Loading ...
Sorry, an error occurred while loading the content.

48760Re: ⚗ Distillation Controller

Expand Messages
  • tyto_negro
    Jan 12, 2013

      Excuses accepted ;-)

      I'm not contradicting you theoretical explanation of things, but:

      1) a traditional alambique/pot still is a far-from-perfect thing. And that is good, we want a tasteful drink not technical C2H5OH in our glass. (People interested in that would go for more sophisticated reflux stills)

      2) We have non perfect wine, that is, it is *not* a simple binary mixture. I'm making mainly natural fruit brandies.

      So, that makes a far from perfect distillation process. If the process would adhere so well to theoretical rules, can you explain the following:

      a) Why do I get my first smelly drips sometimes starting at 65 C?

      b) Why don't we turn up the heat to the max to get everything out a.s.a.p. if boiling points and vapour temps don't change/react to power settings?

      c) Why does the ABV change much more than predicted by the ABV of the wash and the connected vapour temp alone?

      You don't have to answer these questions, they are merely rhetorical.

      I have studied chemical engineering/process technology at university levels and I'm an electronics engineer.


      Like I said, results sofar a very good and when the program is finished it will surely help me using my time more efficiently without doing prejudice to the end results. The program will mimic what I do normally by hand, like you said, adjust power levels.

      To Paul:
      The influence of air pressure and ambient temperature will only induce a shift in time, the raising temp will, literally, sooner or later, pass the point where I get the desired output.

      Maybe the (stripping) process won't be faster when I do it 'automated', but since I don't need to sit next to it, I still have more time for other things, even when the distilling takes longer!

      I thought others here would be interested in something like this as well ;-)



      --- In Distillers@yahoogroups.com, "tgfoitwoods" wrote:
      > Tom,
      > I'm sorry I missed the location of that temperature sensor, and I did
      > look for it. The truth however, is that the new location of the sensor
      > does not change my argument at all in theory, and in practice the
      > difference is small and difficult to predict.
      > If I may be permitted a small simplification of the problem, I can
      > produce data that will help us understand what happens in a potstill. I
      > offer the distiller's "cigar" curve, which tells us everything we need
      > to know about distilling a mixture of ONLY ethanol and water, a simple
      > binary mixture. The reason for this simplification is if we showed all
      > the mixtures of all the volatile liquids found in the typical still
      > wash, the graph would have to be in too many dimensions to understand,
      > certainly to view. Here's the curve.
      > http://www.kelleybarts.com/PhotoXfer/alcoholvaporCelsius.gif
      > Ok, what the blue curve represents is the boiling point (up the left
      > side) versus the percentage of ethanol in the wash (across the bottom).
      > If we look at our theoretical 10% ABV wash, we see that the 10% line on
      > the graph crosses the vapor boiling boint (blue) curve at close to 93C,
      > and here's where we disagree. When that 10% wash boils at 93C, it forms
      > vapor at EXACTLY THE SAME TEMPERATURE, 93C. Now, here's where I'll admit
      > theory can vary from reality: If you have a tall copper riser on your
      > still, and it's un-insulated, and the vapor is moving slowly up it, and
      > the garage door is open for ventilation, and it's windy, and it's 0C
      > outside, that vapor will cool a bit, but it's exactly 93C when it leaves
      > the surface of the wash. and it's really close to 93C when it passes
      > your temperature sensor and heads down the lyne arm to the condenser.
      > Further, if we follow that 93C temperature line to the right, until it
      > crosses the red (vapor) line, we find 93C crosses the red line at the
      > same place the 53% line crosses it, so we know that the 93C vapor,
      > formed by the 93C-boiling wash, when condensed would form a 53% ABV
      > liquid. Notice that there is nothing on that graph to compensate for
      > wash that is barely boiling or hugely vigorously boiling. If it's
      > boiling at all, the vapor going by your temperature sensor is 93C, no
      > matter how you apply the power (assuming my garage door is shut on a
      > warm day, but I'd hate to have to write an algorithm that corrects for
      > that).
      > The only thing that will change the vapor temperature is to change the
      > %ABV of the boiling wash, and that will occur because as the ethanol
      > boils off faster than water, the %ABV of the boiling wash increases,
      > becomes closer to the boiling point of pure water, and that happens very
      > slowly over time.
      > For that reason, many of us use a thermometer in the same position you
      > use, to read the head temperature, assume it's pretty close to the wash
      > boiling boint, and therefrom to infer what our instantaneous distillate
      > ABV is.
      > Re-reading your last post, do I see the possibility that your system
      > design is just to keep a constant flow as the heat of vaporization of
      > the distillate goes up due to higher water concentration? If that's what
      > all this has been about, then I apologize, but you need to know that
      > most potstillers will manually change power input a few times during a
      > run, for reasons of flavor, and not constant output.
      > Zymurgy Bob, a simple potstiller Making Fine Spirits
      > --- In Distillers@yahoogroups.com, "tyto_negro" wrote:
      > >
      > > Dear all,
      > >
      > > As far as I'm concerned, Bob, "... this wash-temperature-control thing
      > won't get a foothold on these lists..."
      > >
      > > Maybe read first and answer then.
      > >
      > > The temperature of the vapour gives a good indication of the
      > alcohol/water proportion of the distillate.
      > >
      > > Sofar results are very good, apart from a few bugs in the software.
      > Of course the easiest is distilling with constant power input, but
      > experience shows you need an increase of power towards the end.
      > >
      > > Having the temperature increase slowly (in time) corresponds with this
      > power rise (in time) and/or vice-versa.
      > >
      > >
      > >
      > >
      > >
      > ----snip----
    • Show all 16 messages in this topic