That’s not beer! A biofuel fermentation controller project

biofuel-fermentation-controller

Any home brewer will recognize the setup pictured above as a temperature controlled fermentation chamber. They wouldn’t be wrong either. But you’re not going to drink what results. This project is aimed at providing a temperature controlled environment for fermenting biofuel.

[Benjamin Havey] and [Michael Abed] built the controller as their final project in his microprocessor class. The idea is to monitor and control the mini-refrigerator so that the strain of Saccharomyces Cerevisiae yeast produce as much ethanol as possible. An MSP430 microcontroller was used. It monitors a thermister with its analog to digital converter and drives a solid state relay to switch mains power to the fridge. At 41 degrees Fahrenheit this is down below what most lager yeasts want (which is usually in the low fifties). But the nice thing about using a microcontroller is you can set a schedule with different stages if you find a program that gives the yeast the best environment but requires more than one temperature level.

Who knew all that beer making was getting you ready to produce alternative fuels?

Comments

  1. Hirudinea says:

  2. Zack says:

    A Lager yeast seems like a strange choice – most have lower alcohol tolerances than Ale yeast, wine and cider yeasts higher still, but what would seem to be the obvious choice would be a Turbo/Distillers Yeast.

    • Ben says:

      We tested it with baker’s yeast for proof of concept, which is optimal at ~50 Fahrenheit (41 was a mistake on our part). Many of the dry Turbo/Distillers yeasdt are in this same family of yeast and would still be best used in the fermentation box.

  3. mike hogan says:

    Why use a terminator? One of.my favorite features.of.the MSP430 is its.built in temperature sensor. In fact, the thermistor costs nearly as much as the MCU ..

    • Ben says:

      Unfortunately I found that the built in temp sensor on the MSP430 is not good for environmental testing. First off, it’s built *inside* the chip, so it’s really only intended to measure the chip temperature. You could use a function to convert it to the equivalent environment temperature, but chances are the equation would cascade on itself, as the internal temperature will increase because of the calculations. You’d have to figure out the temperature response of the chip depending on what instructions were being used and then formulate an appropriate function.

  4. Leithoa says:

    I’d be curious to know the exact strain of yeast they’re using. Every strain of yeast I’ve read about doesn’t “thrive” at 41F. Most beer and wine yeasts convert sugar to ethano just fine up into the low 80’s. sure you get more fusels but since you have to add them back for tax purposes why bother with temperature control? Even distillers yeast that ferments up to 18% in 5 days( 20% in 3 weeks) like temperatures above 60F . This yeast marketed as a fuel ethanol yeast works best from 86-95(http://www.eagleyeast.com/files/C6_Data_sheet_blue.pdf).

    The difference in the amounts of fusel oils produced is below a couple percent even when the yeast are stressed. Since you’re distilling it anyways why bother with trying to minimize fusel oil production? It doesn’t really cut into the profit margin since you don’t have to discard the hearts and tails for fuel ethanol.

    • Ben says:

      The 41 degrees Fahrenheit was actually a mistake. I misread the dormancy temp as the optimal temp. In actuality it should have been ~50° F, although it still fermented. The reasoning for our yeast choice was attenuation rates. S. Cerevisiae has a very high conversion ratio of sugar:alcohol, making it ideal for efficiency, but not necessarily speed/cost. Further testing would be required to find the perfect strain, one that was efficient enough but was fast enough to offset the energy required to run the fridge.

      • Leithoa says:

        S. Cervisiae isn’t a strain it’s a species of yeast. There are dozens of commonly used strains of S. Cervisiae in brewing: K1V-116, EC-1118, WLP 001, Montrachet, WY1028. These are strans, lager and ale yeast doesn’t narrow it down either as dozens of each are commonly used as well.
        When you call up the yeast supplier what’s the name you used? I guess in short what I am getting at is, I wanted to look up the specs( attenuation, temp range, gravity range) of the yeast you used.
        Often the ‘optimum’ temperature listed on the datasheet/sachet is the temperature at which few fusels and other cogeners are produced. Since you aren’t concerned with how the wash tastes and even when horribly abused yeast only produce a few percent higher alcohols it doesn’t matter how high the temperature gets so long as it doesn’t kill the yeast off. This is for fuel use so low temperatures should be avoided so production can be maximized. Quick production is even more key due to the small scale involved.

  5. Chris C. says:

    That’s not biofuel, either. Just an unusable mix of ethanol, water, and other stuff. Fermentation is really the easy part. Purifying the ethanol, without expending more energy than the finished product yields, is the hard part.

    Not that this takes away from the project, in its more limited scope. The website seems to be overloaded, but I’m sure it’s a good write-up, based on the intelligent comments from the creators I see here. And I hope it received an excellent grade.

  6. dan says:

    I’ve written a long comment with the maths to show that this system is uniquely crap at being green.

    perhaps my comment is too long, but it doesn’t seem to want to post.

    When I was at uni, I watched a guy fail his coursework, his idea was to build a new valve for fire fighting equipment. that would regulate water pressure.
    clever idea, he had loads of stuff done mathcad simulations of projects etc.

    long story short, he failed his project, not because of the work he did, but because he did not identify a real life problem, and his solution was completely inappropriate. it was the marking lecturers view that when there is a fire you want as much water as possible on the flames.

    anyway. do the maths.
    sub-obtimum brew maybe between 4 – 6percent (ABV)
    uber climate control brew may be 7 – 9percent (ABV)
    in your 4.5l demijohn that’s only an additional 135ml of ethanol.
    which is 3MJ additional energy compared to just leaving it on the side.

    running your fridge for two weeks, (assuming it’s pretty new and energy efficient will take a little over 3000MJ of energy, (averaged for time).
    your 4.5 litres of fluid only contains 8MJ of energy in the ethanol, that’s at 8% since you’re not using a turbo yeast trying to brew stronger is only going to result in poisoned and dead yeast anyway.

    in short your system is net energy negative.
    sure you produce an extra 3MJ, but given that you use over 3000 to get there it’s a bit of a waste.

    I would expect a similar mediocre mark to the guy who decided that limiting pressure and water flow on a fire hose would be a good idea.
    the problem is that (at least from your write up there) you haven’t identified the key problem, your system is net energy negative, and therefore not really a green hack at all!
    (if you want the maths then just say so and I’ll try and post in multiple comments or something).

    • Leithoa says:

      Most beer yeasts don’t tucker out until 10-11% ethanol, with the low fermentation temperature this is almost assuredly reached, it will just take forever. Many beer yeasts along with nearly every wine yeast are good for up to 15% EtOH. Champagne, distillers yeast, and a few wine strains make it up to 20% withouth much trouble. These numbers don’t affect the overall calculation but they do make a difference.

      Your observation of energy debt is accurate, but the fact that the US has 10% ethanol added to all gasoline, a few cars on the road are entirely ethanol, and the entire country of Brazil uses ethanol for fuel, state that the world doesn’t care.

      This is a microcontroller class, so ‘the problem’ is a client wants a climate controlled fermentation chamber for the production of ethanol requiring a temperature window from 41-100F with the ability to program temperature ramps in for optimum production. BP, Exxon-Mobile, Conoco-Phillips, all have R&D teams engineering yeast and other bacteria to produce ethanol from a variety of feedstock. The temperature controller doesn’t care what is in the chamber, only that the temperature of the chamber and the preset match.

      • dan says:

        true…
        the point I was making it it requires 3395MJ of energy to power the fridge for two weeks.

        if they could brew 4.5l at 100% ABV (pure alcohol)
        at 22MJ/L that’s still only 99MJ

        even if we pretend that as they say the temp should be 10Degrees the fridge uses 1/4 of the energy, and that they can fit 4 demijohns into the fridge that’s still 1000MJ of energy utilised to produce 400MJ of energy.

        This might well be a micro-controller class.
        the guy I mentioned who failed was just studying Mechanics.
        poor problems = poor conclusions = low marks.

        as I said, just like the idea that you should limit the output of a fire-hose is a poor premise for a final project, so is the idea that you should use four times as much energy as a fuel provides, (before distillation) to produce a green fuel is ludicrous.
        designing a mechanical valve = great. making up that it’s got a good roll in an application it’s completely unsuited for, a roll which in fact makes fire fighting equipment less effective = dubious work and lower marks.

        calculations prove that as a green fuel this is completely ineffective.

        best advice to these guys.
        Forget about the green fuel. maths proves that anything you make won’t be green fuel.
        write that you’re students you want cheap free strong beer. then the apparatus designed and the conclusions are perfectly suited to the premise. = good work = good marks.

        • Leithoa says:

          These students aren’t solving the “green fuel” problem. They’re solving the “build a programable thermostat” problem. Which arguabley isn’t novel, but perhaps their approach was( I don’t know much about MCUs and programming).
          And again while not energetically favorable, biofuels are economically favorable due to a variety of subsidies or the presence of cheap feedstock ie; Brazilian bagasse and cane juice.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 96,660 other followers