them into the oven, you bake them, you take them out and you say you want a moisture. Well when you put the brownies in, if you dry it to a 100 degrees C over a certain time period, you're going to get a certain moisture content. If you drive to 120, or 140, 180, to 200, each of those levels is going to give you a different moisture content. There will always be more water and other things getting dried off of your sample as you go up in temperature up until the point where the entire sample is incinerated. So that's the first thing is it's difficult to get consistent readings because the method, the approach that you use will give you different answers. So getting back to that consistency point that you mentioned, the method that you use is going to give you different answers.
Scott Campbell (08:34):
If the lab humidity is different on a particular day, even if you're using the exact same temperature in time, you'll get a different answer because the vapor pressure is different. If you do your sample prep different, so the surface area of your sample changes, you'll get a different answer. And most products that our customers deal with are intermediate to low moisture products. That means that the amount of water that comes off of the sample is relatively low compared to the mass that's left on the balance. And so when that difference between the dry weight and the wet weight is low, that means that even with a really precise balance, you're going to not have a high signal to noise ratio. So getting back to the reason why people want it to measure moisture in the first place, and that is consistency. The method itself of using a balance is not good for establishing consistency.
Scott Campbell (09:21):
And that's why the Schlep Blindness idea comes in. It's this tedious or unpleasant task. I mean it's tedious and unpleasant just to get the readings from a moisture analyzer because you have to do all the sample prep and set up the methods and calibrate them to reference. But just the idea that that the Schlep Blindness relates to trying to use those numbers to then say, “my process is consistent.” If you're doing that, but you don't see that there's a correlation between your process variables. And the moisture content that you're reading, it's probably because of the lack of precision of the measurement method. And that's why we say there's issues with the balance approach.
Who is coming up with these procedures, you know, who decides the temperatures and the length of time. Is this something that is done in house and is this going to vary from company to company?
Scott Campbell (10:10):
It really does and what we see is that different companies do it in different ways. We see large multinationals, you know, and everybody knows these companies, you know the Nestle's, the General Mills, they, a lot of times we'll have standards departments that will say this is how we do this test method at the company and you know, these are the instruments that we use. And trying to have some level of consistency around what's being done. Because even if there is some noise in the measurement, at least if the methods are the same, they can have some comparable results between factories and over the corporation. So those are common at larger companies. At smaller companies, it's much more common for people to purchase a moisture balance off of the internet or through a catalog supplier to take the moisture out, set it on the desk and just start pushing buttons and getting readings.
Scott Campbell (11:02):
And this is particularly damaging because if you don't do any reference methods for moisture, then the reading that you're getting is even less precise. So, so it's kind of a, something people don't talk about a lot. They just take the moisture content meter, they just arbitrarily pick a temperature number and then start taking the readings and just say, Hey, we're going to do one 60 for until the mass starts stops changing at a particular rate. And that's just what we're going to do. But if you read the user manuals on these moisture balances, they say, clearly, for every product you have to run a reference moisture, which is four hours at 105 degrees C, you have to use a desecrator, you have to use a precision balance. Once you have that number, you have to devise a temperature regime that gives you that same answer for every product that you have. But that, as you know, that is a Schlep right there. It's getting tedious to run that for all the products. And most companies that use moisture balances are not doing this.
Zachary Cartwright (12:06):
Yeah. Even as a graduate student, admittedly we had a moisture balance and we did just that. We took it out of the box and started using it right away. But as I've moved along in my career, I've realized that it can take weeks of research to really calibrate that instrument correctly. And, this isn't something that I see most companies doing. Is that something that you would agree with?
Scott Campbell (12:29):
Yeah, exactly. I do see that happening a lot and it's understandable, you know, we're not sitting here saying, Hey, you know, these companies are bad for doing this. Everybody is overworked. They have tons of stuff to do. They have product that needs to ship, and product can't ship if the readings aren't done. And so these quality departments are under tremendous pressure with low staff and without a huge instrument budget to be able to meet those needs. So, it is something that we see a lot and I definitely agree with you on that.
Zachary Cartwright (13:03):
So how long have these instruments been used in the food industry? Do you know?
Scott Campbell (13:11):
Yeah I tried to research this and the best information that I can find was that in the 50s or 60s, some of these balances started to be available. I mean, part of it has to do with the availability of a microchip to be able to do basic firmware on board that would say, hey, here was the starting weight and the ending weight. So, as best I can tell that the practice was used for doing moisture contents based on mass ever since we've had balances. But these machines that would do a simple calculation, which is dry weight and a wet weight and the difference over the total mass, you know, 60s, is the most that I could find on that. But as I did this research, I also saw something really pretty interesting, which is the stuff that we geek out about, which is water activity. If you go and look up what moisture content are, most of your content is found on Wikipedia. And you'll find a little page that talks about it.
Scott Campbell (14:15):
But it's just, you know, there really isn't a lot there. It's because the science of moisture content is not that interesting even though it's used throughout the food industry as the way that people talk about about product quality. There isn't as much science behind it. And in fact, the water activity page on Wikipedia has way more information about how it relates to microbial growth, how to measure it, like all of these, these technical things. So as I was researching the history, that was another interesting aspect that I ran into.
Zachary Cartwright (14:47):
And it sounds like there's just so much variation with how this is being done and,
Scott Campbell (14:52):
And it's hard to know the accuracy that you're getting. So what other limitations do you see with moisture balances and why isn't the food manufacturing industry more aware of these limitations?
Scott Campbell: Yeah. One of the limitations of the moisture balance is that it doesn't have a good tech stack behind it to help people with their workflows. Meaning that a lot of things that people have to do that involve the moisture balance, like writing down readings, verifying that the balance is calibrated properly, those things are all just put on to the user. And partly it's because moisture balances are used in way more industries than just food and food quality. And so the companies that make the balances don't really have any incentive to make the workflow easier. And that's a key thing is looking at the, you know, I talked about this Schlep Blindness.
Scott Campbell (15:46):
Part of the reason that we have invested in trying to make these things easier for customers is because our customers are principally only in the food industry. And so it allows us to look at the workflow and say, Hey, how can we make this whole process easier for people? Rather than just saying, how could we make a better measurement. We'd like to make a better measurement and make it easier for folks. So I think that's one of the limitations of moisture balances is that they're meant as general purpose devices, not engineered to be appropriate for food labs. And I suppose another limitation of the moisture balance, and I mentioned standards before, is if you'd like to make sure that the balance is working properly, there are these standards for moisture do exist. It's interesting because once we started saying, I don't know, 10 or 15 years ago, there aren't any standards for moisture content that they actually do exist now.
Scott Campbell (16:40):
You can get them in these little packets and measure them, but they're not widely used. And in most cases, labs are not checking to see if their instruments are actually functioning properly. There could be drift on the balances, there could be errors in the way that the firmware is working and the calculations are done. They're simply not checked. So that's another limitation of the balance.
Zachary Cartwright: So after looking at these limitations, is there any type of competitive or compelling alternative measurement that food manufacturers could be using instead?
Scott Campbell: Well, I mean, now, Zack, you're asking me a leading question and I'm gonna have to say that, part of this Schlep Blindness thing, it really got us thinking how could we address this in the food industry. In fact, it's something that I've been obsessed with for the last, oh, at least 12 or 13 years.
Scott Campbell (17:39):
Even before… METER is a family company. It was started by my dad, so even before I came back to work at METER full time. I was working at other companies. This was in like 2008 or so, 2007 and I'd always seen the way that people did moisture in the food industry and I thought it could be done better. And I said to the people who are at the company at the time, I was like, why don't we look at a way, because we know that moisture content and water activity are correlated, why don't we look at a way that we could get a water activity reading and show customers the moisture content for that product on the same screen? That's actually what produced a product called Aquila Duo that we've been selling since about 2009. So I came to the company in 2008 and we actually commercialized that idea. And it was great because you know you could take this water activity reading and get a moisture content at the same time.
Scott Campbell (18:36):
So just a quick side note on that, that is the approach that we're advocating for people to take just hit us like a brick. It was: moisture content is important. Moisture content should be measured. But the way to measure moisture content isn't with a moisture analyzer, it's with a water activity meter. And the reason for that is that the correlation between water activity and moisture content is known. So whenever we measure a water activity for a specific product, if we know what the product is, we can also calculate what that moisture content is. And the key there is that for intermediate to low moisture foods, a small change in moisture is correlated with a big change in water activity. And that's the perfect way to measure this parameter is using a way that's super sensitive to it. So because of what our activity is so much more sensitive to changes in moisture and on a product specific basis, we can give you the moisture content number.
Scott Campbell (19:32):
That's why we're so excited about it as a method. Now, in the old implementation of this moisture content, which through this duo product that I was telling you about, we would just make the models and then we had to email them to people and then they had to download them over the RS 232 cable to their Aqualab 4 and then, and then they had to offset them and then they could use them. And for companies that made two or three products, this was a good, a pretty good system. However, for companies that made 30, 40, 50 different products, it was a complete non-starter. It just became unmanageable because the user interface was bad. And that's why we're so excited about the SKALA platform that supports our water activity meter now is that, all those capabilities, we put them on the cloud. And what's more, we found a way on the cloud to model how products equilibrate on a product specific basis, how they equilibrate.
Scott Campbell (20:32):
And so we sped up the water activity reading from five minutes to one minute. So we made it five times faster. We combined the capability to do moisture content on a product specific basis, and getting that product that you want to measure on your machine and reading is as simple as choosing in a dropdown box on an iPad. And so that's why I was saying we tried to address the workflow issues. So essentially in one minute you can get a precise water activity and moisture content reading from the same machine and do it in a way that doesn't require the same amount of sample prep. It's nondestructive to your samples. And it's super, super precise. So, we're really excited about that. We see a tremendous potential for making people's lives easier and finally getting moisture content to where it needs to be. It is something people need to know about. You should be interested and excited about knowing the moisture content of your products, but the way that you should be doing that is different than the way that most people think. And that's what we're trying to communicate. And I also want to make it perfectly clear that I will be getting a haircut soon.
Speaker 2 (21:44):
I don't know. I kind of like the direction I'm going, so I might keep growing it out for a while. I don't think I'm ready for the buzz cut like you.
But, Zach, you got called Taylor Lautner. I mean, let's be serious.
That's true. That's true. Well, Scott, I want to thank you for being on my very first podcast. This is really great. I'm excited to start with you and to see where this podcast goes. So again, this is the podcast of Water In Food. I'll be interviewing a range of execs and scientists and engineers along the way. So if you're out there listening, I hope that you join in on the next conversation. So thanks again, Scott.
Yup. Thank you, Zach.