Instantaneous RR measurement
Moderator: Site Moderator
Instantaneous RR measurement
In a VM still, all of the vapour that gets to the reflux condenser is turned into liquid, and the part that goes to the product condenser is turned into a liquid in that path. So, as I understand it, the reflux ratio is:
RR = [mass of reflux liquid]/[(mass of reflux liquid)+(mass of product liquid)]
Now on page 44 of The Compleat Distiller, this would seem to be the definition of the product ratio, and the definition of the reflux ratio would seem to be the percentage of the vapour that is condensed in the reflux condenser out of the total vapour entering it. If any vapour is not condensensed in the reflux condenser, then it must be going to the product condenser, otherwise you would have vapour venting from the reflux condenser.
So, is there really any difference between Reflux Ratio and Product Ratio in a VM still?
Assuming that my definition is correct, then the rate of liquid condensed in each condenser would be equal to:
liquid condensation rate = (rate of coolant flow) X (temperature difference between in and out coolant) X (specific heat of water [=1]) / (latent heat of vapourisation of vapour)
But in a VM still, the same water flows at the same rate through the product and the reflux condensers, and the same vapour is going into both condensers, so when you plug the two liquids into the RR equation and cancel out the coolant flow rates and the LHV, you get:
RR = [temp diff between reflux in and out]/[(temp diff between reflux in and out)+(temp diff between product in and product out)]
or, since the product out and reflux in are the same:
RR= (temp diff between reflux in and out)/(temp diff between input to product condenser and output of reflux condenser)
Any cooling of the condensate would cause a small error, but since the specific heat of the liquid is so much tinier than the latent heat of vapourisation, it can probably be ignored.
So, if I install temperature sensors at the input to the product condenser, in the line between the product and reflux condensers, and at the output of the reflux condenser, could I measure the instantaneous reflux ratio?
M
RR = [mass of reflux liquid]/[(mass of reflux liquid)+(mass of product liquid)]
Now on page 44 of The Compleat Distiller, this would seem to be the definition of the product ratio, and the definition of the reflux ratio would seem to be the percentage of the vapour that is condensed in the reflux condenser out of the total vapour entering it. If any vapour is not condensensed in the reflux condenser, then it must be going to the product condenser, otherwise you would have vapour venting from the reflux condenser.
So, is there really any difference between Reflux Ratio and Product Ratio in a VM still?
Assuming that my definition is correct, then the rate of liquid condensed in each condenser would be equal to:
liquid condensation rate = (rate of coolant flow) X (temperature difference between in and out coolant) X (specific heat of water [=1]) / (latent heat of vapourisation of vapour)
But in a VM still, the same water flows at the same rate through the product and the reflux condensers, and the same vapour is going into both condensers, so when you plug the two liquids into the RR equation and cancel out the coolant flow rates and the LHV, you get:
RR = [temp diff between reflux in and out]/[(temp diff between reflux in and out)+(temp diff between product in and product out)]
or, since the product out and reflux in are the same:
RR= (temp diff between reflux in and out)/(temp diff between input to product condenser and output of reflux condenser)
Any cooling of the condensate would cause a small error, but since the specific heat of the liquid is so much tinier than the latent heat of vapourisation, it can probably be ignored.
So, if I install temperature sensors at the input to the product condenser, in the line between the product and reflux condensers, and at the output of the reflux condenser, could I measure the instantaneous reflux ratio?
M
Re: Instantaneous RR measurement
Seems like you could figure it that way but your heat input and the ABV of the boiler charge would have to be the same all the time in order to get a consistent " true" RR.....I think.
Well shoot maybe not as it is a ratio then.
Well shoot maybe not as it is a ratio then.
Trample the injured and hurdle the dead.
Re: Instantaneous RR measurement
You don't actually need to know what you reflux ratio is. In addition, the optimum reflux ratio changes during the progress of the run. What is important is the vapour temperature going into your VM port.
Provided you can change your reflux ratio in order to maintain the product purity that you require, then it does not matter what that reflux ratio actually is.
Provided you can change your reflux ratio in order to maintain the product purity that you require, then it does not matter what that reflux ratio actually is.
Re: Instantaneous RR measurement
Don't over-complicate the whole operation of a reflux column... You're trying to sweat too many details, some of which simply aren't as important as you're playing them up to be... Temperature comparisons won't get you there... As myles stated, we really don't care what the exact reflux ratio is, just that it is keeping the column at equilibrium throughout the run... You could spend the rest of your life puttering around trying to place finite numbers on an ever moving target and still not be any smarter so why bother with an exercise in futility... Sometimes we need to leave the magic alone and just be happy we are making better spirits than commercial distilleries... I try to leave as much math and physics out of my enjoyment of this hobby, otherwise I might as well do something else to help me relax...
Some VM columns do have separate coolant flow circuits for reflux and product branches for better fine tuning...
Some VM columns do have separate coolant flow circuits for reflux and product branches for better fine tuning...
Re: Instantaneous RR measurement
Some people like to relax with a good engineering handbook or an instruction manual for a new machine! When I got my first McMaster-Carr catalogue, I spent a month reading it cover-to-cover. Knowledge and understanding keep life interesting.
From the start, I incorporated a microcontroller into my still. Right now, it is only used to control boiler power, which is set precisely by a potentiometer. If I can measure reflux ratio, then I can also control it. Imagine if you had a still with a knob labelled Reflux Ratio, right there to set as you pleased. Would you use it? If so, how can you imagine using it?
M
From the start, I incorporated a microcontroller into my still. Right now, it is only used to control boiler power, which is set precisely by a potentiometer. If I can measure reflux ratio, then I can also control it. Imagine if you had a still with a knob labelled Reflux Ratio, right there to set as you pleased. Would you use it? If so, how can you imagine using it?
M
Re: Instantaneous RR measurement
This is one of the nice things about a VM. The reflux ratio is dependent on the ratio of the cross sectional areas of the two ports, into the reflux condenser chamber and into the product branch. The actual composition of the vapour does not matter. At the start of the run the vapour contains a higher proportion of alcohol than at the end of the run. But it does not matter as you are splitting the vapour and not just the alcohol carried within it.
In contrast with an LM you condense all the vapour and then return a certain volume / flow rate, of that condensate to the column. In this situation your reflux ratio does change with the composition of the vapour, and varies throughout the run.
Even if you did know what it is I am not sure how valuable it would be. Our main interest is in the EFFECT of the reflux ratio, and that is changing the composition of the vapour itself. We currently use the temperature to indicate the composition of that vapour.
Although, I am just waiting for someone to plumb in a vapour sealed hydrometer for a direct measurement.
In contrast with an LM you condense all the vapour and then return a certain volume / flow rate, of that condensate to the column. In this situation your reflux ratio does change with the composition of the vapour, and varies throughout the run.
Even if you did know what it is I am not sure how valuable it would be. Our main interest is in the EFFECT of the reflux ratio, and that is changing the composition of the vapour itself. We currently use the temperature to indicate the composition of that vapour.
Although, I am just waiting for someone to plumb in a vapour sealed hydrometer for a direct measurement.
Re: Instantaneous RR measurement
If you are going to use a microcontroller to assist or even operate your still, you have to set the goal: produce the best quality product. If you want to measure the quality of the product, you must first make a quality sensor. The quality of the product is directly measured with an alcoholometer, but of course there is smell and taste, for which sensors might be hard to find or make.
So if you control for quality via an instrumented alcoholometer, you have to set some background parameters:
1. boiler power
2. coolant flow rate
3. product vapour valve opening
Now, all these can can be set by following a fixed (or maybe partly algorithmic) formula. You, the operator, are the sensor for the still, and you are telling it what to do. Whatever you do, the microcontroller records the actions and the times, and then re-performs them at your will. Many industial robots work this way. It may be preferable, in this case, because the individual can set up his unique routine to produce his distinctive product. Or, if you have identical stills, perform the actions of some other 'stiller.
If you add sensors, you can set goals for finding the optimum values of these background parameters to get the fastest, maximum %ABV product.
These sensors might include:
1. vapour temperature at top of column
2. %ABV (to be invented)
3. coolant temperature at
a. input to product condenser
b. tubing between output of product condenser and input to reflux condenser
c. output of reflux condenser
d. output of coolant cooler (fan-powered radiator, in my case)
4. boiler temperature
5. boiler liquid level
6. gate valve opening
7. coolant flow rate
8... ?
M
So if you control for quality via an instrumented alcoholometer, you have to set some background parameters:
1. boiler power
2. coolant flow rate
3. product vapour valve opening
Now, all these can can be set by following a fixed (or maybe partly algorithmic) formula. You, the operator, are the sensor for the still, and you are telling it what to do. Whatever you do, the microcontroller records the actions and the times, and then re-performs them at your will. Many industial robots work this way. It may be preferable, in this case, because the individual can set up his unique routine to produce his distinctive product. Or, if you have identical stills, perform the actions of some other 'stiller.
If you add sensors, you can set goals for finding the optimum values of these background parameters to get the fastest, maximum %ABV product.
These sensors might include:
1. vapour temperature at top of column
2. %ABV (to be invented)
3. coolant temperature at
a. input to product condenser
b. tubing between output of product condenser and input to reflux condenser
c. output of reflux condenser
d. output of coolant cooler (fan-powered radiator, in my case)
4. boiler temperature
5. boiler liquid level
6. gate valve opening
7. coolant flow rate
8... ?
M
-
- Angel's Share
- Posts: 2228
- Joined: Tue Dec 07, 2010 8:20 am
- Location: Pacific Northwest
Re: Instantaneous RR measurement
Maritimer,
Reflux ratio is the percentage of vapor being condensed by your reflux condenser and returned to the packing as liquid to gain additional evaporation cycles.
Reflux ratio is not controlled or measured by temperature. It is determined by take off rate as compared to power in and then only a rough estimate as pressure changes and alcohol water ratio affect vapor volume. Watt input = vapor speed.
If you run your two inch column at 1000 watts it has a vapor speed of 14.5 inches per second. At 2 inch diameter that is 45.55 cu inches a second or 27330 cubic inches in ten minutes.
At one atmosphere 1 cubic inch of water contains 1600 cubic inches of vapor at 101c. 1 ounce of water contains 1.8 cu inches. 1.8 x 1600 = 2880 cu inches vapor. 2880/273300 = 10.5. If you collect 1 oz of water in ten minutes your reflux ratio is 89.5% This assumes one atmosphere.
That is water. Go work the vapor volume for pure alcohol for the other half and you can calculate the correct liquid vapor ratio to use for your RR % in a 10% wash
Further measuring ABV will tell you the reflux ratio necessary with your particular still / packing to achieve your goals. Calculating HETP is still another step over my head.
Measuring temperature at different locations can be useful. Boiler temperature will tell you wash composition and when nearing tails. Head temp will tell you that vapor has reached the top or when nearing tails. Coolant exit temperature confirms settings necessary to repeat your success. Temperature of distillate allows accurate meter readings.
Reflux ratio is the percentage of vapor being condensed by your reflux condenser and returned to the packing as liquid to gain additional evaporation cycles.
Reflux ratio is not controlled or measured by temperature. It is determined by take off rate as compared to power in and then only a rough estimate as pressure changes and alcohol water ratio affect vapor volume. Watt input = vapor speed.
If you run your two inch column at 1000 watts it has a vapor speed of 14.5 inches per second. At 2 inch diameter that is 45.55 cu inches a second or 27330 cubic inches in ten minutes.
At one atmosphere 1 cubic inch of water contains 1600 cubic inches of vapor at 101c. 1 ounce of water contains 1.8 cu inches. 1.8 x 1600 = 2880 cu inches vapor. 2880/273300 = 10.5. If you collect 1 oz of water in ten minutes your reflux ratio is 89.5% This assumes one atmosphere.
That is water. Go work the vapor volume for pure alcohol for the other half and you can calculate the correct liquid vapor ratio to use for your RR % in a 10% wash
Further measuring ABV will tell you the reflux ratio necessary with your particular still / packing to achieve your goals. Calculating HETP is still another step over my head.
Measuring temperature at different locations can be useful. Boiler temperature will tell you wash composition and when nearing tails. Head temp will tell you that vapor has reached the top or when nearing tails. Coolant exit temperature confirms settings necessary to repeat your success. Temperature of distillate allows accurate meter readings.
Last edited by mash rookie on Wed Oct 10, 2012 6:00 pm, edited 1 time in total.
Re: Instantaneous RR measurement
Thanks for your replies, Mash Rookie and gang.
I've looked over the posts from the people who have replied to this thread, and I am really impressed. Some day I hope to be able to keep up with you. I'm a retired research lab technician--35 years at a big university--so I like working with professorial types!
MR, here is your definition of reflux ratio: "Reflux ratio is the percentage of vapor being condensed be your reflux condenser and returned to the packing as liquid to gain additional evaporation cycles."
Did you mean: Reflux ratio is the ratio of vapor being condensed in your reflux condenser to that returned to the packing as liquid to gain additional evaporation cycles.
Myles says: "This is one of the nice things about a VM. The reflux ratio is dependent on the ratio of the cross sectional areas of the two ports, into the reflux condenser chamber and into the product branch. The actual composition of the vapour does not matter. At the start of the run the vapour contains a higher proportion of alcohol than at the end of the run. But it does not matter as you are splitting the vapour and not just the alcohol carried within it."
MR, how could the reflux ratio as you define it be anything but 100%? If the vapour going into the reflux condenser is not condensed, then it will be vented. The RR as Myles defines it takes the vapour coming from the column and splits it into two paths; the fraction of the column flow that goes to the reflux condenser is the RR. That is how I understand RR, too. Maybe I need and understanding update?
My contention is that the vapour that is condensed in each path transforms its LHV into coolant heating. By measuring the coolant heating through each condenser, we can determine the amount of vapour that went to each condenser (if we know its composition, of course). By taking the ratio of the coolant heating in each condenser, the composition of the vapours cancels out, and since the rate of coolant flow is the same in each condenser, the long-term temperature difference across each one will settle down as the coolant in each condenser is refreshed and heated.
I've got 15 liters of 45% low wines sitting with bicarb for a week, and then I'll be doing another run.
I've got four quite accurate temperature sensors (Accu-temp from Canadian Tire).
The condensers will have to be insulated.
First check will be to keep a constant gate valve opening and then change boiler power. That will change vapour rate. If the ratio of temperature differences remains the same, that will be one step to verifying the idea.
Then the coolant rate will be changed. That should also cancel out.
And the vapour composition will change as the run progresses. That should also cancel out.
The coolant temperature is going to change. I'm using a pond pump and fans blowing through a radiator to extract the heat from the coolant. There will be temperature changes. That should also cancel out, as we are using temperature differences.
Lab report will follow.
M
I've looked over the posts from the people who have replied to this thread, and I am really impressed. Some day I hope to be able to keep up with you. I'm a retired research lab technician--35 years at a big university--so I like working with professorial types!
MR, here is your definition of reflux ratio: "Reflux ratio is the percentage of vapor being condensed be your reflux condenser and returned to the packing as liquid to gain additional evaporation cycles."
Did you mean: Reflux ratio is the ratio of vapor being condensed in your reflux condenser to that returned to the packing as liquid to gain additional evaporation cycles.
Myles says: "This is one of the nice things about a VM. The reflux ratio is dependent on the ratio of the cross sectional areas of the two ports, into the reflux condenser chamber and into the product branch. The actual composition of the vapour does not matter. At the start of the run the vapour contains a higher proportion of alcohol than at the end of the run. But it does not matter as you are splitting the vapour and not just the alcohol carried within it."
MR, how could the reflux ratio as you define it be anything but 100%? If the vapour going into the reflux condenser is not condensed, then it will be vented. The RR as Myles defines it takes the vapour coming from the column and splits it into two paths; the fraction of the column flow that goes to the reflux condenser is the RR. That is how I understand RR, too. Maybe I need and understanding update?
My contention is that the vapour that is condensed in each path transforms its LHV into coolant heating. By measuring the coolant heating through each condenser, we can determine the amount of vapour that went to each condenser (if we know its composition, of course). By taking the ratio of the coolant heating in each condenser, the composition of the vapours cancels out, and since the rate of coolant flow is the same in each condenser, the long-term temperature difference across each one will settle down as the coolant in each condenser is refreshed and heated.
I've got 15 liters of 45% low wines sitting with bicarb for a week, and then I'll be doing another run.
I've got four quite accurate temperature sensors (Accu-temp from Canadian Tire).
The condensers will have to be insulated.
First check will be to keep a constant gate valve opening and then change boiler power. That will change vapour rate. If the ratio of temperature differences remains the same, that will be one step to verifying the idea.
Then the coolant rate will be changed. That should also cancel out.
And the vapour composition will change as the run progresses. That should also cancel out.
The coolant temperature is going to change. I'm using a pond pump and fans blowing through a radiator to extract the heat from the coolant. There will be temperature changes. That should also cancel out, as we are using temperature differences.
Lab report will follow.
M
Re: Instantaneous RR measurement
I think the actual definition of the reflux ratio is
The percentage alcohol returned to the column : the percentage alcohol taken as product. (Percentages of the total alcohol available at the top of the column, at that moment in time.)
With the VM if your vapour paths are split 50:50 you will have a certain reflux ratio. That reflux ratio remains constant. It does not matter if the vapour is 70% alcohol or 10% alcohol.
Because the vapour path is split by the same amount during the run, so is the alcohol carried in that vapour path irrespective of the composition of that vapour.
The percentage alcohol returned to the column : the percentage alcohol taken as product. (Percentages of the total alcohol available at the top of the column, at that moment in time.)
With the VM if your vapour paths are split 50:50 you will have a certain reflux ratio. That reflux ratio remains constant. It does not matter if the vapour is 70% alcohol or 10% alcohol.
Because the vapour path is split by the same amount during the run, so is the alcohol carried in that vapour path irrespective of the composition of that vapour.
Re: Instantaneous RR measurement
Myles,
"The percentage alcohol returned to the column : the percentage alcohol taken as product. (Percentages of the total alcohol available at the top of the column, at that moment in time.)"
"It does not matter if the vapour is 70% alcohol or 10% alcohol."
If we have 100 parts vapour in the column and 80 parts goes back to the column and 20 parts goes to the product, then the RR is 80/20=4.
So instead of the formula I started with, it should have been:
RR = (mass of liquid condensed in reflux condenser) / (mass of liquid condensed in product condenser)
which, in terms of temperature differences, with identical flows in both condensers, becomes:
RR = (temperture difference across reflux condenser) / (temperture difference across product condenser)
at steady state, of course.
In The Compleat Distiller, RRs of 80% are suggested, but when equilibrating the column, it is run at 100%. This suggests that the original equation is closer to TCD's definition. If we use your definition, no product taken would be 100/0 = infinity.
In the original equation, no product gives: RR = (reflux mass condensed)/(reflux mass condensed + zero product) = 100/(100 + 0) = 1, or 100%.
If the still is run as a pot still, 0% reflux, RR = 0/(0 + 100) = 0, or 0%.
So the original equation jibes with TCD at 0 and 100, at least.
M
"The percentage alcohol returned to the column : the percentage alcohol taken as product. (Percentages of the total alcohol available at the top of the column, at that moment in time.)"
"It does not matter if the vapour is 70% alcohol or 10% alcohol."
If we have 100 parts vapour in the column and 80 parts goes back to the column and 20 parts goes to the product, then the RR is 80/20=4.
So instead of the formula I started with, it should have been:
RR = (mass of liquid condensed in reflux condenser) / (mass of liquid condensed in product condenser)
which, in terms of temperature differences, with identical flows in both condensers, becomes:
RR = (temperture difference across reflux condenser) / (temperture difference across product condenser)
at steady state, of course.
In The Compleat Distiller, RRs of 80% are suggested, but when equilibrating the column, it is run at 100%. This suggests that the original equation is closer to TCD's definition. If we use your definition, no product taken would be 100/0 = infinity.
In the original equation, no product gives: RR = (reflux mass condensed)/(reflux mass condensed + zero product) = 100/(100 + 0) = 1, or 100%.
If the still is run as a pot still, 0% reflux, RR = 0/(0 + 100) = 0, or 0%.
So the original equation jibes with TCD at 0 and 100, at least.
M
Re: Instantaneous RR measurement
It is all interesting but lets not forget reality. You never have a reflux ratio of zero. Even the most basic pot still has a tiny bit of reflux. Also the reflux ratio is not a percentage, it is a number. Under close to full reflux this could be 1000:1 or even 10,000:1.
In practice what is important is what we do. If the product quality drops we restrict the product qty taken to increase the reflux ratio. Take more product and the product ABV drops. Towards the end of the run when there is less alcohol available, even increasing the reflux ratio may not return enough alcohol to the column to maintain stability. Eventually the column will fail.
Also remember that the opposite is also true. At high vapour speeds increasing the reflux ratio can lead to the situation where the quality also drops.
The only time you have equal flows in both condensers is when you have a reflux ratio of 1:1, so utilising temperature differences might be of limited value.
I still think that knowing the actual reflux ratio is not going to help you a lot. To be able to apply it to control the column, you would also need to know the vapour composition at both the base and top of the column. If you are using LM then you also need to know the run profile, as the reflux ratio changes constantly throughout the run as the available alcohol is depleted.
You can't select a reflux ratio and set it for the entire run. Even with a VM towards the end of the run you cycle the column between full reflux and partial reflux, and this is determined by the product quality at that moment in time. When it starts to drop the column is put back into full reflux to restabilise, then the reflux ratio is slowly reduced to take a bit more product.
EDIT: 1 other minor point. The rate of coolant flow in both condensers is NOT the same. You should be running your reflux condenser at much lower coolant flow rate than your product condenser. You want the coolant exiting your reflux condenser to be warm, if not hot!!
In practice what is important is what we do. If the product quality drops we restrict the product qty taken to increase the reflux ratio. Take more product and the product ABV drops. Towards the end of the run when there is less alcohol available, even increasing the reflux ratio may not return enough alcohol to the column to maintain stability. Eventually the column will fail.
Also remember that the opposite is also true. At high vapour speeds increasing the reflux ratio can lead to the situation where the quality also drops.
The only time you have equal flows in both condensers is when you have a reflux ratio of 1:1, so utilising temperature differences might be of limited value.
I still think that knowing the actual reflux ratio is not going to help you a lot. To be able to apply it to control the column, you would also need to know the vapour composition at both the base and top of the column. If you are using LM then you also need to know the run profile, as the reflux ratio changes constantly throughout the run as the available alcohol is depleted.
You can't select a reflux ratio and set it for the entire run. Even with a VM towards the end of the run you cycle the column between full reflux and partial reflux, and this is determined by the product quality at that moment in time. When it starts to drop the column is put back into full reflux to restabilise, then the reflux ratio is slowly reduced to take a bit more product.
EDIT: 1 other minor point. The rate of coolant flow in both condensers is NOT the same. You should be running your reflux condenser at much lower coolant flow rate than your product condenser. You want the coolant exiting your reflux condenser to be warm, if not hot!!
Re: Instantaneous RR measurement
Looking back at the OP you said the same thing there, and I should have picked up on it sooner. Coolant flow. The coolant flow should be independent between the two condensers. You can run them in series but this is a design flaw. There is nothing wrong with feeding a SAMPLE of the coolant output from your product condenser into the reflux condenser, but there should be a bypass to return the bulk of the coolant back to the reservoir or drain.
The 2 condensers have different purposes. The reflux condenser converts hot vapour into hot liquid. If it cools that liquid then that is a flaw and is just wasting enrergy. Overcooling of the reflux condensate can adversely affect the temperature gradient established withn the column. The coolant output from the reflux condenser should be hot.
The product condenser on the other hand is supposed to convert hot vapour into cool liquid. Not neccesarily cold but definitely below the flash point of that liquid or the vapour evapourating from it. Hot high proof liquid alcohol, exposed to atmosphere, is not something to encourage. The coolant output from the product condenser should be cool.
The 2 condensers have different purposes. The reflux condenser converts hot vapour into hot liquid. If it cools that liquid then that is a flaw and is just wasting enrergy. Overcooling of the reflux condensate can adversely affect the temperature gradient established withn the column. The coolant output from the reflux condenser should be hot.
The product condenser on the other hand is supposed to convert hot vapour into cool liquid. Not neccesarily cold but definitely below the flash point of that liquid or the vapour evapourating from it. Hot high proof liquid alcohol, exposed to atmosphere, is not something to encourage. The coolant output from the product condenser should be cool.
-
- Angel's Share
- Posts: 2228
- Joined: Tue Dec 07, 2010 8:20 am
- Location: Pacific Northwest
Re: Instantaneous RR measurement
No. Reflux ratio refers to the amount of vapor being refluxed (condensed by the reflux condenser and returned to packing or plates) compared to the amount of vapor that is allowed to pass and condensed by the product condenser then taken as product.Did you mean: Reflux ratio is the ratio of vapor being condensed in your reflux condenser to that returned to the packing as liquid to gain additional evaporation cycles.
On a CM still 100% reflux is used initially to establish equilibrium in the column. Once established reflux coolant is reduced allowing some vapor to pass and be condensed as product. (you only take off a percentage of vapor being produced) The amount taken off as compared to the amount of vapor produced. = RR. It can be expressed as a ratio or percentage.MR, how could the reflux ratio as you define it be anything but 100%? If the vapour going into the reflux condenser is not condensed, then it will be vented.
Temperature can give you much valuable information but can not measure RR.
Very few here get the concept I have explained or care to follow the math. I know Myles does. He's a big egg head. RR numbers are really not of much importance unless you are using it for comparative testing to study different designs. I build and test new designs on a regular basis.
Build yuor still. Install temp probes everywhere. Learning what they are telling you will be half the fun. A guy with your brains and background will have a great time here. It can be quite challenging and rewarding.
Welcomw aboard.
MR
Re: Instantaneous RR measurement
I have one of those knobs on my LM column... It's called the take off/reflux ratio needle valve...Maritimer wrote:Imagine if you had a still with a knob labelled Reflux Ratio, right there to set as you pleased. Would you use it? If so, how can you imagine using it?
Do I use it...??? Yes...
How do I use it...??? I can control the reflux ratio to keep the column in equilibrium throughout the run and the take off rate is merely a result of adjusting the reflux ratio...
Do I need to know the exact reflux ratio...??? No because there are too many other variables that come into play over the course of each and every run due to wash composition, ambient weather conditions, the alignment of the stars, and more other variables than worth considering...
I am the process controller and I refuse to surrender control to a plethora of electronic sensors and a programmable controller...
As for what Reflux Ratio means, that was determined years ago and regardless of how someone wants to word it the fundamentals of the ratio remains the same... Spirits Returned : Spirits Collected... Reflux is measured in liquid not vaporous or temperature form and trying to bend those other methods into something worthwhile may very likely simply be fools folly and impossible to master at the hobby scale...
<rant>
It would be nice if engineers would leave their degrees at the login page... Constantly trying to persuade them to throttle back on the thinking and just enjoying themselves gets old and the majority of the membership doesn't give two shits about making home distillation high tech... Leave that thought process for your smart phone or whatever other high tech gadgets you feel the need to play with far too much and turning your brain into overstressed mush... I think hard enough for a living and and everyday life and have less than no desire to think that hard while trying to relax... But that's just me...
</rant>
-
- Angel's Share
- Posts: 2228
- Joined: Tue Dec 07, 2010 8:20 am
- Location: Pacific Northwest
Re: Instantaneous RR measurement
<rant>
It would be nice if engineers would leave their degrees at the login page... Constantly trying to persuade them to throttle back on the thinking and just enjoying themselves gets old and the majority of the membership doesn't give two shits about making home distillation high tech... Leave that thought process for your smart phone or whatever other high tech gadgets you feel the need to play with far too much and turning your brain into overstressed mush... I think hard enough for a living and and everyday life and have less than no desire to think that hard while trying to relax... But that's just me...
</rant>[/quote]
LOL nice rant. Most guys that start out with posts like this never end up distilling a drop.
It would be nice if engineers would leave their degrees at the login page... Constantly trying to persuade them to throttle back on the thinking and just enjoying themselves gets old and the majority of the membership doesn't give two shits about making home distillation high tech... Leave that thought process for your smart phone or whatever other high tech gadgets you feel the need to play with far too much and turning your brain into overstressed mush... I think hard enough for a living and and everyday life and have less than no desire to think that hard while trying to relax... But that's just me...
</rant>[/quote]
LOL nice rant. Most guys that start out with posts like this never end up distilling a drop.
Re: Instantaneous RR measurement
OK, when I first discovered home distilling, I read the parent site obsessively and ordered three books from the Amphora society. I made a VM still using every bit of knowledge I could gather, added a few of my own tweaks, and got 95% alcohol on my first run. The Compleat Distiller is cited everywhere and seems to be the bible of the hobby. So, if we are going to argue definitions, I'll take TCD's as gospel. Here are the definitions from page 44, second edition:
1) The Reflux Ratio is the ratio of the vapor returned to the still as reflux to the total amount of vapor produced. [bold in original]
2) The ratio of the amount withrawn to the total vapor is the product ratio, or "draw down" of the system.
Another book I got from the Amphora Society was Riku's Designing & Building Automatic Stills, maybe not quite as sacred as TCD, but certainly highly respected. His standard practice is to connect the product and reflux condensers in series, see Thor's hammer on page 86, for example. Even TCD, page 199, shows the product and reflux condensers in series. The cold coolant first condenses the product, warming up a little, and then goes to the reflux condenser, where it heats up a lot, coming out hot.
It was this obvious large difference in temperature rises that sparked this discussion. If you have ever made a voltage divider using resistors in series, in which a common current flows through them and the voltage drop across each resistor is proportional to its resistance, an analogy to condensers in series immediately springs to mind. Instead of voltage drops, we have temperature drops. Instead of electric current, we have vapour flow (if the vapour flow doubles, for example, the temperature drops will double). The coolant flow is like a multiplier, setting the vapour-flow to temperature-drop conversion factor.
Again, see page 170 of TCD. Here they show that, since the specific heat of ethanol is only 1/367 of the LHV, cooling the reflux liquid is not much of a problem, in fact, it isn't a problem at all!
M
1) The Reflux Ratio is the ratio of the vapor returned to the still as reflux to the total amount of vapor produced. [bold in original]
2) The ratio of the amount withrawn to the total vapor is the product ratio, or "draw down" of the system.
Another book I got from the Amphora Society was Riku's Designing & Building Automatic Stills, maybe not quite as sacred as TCD, but certainly highly respected. His standard practice is to connect the product and reflux condensers in series, see Thor's hammer on page 86, for example. Even TCD, page 199, shows the product and reflux condensers in series. The cold coolant first condenses the product, warming up a little, and then goes to the reflux condenser, where it heats up a lot, coming out hot.
It was this obvious large difference in temperature rises that sparked this discussion. If you have ever made a voltage divider using resistors in series, in which a common current flows through them and the voltage drop across each resistor is proportional to its resistance, an analogy to condensers in series immediately springs to mind. Instead of voltage drops, we have temperature drops. Instead of electric current, we have vapour flow (if the vapour flow doubles, for example, the temperature drops will double). The coolant flow is like a multiplier, setting the vapour-flow to temperature-drop conversion factor.
Again, see page 170 of TCD. Here they show that, since the specific heat of ethanol is only 1/367 of the LHV, cooling the reflux liquid is not much of a problem, in fact, it isn't a problem at all!
M
Reply to Mash Rookie
Mash Rookie,
Here is a reply to your calculations:
Suppose we run the boiler at 1000 watts = 238.8 cal/sec.
This will be converted to vapour which will be condensed, heating the coolant with this 1000 watts.
If the coolant is flowing at 50 ml/sec = 50 gm/sec for water, it will be heated at the rate of (238.8 cal/sec)/(50 gm/sec) = 4.776 cal/gm.
1 calorie is the heat required to raise 1 gram of water 1*C, so the coolant will rise 4.776*C.
Does the volume of the condenser matter? Suppose we have two condensers, one of 100 ml volume, and the other at 400 ml volume. At 50 ml/sec coolant flow, the 100 ml condenser is flushed in 2 seconds, and the 400 ml is flushed in 8 seconds.
The 100 ml condenser is heating up at 238.8 cal/sec, but it takes 2 seconds to flush the condenser, so the temperature rise is (238.8 cal/sec)/(100 gm/2 sec) = 4.776 cal/gm = 4.776*C for water.
The 400 ml condenser is the same: (238.8 cal/sec)/(400 gm/8 sec) = 4.776 cal/gm = 4.776*C.
So, the temperature rise is proportional to the vapour power, and inversely proportional to the coolant flow rate:
tempDiff = vapourPower/coolantFlowRate [(cal/sec)/(gm/sec) = cal/gm = *C]
And this is proportional to the mass of condensate because the mass vapourized is the boilerPower/LHV: (238.8 cal/sec) / (220.8 cal/gm) = 1.082 gm/sec for pure ethanol, for example.
So,
tempDiff = Constant X rateOfCondensateProduction [the Constant will depend on the vapour composition] With the same flow rate in the two condensers, the ratio of the temperture rises is then proportional to the rate of masses produced by the condensates:
tempDiffOfRefluxCondenser / tempDiffOfProductCondenser = massOfReflux/massOfProduct [the Constants cancel, and the rate i.e., 1/sec cancel]
And ultimately,
RR = (mass of reflux condensate)/(total mass of vapour from boiler)
RR = tempDiffOfRefluxCondenser / (tempDiffOfRefluxCondenser + tempDiffOfProductCondenser )
Hope this convinces you a little. Or please critique my calculations.
M
Here is a reply to your calculations:
Suppose we run the boiler at 1000 watts = 238.8 cal/sec.
This will be converted to vapour which will be condensed, heating the coolant with this 1000 watts.
If the coolant is flowing at 50 ml/sec = 50 gm/sec for water, it will be heated at the rate of (238.8 cal/sec)/(50 gm/sec) = 4.776 cal/gm.
1 calorie is the heat required to raise 1 gram of water 1*C, so the coolant will rise 4.776*C.
Does the volume of the condenser matter? Suppose we have two condensers, one of 100 ml volume, and the other at 400 ml volume. At 50 ml/sec coolant flow, the 100 ml condenser is flushed in 2 seconds, and the 400 ml is flushed in 8 seconds.
The 100 ml condenser is heating up at 238.8 cal/sec, but it takes 2 seconds to flush the condenser, so the temperature rise is (238.8 cal/sec)/(100 gm/2 sec) = 4.776 cal/gm = 4.776*C for water.
The 400 ml condenser is the same: (238.8 cal/sec)/(400 gm/8 sec) = 4.776 cal/gm = 4.776*C.
So, the temperature rise is proportional to the vapour power, and inversely proportional to the coolant flow rate:
tempDiff = vapourPower/coolantFlowRate [(cal/sec)/(gm/sec) = cal/gm = *C]
And this is proportional to the mass of condensate because the mass vapourized is the boilerPower/LHV: (238.8 cal/sec) / (220.8 cal/gm) = 1.082 gm/sec for pure ethanol, for example.
So,
tempDiff = Constant X rateOfCondensateProduction [the Constant will depend on the vapour composition] With the same flow rate in the two condensers, the ratio of the temperture rises is then proportional to the rate of masses produced by the condensates:
tempDiffOfRefluxCondenser / tempDiffOfProductCondenser = massOfReflux/massOfProduct [the Constants cancel, and the rate i.e., 1/sec cancel]
And ultimately,
RR = (mass of reflux condensate)/(total mass of vapour from boiler)
RR = tempDiffOfRefluxCondenser / (tempDiffOfRefluxCondenser + tempDiffOfProductCondenser )
Hope this convinces you a little. Or please critique my calculations.
M
Re: Instantaneous RR measurement
You are not accounting for heat loss or gain due to ambient outside air (temp and/or wind chill), column insulation & [the Constant will depend on the vapour composition] vapor composition will not remain constant throughout the entire system(i.e. stacking the column) & even less so throughout the entire run, that's how cuts are made. As an example, your reflux condenser will knock down the mixture differently as it changes throughout the run. In my flute I can equalize the column and provide just enough additional heat to pass the foreshots but not the heads/hearts, only removing a few hundred ml of product, already the composition has changed enough to modify the amount of heat required to pass the reflux condenser at a constant rate.
Just as a thought, IMO nothing is going to be instantaneous & exact in a system of constant flowing everything: water temp, water flow rate, vapor temp, vapor flow & vapor composition can all change depending on several factors. Hell, if you flush your toilet you probably lose water pressure a little and throw the whole calculation into a tailspin. You will not be able to monitor anything like this instantaneously and exactly. What you are going to get is more of an ever flowing / changing system that you can monitor for a range of values that fit a range of good or bad figures for your specific system. Every system acts differently and every wash is composed differently. No two washes are going to have the exact same amount of fores/heads/hearts/tails.
I agree with MR on getting geeky with it but if you are having fun doing it then more power to you. It is possible doing this kind of research could lead to the next big thing so I support it whole heartedly.
Just as a thought, IMO nothing is going to be instantaneous & exact in a system of constant flowing everything: water temp, water flow rate, vapor temp, vapor flow & vapor composition can all change depending on several factors. Hell, if you flush your toilet you probably lose water pressure a little and throw the whole calculation into a tailspin. You will not be able to monitor anything like this instantaneously and exactly. What you are going to get is more of an ever flowing / changing system that you can monitor for a range of values that fit a range of good or bad figures for your specific system. Every system acts differently and every wash is composed differently. No two washes are going to have the exact same amount of fores/heads/hearts/tails.
I agree with MR on getting geeky with it but if you are having fun doing it then more power to you. It is possible doing this kind of research could lead to the next big thing so I support it whole heartedly.
Current Evolution:
MrDistiller > 2" potstill > copper 4" perf 4 plate flute
"I seal the lid with Silly Putty, that's OK ain't it ?"
~ kekedog13
"Attach a vibrator to it and hang it upside down. Let it work"
~Mr. P
MrDistiller > 2" potstill > copper 4" perf 4 plate flute
"I seal the lid with Silly Putty, that's OK ain't it ?"
~ kekedog13
"Attach a vibrator to it and hang it upside down. Let it work"
~Mr. P
Re: Instantaneous RR measurement
Hi Durace11,
I appreciate that you understand that I'm just having fun. There sure are a lot of crankypants on this forum!
Of course there are other factors to consider; the calculations are ideal.
If you follow my calculations, you can see that the variations in LHV as the ethanol is extracted, flow rate changes, and static coolant temperature will cancel out.
My still is encased in 2" blue styrofoam. R-value of 10.
I'll insulate both condensers, too.
As for instantaneous, that is also ideal. There will be lags as the surrounding metal settles to a steady temperature.
Sunday or Tuesday will probably be the test day. We shall see.
M
I appreciate that you understand that I'm just having fun. There sure are a lot of crankypants on this forum!
Of course there are other factors to consider; the calculations are ideal.
If you follow my calculations, you can see that the variations in LHV as the ethanol is extracted, flow rate changes, and static coolant temperature will cancel out.
My still is encased in 2" blue styrofoam. R-value of 10.
I'll insulate both condensers, too.
As for instantaneous, that is also ideal. There will be lags as the surrounding metal settles to a steady temperature.
Sunday or Tuesday will probably be the test day. We shall see.
M
-
- Swill Maker
- Posts: 315
- Joined: Fri Sep 20, 2013 3:00 pm
Re: Instantaneous RR measurement
I've been thinking about this.Maritimer wrote:
Any cooling of the condensate would cause a small error, but since the specific heat of the liquid is so much tinier than the latent heat of vaporisation, it can probably be ignored.
M
Any cooling of the condensate in the reflux condenser is just going to further condense the vapour coming up. This means there is effectively zero sub-cooling at that point, only condensing.
Your condenser is just working remotely further down the column.
I'm thinking of doing something similar with a CM flute style set up.
I could use one of these at the inlet and outlet of each condenser and another at the product outlet to maintain 25°C and drive both condensers separately via two variable speed PD pumps.
That way I could dial in from 0% to 100% reflux consistently regardless of power in (could work for propane too) and be using the absolute minimum amount of water all the time.
I could get the flow rate from the pump speed, but do I need it
If I wanted to maintain a specific still head temp in neutral mode then that could also be easily dialed too in with a bit of extra programing but I'd still display the RR being used to achieve it.
Re: Instantaneous RR measurement
Hi JB,
Here are my calculations for the RR and considerations of condensate cooling.
This one takes into account condensate cooling to measure product rate. And this one uses the product rates for the reflux and product condensers. At the end, in assuming a difference of 20*C in the product and reflux condensates, the error amounts to 0.7%. To use this technique in a CM still, it is necessary to arrange the formula so that everything cancels out except for the coolant temperature differences. I fear that using a straight series connection of the condensers would produce a very hot product and maybe let some vapour escape. That might not be true, if you have run your condensers successfully in series, then the same equation can be used.
Edit: An alternative would be to run the coolant in series through the product and reflux condensers and use the temperature differences of the coolant to calculate the RR, and then add another condenser in series with the product condenser to finalize the condensation and cooling of the product.
But if additional product condenser coolant flow is needed, then it will have to be a multiple of the reflux condenser flow.
Let's say that FRpc = a x FRrc
where FRpc is the flow rate of the product condenser, etc.
Then,
RR = deltaTrc/(deltaTrc + a x deltaTpc)
for instance, if the FRpc = 2 x FRrc, then the delataTpc would be 1/2 of what it would be if the flow rates were the same. So multiply by 2 to bring it back. Of course, this will cause some product cooling, which could be measured and the deltaTpc compensated.
That digital temperature sensor looks impressive, but note that the spec is +/-0.5*C. It's more wiring, for sure, but I'm sticking to the LM35 at +/-0.2*C. Also, I can take differences in analoque and tweak any error.
M
Here are my calculations for the RR and considerations of condensate cooling.
This one takes into account condensate cooling to measure product rate. And this one uses the product rates for the reflux and product condensers. At the end, in assuming a difference of 20*C in the product and reflux condensates, the error amounts to 0.7%. To use this technique in a CM still, it is necessary to arrange the formula so that everything cancels out except for the coolant temperature differences. I fear that using a straight series connection of the condensers would produce a very hot product and maybe let some vapour escape. That might not be true, if you have run your condensers successfully in series, then the same equation can be used.
Edit: An alternative would be to run the coolant in series through the product and reflux condensers and use the temperature differences of the coolant to calculate the RR, and then add another condenser in series with the product condenser to finalize the condensation and cooling of the product.
But if additional product condenser coolant flow is needed, then it will have to be a multiple of the reflux condenser flow.
Let's say that FRpc = a x FRrc
where FRpc is the flow rate of the product condenser, etc.
Then,
RR = deltaTrc/(deltaTrc + a x deltaTpc)
for instance, if the FRpc = 2 x FRrc, then the delataTpc would be 1/2 of what it would be if the flow rates were the same. So multiply by 2 to bring it back. Of course, this will cause some product cooling, which could be measured and the deltaTpc compensated.
That digital temperature sensor looks impressive, but note that the spec is +/-0.5*C. It's more wiring, for sure, but I'm sticking to the LM35 at +/-0.2*C. Also, I can take differences in analoque and tweak any error.
M
-
- Novice
- Posts: 13
- Joined: Wed Jun 25, 2014 9:36 am
Re: Instantaneous RR measurement
Im confused here, I am fascinated by this thread, With school the way it is right now I don't have the time to fully understand and research the points mentioned as to make a valid contribution, but the fact that its a complex enough of a question to require the time is what makes it so appealing to me.rad14701 wrote: .....
<rant>
It would be nice if engineers would leave their degrees at the login page... Constantly trying to persuade them to throttle back on the thinking and just enjoying themselves gets old and the majority of the membership doesn't give two shits about making home distillation high tech... Leave that thought process for your smart phone or whatever other high tech gadgets you feel the need to play with far too much and turning your brain into overstressed mush... I think hard enough for a living and and everyday life and have less than no desire to think that hard while trying to relax... But that's just me...
</rant>
I guess what i'm asking is If this isn't an area of interest for you, why partake in the discussion. The threads is what it is because of shared interest. your implying it bothers you, why let it?
(im not judging, I don't know your situation and I have alot of respect for easy going philosophies, I was just wondering about it)
thank you all for the mental stimulation
Re: Instantaneous RR measurement
This is what I call a rad-trap. Rad does this to newbies. The required response is anger, to which he replies in a frustrating, "What, I'm only giving my opinion, what are you so mad about?" I had observed him doing this before I started posting, so when it came, I didn't respond. However, I do like to point out when he does this and to advise anyone who is taunted not to respond. He knows this game too well for a newbie to win.
M
M
Last edited by Maritimer on Fri Jul 04, 2014 6:09 am, edited 1 time in total.
-
- Novice
- Posts: 13
- Joined: Wed Jun 25, 2014 9:36 am
Re: Instantaneous RR measurement
thats funny, good looking out
Re: Instantaneous RR measurement
Instantaneous RR measurement would be nice. It will give another set of readings to look at alongside the many other things to watch. Opening the needle valve wide open during the run and comparing the product output before and after will give the numbers, but so doing will de-stabilize the column and it must be re-stabilzed after each reading.
Scientist- analysis of numbers leads to guaranteed repeatability, easily verified by repeating the numbers.
Artist- analysis of feelings leads to intuition, intuition leads to great art which is extremely hard to duplicate.
They feed upon each other, let them unite in heavenly union...close the curtains!
Scientist- analysis of numbers leads to guaranteed repeatability, easily verified by repeating the numbers.
Artist- analysis of feelings leads to intuition, intuition leads to great art which is extremely hard to duplicate.
They feed upon each other, let them unite in heavenly union...close the curtains!
cornflakes...stripped and refluxed
Re: Instantaneous RR measurement
I agree completely. However, my only interest in distilling is to make good neutral. I know the sweet spot for each stage of the process and use these settings.Ayay wrote:Scientist- analysis of numbers leads to guaranteed repeatability, easily verified by repeating the numbers.
Artist- analysis of feelings leads to intuition, intuition leads to great art which is extremely hard to duplicate.
However, it might be possible to venture into new territory, call it algorithmic distilling, where new regimes could lead to new flavours. Once new tools are available, there is the possibility new discoveries.
M