Calculates the course of column distillations and displays them graphically.
The individual plates of a column are shown.
First the state after filling and heating up the boiler and filling the column.
Distillation can then be started.
The alcohol strengths on the plates, the quantities and alcohol strengths of the vapor and the reflux and all temperatures are shown.
And after starting the distillation, the quantity and alcohol strength of the collected distillate, the elapsed time and the energy consumed are also displayed.
The effect of the product draw-off on the alcohol concentrations in the column is not carried out according to McCabe-Thiele in this calculator. The same boiling point diagram data is used as in our other calculators, but instead of using the simple rules of the McCabe-Thiele method, the values are calculated according to the following rules:
A calculation that follows these two rules can only be carried out using a simulation. The calculator first roughly calculates the ratios in the column at either 0 % or 100 % reflux and then calculates in very small steps how the values in the column slowly change. If there is not much change, the result is displayed.
This type of calculation takes some time. That's why this simulator doesn't react quite as quickly as our old McCabe-Thiele Column Simulator.
For this reason, and because it is not so important there, this type of calculation is not built into our Potstill and Thumper Simulator, but it is still calculated there according to McCabe-Thiele.
The simulator also calculates the boiler content after filling the column. This means that it subtracts the liquid in the plates from the contents of the boiler, so that there is less liquid in the boiler, usually with a lower alcohol strength, than was "filled in" using the sliders. This liquid in the column is also called "holdup". The alcohol strength in the boiler is therefore set using the %abv slider before the plates are filled. Small sliders can be used to enter the liters in the boiler before distillation and the ml volume of the plates. If you do not want this function, i.e. you want the boiler to contain exactly the alcohol strength and quantity that you enter, you can simply set the "ml per plate" to 0.
The plate efficiency can be set between 50 and 100 %. In our calculation, an efficiency of 50 % means that for each plate, half of the vapor flowing into the plate from below passes through the liquid in the plate unchanged. This incomplete interaction of vapor and liquid leads to less alcohol concentrating in the column.
If you have a column with packing, you should normally set the efficiency slider to 100 %, as a packing height corresponding to a real plate in terms of separation capacity means a real plate with 100 % efficiency.
Most reflux stills have a reflux condenser that condenses all the vapor. The product and reflux are either separated before condensation, for example with VM, or afterwards, for example with LM. With this full condensation, the vapor below the reflux condenser, the reflux and the product have the same alcohol strength. However, the temperature of the reflux is lower, exactly one step lower on the boiling diagram. CM reflux stills, on the other hand, have a partial condenser. Only the reflux portion is condensed, the product continues to flow in vapor form to the product condenser. This partial condensation results in additional alcohol concentration. The vapor above the reflux condenser, which then becomes the product, has a higher alcohol strength and lower temperature than the vapor below the reflux condenser and the reflux has both a lower alcohol strength and a lower temperature than the vapor below the reflux condenser. There is a switch in the simulator that allows you to switch between complete condensation and partial condensation. If it is set to partial condensation, you can set an efficiency of between 70 and 100 % for partial condensation. 100% efficiency means that at 100 % reflux, the condenser produces an additional alcohol concentration of exactly one theor. plate and at each reflux setting, the alcohol strength of the vapor above the partial condenser and that of the reflux flowing down from the partial condenser are exactly one theor. plate apart. This is what is achieved in theory by partial condensation. 70% efficiency means that only 70 % of the alcohol concentration is produced, which would take place at 100 % efficiency. And the distance of alcohol strengths between product and reflux is then less than one theor. plate. At first glance, this means that you can achieve up to one theor. plate more with a partial condenser with 100 % efficiency than with a complete condenser. However, this is not entirely true for two reasons:
Firstly, just like a real plate, which can only cause one theor. plate at 100 % reflux, a partial condenser can only do this at 100 % reflux, because the plate or partial condensation not only results in a higher alcohol strength of the vapor above it, but also a lower alcohol strength of the reflux. This lower alcohol reflux flows downwards and reduces the alcohol strength in the plates below it, causing lower alcohol vapor to flow from these plates, which partially offsets the additional alcohol concentration caused by the plate or partial condensation. The simulation shows that a still with few plates can still achieve quite a lot with partial condensation, but much less a high column for neutral alcohol. Further information: Partial Condensation
The calculated theor. plates refer to the alcohol strength in the boiler after filling the column. In other words, not to the alcohol strength entered before distillation, but to the actual current alcohol strength in the boiler.
The colors of the column refer to the alcohol strength. The color code is: black for 0 %abv, red for 50 %abv, yellow for 75 %abv, green for 90 %abv and blue for 100 %abv.
You can switch between g/min and ml/min. The ml refers to the volume at 20 °C. This is because the unit %abv also refers to 20 °C by definition. And 20 °C naturally means a liquid state. So 100 ml/min of steam at 50 %abv means that it would have 100 ml/min and 50 %abv when condensed and cooled down to 20 °C. However, its displayed temperature naturally refers to the actual temperature of the steam in the column.
If you change the sliders, an initial distillation state is calculated each time. This is the state that is reached after heating and filling the plates and before the distillate is drawn off. However, since stable conditions are not yet established during heating and filling, and this takes some time, this is not entirely realistic.
Once the initial state has been calculated, distillation can be simulated. The quantity and alcohol strength of the collected distillate, the elapsed time and the energy expended are displayed and the column changes over time due to the decreasing alcohol content in the boiler. Three speeds can be selected: ►, ►► or ►►►. The simulation can be paused with ❙❙. The control system can then be adjusted. What can be adjusted is based on practice: The control of the reflux, the heating capacity, the plate efficiency, the air pressure and, in the case of CM, the cooling water inlet temperature. The new state is then calculated and distillation can continue. In practice, the new state would of course only be reached after a delay. Distillation is ended with ◾. You can then go through the distillation again using the slider at the bottom. Pressing Reset deletes these values and an initial state is calculated and displayed again.
Four different column controls can be calculated. The first three are more theoretical ones:
The needle valves in LM and RLM change their flow rate when the alcohol content changes. Water flows through constrictions more easily than ethanol. But temperature also has an effect: Hot liquids flow through more easily than cold ones. And the temperature, in turn, also depends on the alcohol content. And needle valves reduce the flow rate somewhat as the temperature rises. This means that they allow considerably more distillate to pass through at the start of distillation than when they are heated a short time later. We know from practical experience that an LM reduces the % reflux as the boiler alcohol strength falls, which then further reduces the alcohol strength of the distillate. This causes the alcohol strength to decrease quite suddenly towards the end of the distillation without warning, which can ruin the clean distillate collected in the same container. This is a major disadvantage of this system. This is why some people prefer to use the more complex RLM. Unlike LM, the alcohol strength does not decrease towards the end of the distillation, but the % reflux increases, which counteracts the decrease.
The behavior of VM is extremely complex. Theoretically, a VM seems to divide the vapor into two parts and thus set a constant % reflux. In practice, however, it turns out that it makes a difference for the vapor whether it goes up to the reflux condenser or sideways through the valve to the product condenser. And this depends not only on the pipe diameters and the diameter of the valve, but also to a large extent on the alcohol content of the vapor. Vapor with a low alcohol content generally only goes upwards. This VM-specific characteristic of switching to 100 % reflux when there is no longer enough alcohol is often referred to as "shut down", as the still virtually switches off as soon as there is no more clean distillate to be obtained. The reason for this is still a matter of conjecture. It seems plausible that it is due to the very low density of water vapor. It is lower than that of air. Ethanol vapor, on the other hand, has a higher density than air. This could create a kind of chimney effect at low alcohol content, i.e. there is an air movement from the product cooler outlet to the upper end of the reflux condenser, which prevents vapor from moving to the product condenser. The path to the product condenser would then be the opposite direction to the air movement. However, our tests have shown that this effect is not stopped if the reflux condenser is closed at the top, i.e. there can be no draught. That throws this theory out the window. However, it does not mean that the different densities cannot still be the cause, or at least one cause, of the VM-specific behavior. And the behavior of VM seems to be very easily disturbed. For example, you can change the amount of distillate by blowing into one of the two openings and this new state is then partially maintained without further blowing.
It is not easy to rely on the visual impression when it comes to correctly estimating the amount of distillate. High-alcohol distillate flows much more evenly than low-alcohol distillate, giving the impression that it must be a larger quantity. However, a high-alcohol stream of liquid is much thinner than a low-alcohol stream, which is barely visible. As a result, the higher the alcohol strength, the greater the overestimation of the strength of the distillate flow.
Or even if you only collect distillate drop by drop, you cannot rely on the fact that 20 drops are 1 ml, as you often read. Although this is very accurate when dripping water from a pipette, it is neither accurate for alcohol solutions dripped from a pipette nor for drops of water from a tube. On the one hand, drops of alcohol solutions are smaller than drops of water, on the other hand, drops from a tube are considerably larger than drops from a pipette. Here we are dealing with orders of magnitude of doubling or halving.
Now the question is what you can do with the simulator if the controls are largely theoretical. In other words, to what extent can this be transferred into practice? In principle, the simulator's controls can be used to simulate real distillation controls. For example, if you set it to collect 100 ml/min of product, this corresponds to any distillery that collects 100 ml/min of product in reality, regardless of whether it is an LM or a VM, for example. However, in practice you cannot expect these 100ml/min to remain constant if the alcohol content decreases over time. Therefore, it would be nice if you could directly select LM or VM, where the behavior of these controls is realistically simulated. However, for the reasons mentioned above, this would be extremely speculative and would therefore only give the impression of more benefit. This is a basic problem with calculations like ours: That at least users with little background knowledge and no patience to read the notes carefully assume absolute accuracy.
The simulator does not take heating power losses into account. The losses are generalized rather than relative. So a still operated with 1000 watts does not lose half as much as one operated with 2000 watts, but exactly the same or only slightly less, depending on the heating method (internal or external heating). Our Potstill and Thumper Simulator has a simulation of heating power losses. Here you can play around and perhaps get a feel for how much energy you are losing. You can read more about this in the note there.
Atmospheric pressures between 75 and 10000 hPa can be calculated. The consideration of the atmospheric pressure has a substantial influence on the result. If nothing is entered, the calculator assumes the local atmospheric pressure 1013.25 hPa.
Since almost no one has an absolutely accurate indicating thermometer, an additional "thermometer error" can be specified. This can be determined with the help of the calculator Thermometer Error. The temperatures calculated here are then those displayed on this thermometer, not the real ones. However, the cooling water temperature to be specified for CM must be the real temperature, regardless of the thermometer error entered.
Information about our boiling point data and about the influence of atmospheric pressure
The effect of the product draw-off on the alcohol concentrations in the column is not carried out according to McCabe-Thiele in this calculator. The same boiling point diagram data is used as in our other calculators, but instead of using the simple rules of the McCabe-Thiele method, the values are calculated according to the following rules:
- What leaves each plate downwards as reflux and upwards as vapor is exactly the same as what is added to each plate from above as reflux and from below as vapor.
- What is taken from the top as distillate is equal to the difference between the vapor produced by the boiler and what flows back into the boiler as reflux.
A calculation that follows these two rules can only be carried out using a simulation. The calculator first roughly calculates the ratios in the column at either 0 % or 100 % reflux and then calculates in very small steps how the values in the column slowly change. If there is not much change, the result is displayed.
This type of calculation takes some time. That's why this simulator doesn't react quite as quickly as our old McCabe-Thiele Column Simulator.
For this reason, and because it is not so important there, this type of calculation is not built into our Potstill and Thumper Simulator, but it is still calculated there according to McCabe-Thiele.
The simulator also calculates the boiler content after filling the column. This means that it subtracts the liquid in the plates from the contents of the boiler, so that there is less liquid in the boiler, usually with a lower alcohol strength, than was "filled in" using the sliders. This liquid in the column is also called "holdup". The alcohol strength in the boiler is therefore set using the %abv slider before the plates are filled. Small sliders can be used to enter the liters in the boiler before distillation and the ml volume of the plates. If you do not want this function, i.e. you want the boiler to contain exactly the alcohol strength and quantity that you enter, you can simply set the "ml per plate" to 0.
The plate efficiency can be set between 50 and 100 %. In our calculation, an efficiency of 50 % means that for each plate, half of the vapor flowing into the plate from below passes through the liquid in the plate unchanged. This incomplete interaction of vapor and liquid leads to less alcohol concentrating in the column.
If you have a column with packing, you should normally set the efficiency slider to 100 %, as a packing height corresponding to a real plate in terms of separation capacity means a real plate with 100 % efficiency.
Most reflux stills have a reflux condenser that condenses all the vapor. The product and reflux are either separated before condensation, for example with VM, or afterwards, for example with LM. With this full condensation, the vapor below the reflux condenser, the reflux and the product have the same alcohol strength. However, the temperature of the reflux is lower, exactly one step lower on the boiling diagram. CM reflux stills, on the other hand, have a partial condenser. Only the reflux portion is condensed, the product continues to flow in vapor form to the product condenser. This partial condensation results in additional alcohol concentration. The vapor above the reflux condenser, which then becomes the product, has a higher alcohol strength and lower temperature than the vapor below the reflux condenser and the reflux has both a lower alcohol strength and a lower temperature than the vapor below the reflux condenser. There is a switch in the simulator that allows you to switch between complete condensation and partial condensation. If it is set to partial condensation, you can set an efficiency of between 70 and 100 % for partial condensation. 100% efficiency means that at 100 % reflux, the condenser produces an additional alcohol concentration of exactly one theor. plate and at each reflux setting, the alcohol strength of the vapor above the partial condenser and that of the reflux flowing down from the partial condenser are exactly one theor. plate apart. This is what is achieved in theory by partial condensation. 70% efficiency means that only 70 % of the alcohol concentration is produced, which would take place at 100 % efficiency. And the distance of alcohol strengths between product and reflux is then less than one theor. plate. At first glance, this means that you can achieve up to one theor. plate more with a partial condenser with 100 % efficiency than with a complete condenser. However, this is not entirely true for two reasons:
Firstly, just like a real plate, which can only cause one theor. plate at 100 % reflux, a partial condenser can only do this at 100 % reflux, because the plate or partial condensation not only results in a higher alcohol strength of the vapor above it, but also a lower alcohol strength of the reflux. This lower alcohol reflux flows downwards and reduces the alcohol strength in the plates below it, causing lower alcohol vapor to flow from these plates, which partially offsets the additional alcohol concentration caused by the plate or partial condensation. The simulation shows that a still with few plates can still achieve quite a lot with partial condensation, but much less a high column for neutral alcohol. Further information: Partial Condensation
The calculated theor. plates refer to the alcohol strength in the boiler after filling the column. In other words, not to the alcohol strength entered before distillation, but to the actual current alcohol strength in the boiler.
The colors of the column refer to the alcohol strength. The color code is: black for 0 %abv, red for 50 %abv, yellow for 75 %abv, green for 90 %abv and blue for 100 %abv.
You can switch between g/min and ml/min. The ml refers to the volume at 20 °C. This is because the unit %abv also refers to 20 °C by definition. And 20 °C naturally means a liquid state. So 100 ml/min of steam at 50 %abv means that it would have 100 ml/min and 50 %abv when condensed and cooled down to 20 °C. However, its displayed temperature naturally refers to the actual temperature of the steam in the column.
If you change the sliders, an initial distillation state is calculated each time. This is the state that is reached after heating and filling the plates and before the distillate is drawn off. However, since stable conditions are not yet established during heating and filling, and this takes some time, this is not entirely realistic.
Once the initial state has been calculated, distillation can be simulated. The quantity and alcohol strength of the collected distillate, the elapsed time and the energy expended are displayed and the column changes over time due to the decreasing alcohol content in the boiler. Three speeds can be selected: ►, ►► or ►►►. The simulation can be paused with ❙❙. The control system can then be adjusted. What can be adjusted is based on practice: The control of the reflux, the heating capacity, the plate efficiency, the air pressure and, in the case of CM, the cooling water inlet temperature. The new state is then calculated and distillation can continue. In practice, the new state would of course only be reached after a delay. Distillation is ended with ◾. You can then go through the distillation again using the slider at the bottom. Pressing Reset deletes these values and an initial state is calculated and displayed again.
Four different column controls can be calculated. The first three are more theoretical ones:
- Control of % reflux:
Here, the % reflux is simply set using a slider.
This is easy to understand, but unfortunately there is no still design that allows the % reflux to be set in a stable manner regardless of other factors.
Vapor Management (VM) seems to come closest to this control method.
But only at very high alcohol strengths.
This is because the % reflux increases very sharply with VM as the alcohol strength at the valve level falls.
In practice, it is therefore normally not even possible to set 0 % reflux.
The % reflux remain constant at 90.2 as set on the slider.
The alcohol strength decreases from 94.8 to 87.5 %abv within one hour. - Control of product amount:
Here, a slider is used to set how many ml or grams of product are withdrawn per minute.
Liquid management (LM) seems to correspond most closely to this control method, i.e. product withdrawal from the top reflux with a needle valve.
But only if the alcohol strength remains the same.
This is because if the alcohol strength changes, the flow rate through the needle valve also changes.
If the same amount of product is constantly withdrawn, the % reflux decrease over time, which further reduces the alcohol strength:
The % reflux decrease during distillation from 90.2 to 78.7.
And the alcohol strength decreases heavily from 94.8 to 68.8 %abv. - Control of reflux amount:
The same here, but the reflux quantity is controlled directly rather than the product quantity.
Reverse Liquid Management (RLM) seems to correspond most closely to this control, i.e. the top reflux is fed back into the column through a needle valve, whereby the overflow is then drawn off as product.
However, this is also only possible if the alcohol strength remains constant, as the flow rate through the needle valve also depends on the alcohol strength.
Cooling Management (CM) is also similar.
This is because the control of the cooling water directly determines the specific reflux quantity.
However, the reflux quantity here also depends on the alcohol strength.
This is mainly because a high alcohol strength means a low temperature and therefore a smaller temperature difference to the cooling water and therefore less reflux per cooling water.
By constantly returning the same amount of reflux to the column, the % reflux will increase over time, stabilizing the alcohol strength:
The % reflux increase during distillation from 90.2 to 95.9.
The alcohol strength still decreases, but only very slightly from 94.8 to 93.2 %abv. - Cooling management (CM):
Here, a slider is used to open either a small or a large valve, which supplies the reflux condenser with more or less cooling water.
The small valve lets a maximum of 200 ml/min of cooling water through, the large valve a maximum of 2000 ml/min.
Depending on the amount of cooling water, there is more or less reflux.
The rest passes through the reflux condenser as vapor and via a bend to the product condenser.
The reflux condenser is therefore always a partial condenser here.
The calculator assumes that the cooling water in the reflux condenser is heated up to the reflux temperature. This is the theoretically maximum possible cooling water efficiency.
The inlet temperature of the cooling water can be set.
With this control, the calculator also takes " subcooled" reflux into account. This means that if more cooling water is used than is necessary for 100 % reflux, the reflux from the reflux condenser becomes colder. This results in less vapor development on the top plate. And since this vapor then condenses completely at 100 % reflux and drips back onto the top plate as reflux, there is naturally also less reflux on the top plate. The simulation shows that no other changes occur. Subcooled reflux therefore appears to have no effect apart from wasting water. The temperature of the subcooled reflux is also shown in the simulation.
The cooling water also has this temperature after cooling.
If you constantly use the same amount of cooling water at the same temperature in a CM, the % reflux remains very stable. It increases a little over time, i.e. as the alcohol content decreases, which stabilizes the alcohol strength somewhat:
The % reflux increase during distillation from 90.2 to 91.9.
As a result, the alcohol strength decreases quite moderately from 94.8 to 89.3 %abv.
The needle valves in LM and RLM change their flow rate when the alcohol content changes. Water flows through constrictions more easily than ethanol. But temperature also has an effect: Hot liquids flow through more easily than cold ones. And the temperature, in turn, also depends on the alcohol content. And needle valves reduce the flow rate somewhat as the temperature rises. This means that they allow considerably more distillate to pass through at the start of distillation than when they are heated a short time later. We know from practical experience that an LM reduces the % reflux as the boiler alcohol strength falls, which then further reduces the alcohol strength of the distillate. This causes the alcohol strength to decrease quite suddenly towards the end of the distillation without warning, which can ruin the clean distillate collected in the same container. This is a major disadvantage of this system. This is why some people prefer to use the more complex RLM. Unlike LM, the alcohol strength does not decrease towards the end of the distillation, but the % reflux increases, which counteracts the decrease.
The behavior of VM is extremely complex. Theoretically, a VM seems to divide the vapor into two parts and thus set a constant % reflux. In practice, however, it turns out that it makes a difference for the vapor whether it goes up to the reflux condenser or sideways through the valve to the product condenser. And this depends not only on the pipe diameters and the diameter of the valve, but also to a large extent on the alcohol content of the vapor. Vapor with a low alcohol content generally only goes upwards. This VM-specific characteristic of switching to 100 % reflux when there is no longer enough alcohol is often referred to as "shut down", as the still virtually switches off as soon as there is no more clean distillate to be obtained. The reason for this is still a matter of conjecture. It seems plausible that it is due to the very low density of water vapor. It is lower than that of air. Ethanol vapor, on the other hand, has a higher density than air. This could create a kind of chimney effect at low alcohol content, i.e. there is an air movement from the product cooler outlet to the upper end of the reflux condenser, which prevents vapor from moving to the product condenser. The path to the product condenser would then be the opposite direction to the air movement. However, our tests have shown that this effect is not stopped if the reflux condenser is closed at the top, i.e. there can be no draught. That throws this theory out the window. However, it does not mean that the different densities cannot still be the cause, or at least one cause, of the VM-specific behavior. And the behavior of VM seems to be very easily disturbed. For example, you can change the amount of distillate by blowing into one of the two openings and this new state is then partially maintained without further blowing.
It is not easy to rely on the visual impression when it comes to correctly estimating the amount of distillate. High-alcohol distillate flows much more evenly than low-alcohol distillate, giving the impression that it must be a larger quantity. However, a high-alcohol stream of liquid is much thinner than a low-alcohol stream, which is barely visible. As a result, the higher the alcohol strength, the greater the overestimation of the strength of the distillate flow.
Or even if you only collect distillate drop by drop, you cannot rely on the fact that 20 drops are 1 ml, as you often read. Although this is very accurate when dripping water from a pipette, it is neither accurate for alcohol solutions dripped from a pipette nor for drops of water from a tube. On the one hand, drops of alcohol solutions are smaller than drops of water, on the other hand, drops from a tube are considerably larger than drops from a pipette. Here we are dealing with orders of magnitude of doubling or halving.
Now the question is what you can do with the simulator if the controls are largely theoretical. In other words, to what extent can this be transferred into practice? In principle, the simulator's controls can be used to simulate real distillation controls. For example, if you set it to collect 100 ml/min of product, this corresponds to any distillery that collects 100 ml/min of product in reality, regardless of whether it is an LM or a VM, for example. However, in practice you cannot expect these 100ml/min to remain constant if the alcohol content decreases over time. Therefore, it would be nice if you could directly select LM or VM, where the behavior of these controls is realistically simulated. However, for the reasons mentioned above, this would be extremely speculative and would therefore only give the impression of more benefit. This is a basic problem with calculations like ours: That at least users with little background knowledge and no patience to read the notes carefully assume absolute accuracy.
The simulator does not take heating power losses into account. The losses are generalized rather than relative. So a still operated with 1000 watts does not lose half as much as one operated with 2000 watts, but exactly the same or only slightly less, depending on the heating method (internal or external heating). Our Potstill and Thumper Simulator has a simulation of heating power losses. Here you can play around and perhaps get a feel for how much energy you are losing. You can read more about this in the note there.
Atmospheric pressures between 75 and 10000 hPa can be calculated. The consideration of the atmospheric pressure has a substantial influence on the result. If nothing is entered, the calculator assumes the local atmospheric pressure 1013.25 hPa.
Since almost no one has an absolutely accurate indicating thermometer, an additional "thermometer error" can be specified. This can be determined with the help of the calculator Thermometer Error. The temperatures calculated here are then those displayed on this thermometer, not the real ones. However, the cooling water temperature to be specified for CM must be the real temperature, regardless of the thermometer error entered.
Information about our boiling point data and about the influence of atmospheric pressure