You should perhaps investigate Proview, which is a open-source industrial process control system. It would make the discrete PID-controllers redundant and would allow you to hookup all sensors etc into a unified system. I have been thinking of doing the same thing.
I got a bit obsessed with pump performance as it seemed such a limitation but when I started using closed loop feedback control it seemed to nail the problem. I guess fish tank pumps would be fine if their speed can be controlled electronically.
In your experience, how is the system affected by changing the diameter of the rectifying column? If one was to use a 38mm OD copper tube, would the stripping column need to increase by a similar proportion EG 19mm?
I've tried the same 15mm stripping column with a 35mm rectifying column and it works fine. When I was testing the stripping column for optimal length and angle, I found the critical thing was how much down flowing liquid from the rectifying column it could tolerate before the upcoming vapor drew the down coming liquid with it, making it flood, but that only seemed to happen with power settings of several hundred watts. So my guess is that the stripping column diameter would not have to be increased until the rectifying column was well over 50mm.
Don't fret the mess, @@tight1449 ! I really appreciate all the great intel you have to share -but as a visual learner, would love to see more of the apparatus
I'll show it in more detail. I have just made a new rectifying column from scratch. The old one was so modified and remodified it was not a great model to copy. I'm currently testing it and will post videos on making it if it works OK. The stripping column is the same and is very simple.
That's a very good question with a complex answer. Volatile components of the incoming wash tend to remain as vapour, and preferentially leave the column via the head's vapour fraction, and not the hearts liquid fraction. However in the case of methanol, although it is more volatile than ethanol, it has a significant affinity for water so while it does partition between heads and hearts, there is still methanol present in the hearts, although at very small concentrations, particularly in my case where I'm using cane sugar derived wash which has virtually no methanol in it to begin with. Methanol distillation is a complex subject and I've found it's very poorly understood by home distillers. For example, the idea that for-shots contain more methanol than do hearts, which home distillers will swear to, is false.
Silly question for silly questions sake... what if your wash was in say a keg at a minimal pressure even air pressure not co2. You then had the wash feed thru a regulator so it's fed at a constant pressure into the analyser? Cheap way of dialing in delivery or as I said silly question?
Could work but you need a precisely controllable flow rate. A regulator would give you a constant pressure. You could add a fixed resistance to get constant flow but it would probably be difficult to vary it precisely as needed. I have used a system with gravity feed through a constant resistance varying flow by changing the height of a header tank but it was more difficult to set up than roller pumps.
I had a look at the datasheets. They are oscillating AC pumps so would be controlled by varying the AC frequency. That is easy enough to set up but their flow rates at 50-60 Hz with low pressure are quite high at 70-300 ml per minute so for my thin column you'd need a frequency of under 5Hz. Not sure how well they work at that frequency.
Can you reference that? 96.51% is by volume. By mass it is about 95.6% and by mole fraction about 91%. I have not seen 97.2 at STP by any measure. 96.51 by volume accords with my experience because that is the barrier I have never gone past.
@@tight1449 Wikipedia's ethanol page? "Mixtures of ethanol and water form an azeotrope at about 89 mole-% ethanol and 11 mole-% water[78] or a mixture of 95.6% ethanol by mass (or about 97% alcohol by volume) at normal pressure, which boils at 351 K (78 °C)." Or from the rectified spirit page "The purity of rectified spirit has a practical limit of 97.2% ABV (95.6% by mass)[2" The 2 there sources the claim to Whiskey Technology Production and Marketing (great book)
@@MrJhchrist I'm not surprised there is a discrepancy. It is almost always quoted by mass as 95.63 or mole fraction. I could not find any direct measurements by volume which is what we need as hydrometers are calibrated by volume. The conversion seems to be made by calculation. I have not previously seen it as 97.2 but have as 96.988 and 96.51. My measurements are limited by hydrometer and thermometer calibration but according to my measurements it is somewhere between 96.5 and 97. I cannot prove 97.2 wrong though!
![Felix "Unfair" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/Oswujxm2Ag0/mqdefault.jpg)
You should perhaps investigate Proview, which is a open-source industrial process control system. It would make the discrete PID-controllers redundant and would allow you to hookup all sensors etc into a unified system. I have been thinking of doing the same thing.
Thank you for the suggestion, I will have a look at it. Looks like it is just what I need for the infernal feedback control loop problem.
I use a submersible fish tank pump with a needle valve.....I feel so cheep.
I got a bit obsessed with pump performance as it seemed such a limitation but when I started using closed loop feedback control it seemed to nail the problem. I guess fish tank pumps would be fine if their speed can be controlled electronically.
In your experience, how is the system affected by changing the diameter of the rectifying column? If one was to use a 38mm OD copper tube, would the stripping column need to increase by a similar proportion EG 19mm?
I've tried the same 15mm stripping column with a 35mm rectifying column and it works fine. When I was testing the stripping column for optimal length and angle, I found the critical thing was how much down flowing liquid from the rectifying column it could tolerate before the upcoming vapor drew the down coming liquid with it, making it flood, but that only seemed to happen with power settings of several hundred watts. So my guess is that the stripping column diameter would not have to be increased until the rectifying column was well over 50mm.
Interesting, would like to see it in operation.
I'll film it. But I'll have to tidy up my distillery first or it would be too embarrassing!
Don't fret the mess, @@tight1449 ! I really appreciate all the great intel you have to share -but as a visual learner, would love to see more of the apparatus
I'll show it in more detail. I have just made a new rectifying column from scratch. The old one was so modified and remodified it was not a great model to copy. I'm currently testing it and will post videos on making it if it works OK. The stripping column is the same and is very simple.
I find this fascinating but still a bit confusing - how do you get rid of the methanol from a newer wash coming in?
That's a very good question with a complex answer. Volatile components of the incoming wash tend to remain as vapour, and preferentially leave the column via the head's vapour fraction, and not the hearts liquid fraction. However in the case of methanol, although it is more volatile than ethanol, it has a significant affinity for water so while it does partition between heads and hearts, there is still methanol present in the hearts, although at very small concentrations, particularly in my case where I'm using cane sugar derived wash which has virtually no methanol in it to begin with.
Methanol distillation is a complex subject and I've found it's very poorly understood by home distillers. For example, the idea that for-shots contain more methanol than do hearts, which home distillers will swear to, is false.
Silly question for silly questions sake... what if your wash was in say a keg at a minimal pressure even air pressure not co2. You then had the wash feed thru a regulator so it's fed at a constant pressure into the analyser? Cheap way of dialing in delivery or as I said silly question?
Could work but you need a precisely controllable flow rate. A regulator would give you a constant pressure. You could add a fixed resistance to get constant flow but it would probably be difficult to vary it precisely as needed. I have used a system with gravity feed through a constant resistance varying flow by changing the height of a header tank but it was more difficult to set up than roller pumps.
Have a look at Ulka pumps.
I had a look at the datasheets. They are oscillating AC pumps so would be controlled by varying the AC frequency. That is easy enough to set up but their flow rates at 50-60 Hz with low pressure are quite high at 70-300 ml per minute so for my thin column you'd need a frequency of under 5Hz. Not sure how well they work at that frequency.
@@tight1449 My name's Peter, what's yours?
The azeotrope is not 96.51% abv, it is 97.2%. See any source that doesn't confuse abw for abv...
Can you reference that? 96.51% is by volume. By mass it is about 95.6% and by mole fraction about 91%. I have not seen 97.2 at STP by any measure. 96.51 by volume accords with my experience because that is the barrier I have never gone past.
@@tight1449 Wikipedia's ethanol page? "Mixtures of ethanol and water form an azeotrope at about 89 mole-% ethanol and 11 mole-% water[78] or a mixture of 95.6% ethanol by mass (or about 97% alcohol by volume) at normal pressure, which boils at 351 K (78 °C)."
Or from the rectified spirit page "The purity of rectified spirit has a practical limit of 97.2% ABV (95.6% by mass)[2" The 2 there sources the claim to Whiskey Technology Production and Marketing (great book)
@@MrJhchrist I'm not surprised there is a discrepancy. It is almost always quoted by mass as 95.63 or mole fraction. I could not find any direct measurements by volume which is what we need as hydrometers are calibrated by volume. The conversion seems to be made by calculation. I have not previously seen it as 97.2 but have as 96.988 and 96.51. My measurements are limited by hydrometer and thermometer calibration but according to my measurements it is somewhere between 96.5 and 97. I cannot prove 97.2 wrong though!