The mechanics of the Reflux Column

[Andrew_90]

So this is my understanding of the process. In simple terms the reflux column is about reprocessing an ethanol "molecule", up and down a column, heating and cooling it and allowing the molecule allowing it to shed the unwanted wash attached to it. When the ethanol molecule is light enough (shed enough unwanted stuff) it will migrate to the PC to be liquified again.

In my reasoning, bubbles that appear at the top of column packing is a function of the ascending vapor breaking through a the wall of liquid product resting on top of the packing. If the product did not fill the top of the packing then the vapor would simply rise up to the RC with no visible signs. If there is a layer of product on the top of the packing then there is a pressure beneath it forcing the vapor to escape.

If this is this is the case then forcing the vapor through a wall of product will cross contaminate the vapor and the product.

Surely it would be more expedient to have the ascending vapor have minimal contact with descending product. Why would we not want a parallel flow the vapor past the product?

Currently I try maintain 0.5" of bubbling activity on top of my packing. Why would I not want to back off on the power and have pure vapor ascending without have to break through the wall of product, and with the reduced power have the condensate flow back down right past the ascending vapor? I understand that this would reduce output but I am trying to understand the theory of the flow.

With the layer of product on top of the packing how much product never gets the chance to flow back down the column?

[Setsumi]

What you describe is the fluidized column. Originally VM and CM columns were not run this way. It became popular with SPP. There are currently a discussion active on the issue. And Odin noted it well in his original thread. Not all run their VM's this way.

[bluefish_dist]

The one thing you are missing is heat transfer. As the cold liquid falls, it absorbs heat from the raising vapor. This cooling/heating creates a temperature gradient from top to bottom with the top being cold and the bottom being warmer. You don’t need to bubble vapor through a liquid for this to happen. It can happen simply by mixing the raising/falling streams. This is why we have packing. It mixes the raising vapor with the falling liquid so there is time for the heat transfer to occur.

[kimbodious]

@Bluefish - exactly! There’s phase changes happening all through the packing where the falling condensate is heated to boiling (vaporising) again by the rising vapour. The greater this mingling of vapour and condensate, the greater the number of times of re-boiling. This is where the amount and quality of packing comes in to play OR the number of plates.

[Yummyrum]

Andrew ,I think you have realised whats happening .
It’s nothing to do with all this talk of fluid stuff at the top .
And everything to do with what happens in the other 30 or 40” of packing below it . That is where the real stuff happens

Edit: Posted with Kimbo

[NormandieStill]

The one thing you are missing is heat transfer. As the cold liquid falls, it absorbs heat from the raising vapor. This cooling/heating creates a temperature gradient from top to bottom with the top being cold and the bottom being warmer. You don’t need to bubble vapor through a liquid for this to happen. It can happen simply by mixing the raising/falling streams. This is why we have packing. It mixes the raising vapor with the falling liquid so there is time for the heat transfer to occur.

As I understand it then, the lower the thermal mass of the packing, the more this happens as a direct consequence of the vapour / liquid interaction, and the more reactive (less stable) the column. Inversely packing with high thermal mass will take longer to stabilise, but the revaporisation of descending liquid will be more independent of the rising vapour. I could imagine that a "perfectly reactive" column would have no packing, but the descending liquid would be aerosolised allowing for the vapour / liquid interaction required. And conversely a perfectly stable column would have packing which was externally maintained at the required temperature gradient so completely independent of the heat input from the vapour.

Andrew_90. What you're trying to achieve is complete separation of the various components in your wash by temperature. This would result in a column with a temperature gradient from top to bottom which would correspond to the boiling points of the various components in your wash. In reality, this is an approximate process and wash is being continuously separated into its constituent parts. By taking off product at a slow enough rate, the separation in the column persists and you end up just removing the lightest components.

[jessespa]

I think Andrew is trying to figure out why he was deleted from the other thread he was in earlier.
From what I was reading before it was deleted someone thought there was a language barrier or something that I wasn't picking up on.
I did read that you are using rashing rings as your packing. I too use the 1/4"or3/16" rashing rings ( not sure how they measure them) when using my column I never jumped on the SPP wagon.
I'm with Yummy, I don't think fluid on top is all that important but the rest of the column full of packing. Normally I might just see bubbles above my packing and not much more than that.

[bluefish_dist]

As I understand it, we do separate constituents by boiling point, but since we are batch distilling, those constituents are continually changing. In a perfect still we would pull all the lightest components, then the next lightest, until only the heavy components are left. But our stills are not perfect, so we get a mix of components at the top. We still get a change of components vs time as we deplete the lightest molecular weight and move up in weights during the run.

If you run a continuous still, you can pull each component by height. Batch stilling changes that to pulling by time and in order of molecular weight. Thus the need to collect in jars and do cuts. It’s simply selecting your components vs time you want to keep. The better a still performers, the more pure each component. A pot still would be the worst for separating components, but imho whiskey is not all about individual components, but a mix.

[bluefish_dist]

You may be correct on more thermal mass holding heat to do the work, but I expect in our stills the vapor has a far higher heat capacity that the packing. I think less thermal mass would be preferable so the still reacts more quickly to the change in components and boiling points.

[Demy]

I do not have a high technical knowledge, but I think it depends on the type of packaging, I use scrubber and I am not a fan of the "liquid bed", I think it is a way to push the column to the limit ... but from what I have read for who uses spp is something to look for to get the most out of this packaging.

[still_stirrin]

Bluefish & others have pointed out the physics involved.

Heat exchange along the column length creates the condensing and reboiling of the vapor and that causes more volatile (lower boiling point) fluids to accumulate higher and higher while the higher boiling point fluids (water, for example) to continue to fall back down the column. The packing provides a surface on which the heat exchange occurs and the rising (hot) vapors give up some of their heat (not temperature, but heat which is a form of energy) to the condensate.

Reflux is (by definition) the process of condensing and reboiling over and over. It is the mechanism that separates the volatiles from non-volatiles. I don’t believe there is an advantage to a fluidized (flooded) column UNLESS the column is a flute. Those hold the condensate at discrete stages (plates). So, keeping the liquid high in the column is necessary for reflux in a flute.

Andrew, do you have better “performance” when you run a flooded column? I guess it would represent the upper limit of “throughput”, but may not actually give you the highest purity of product.
ss

[Andrew_90]

I think Andrew is trying to figure out why he was deleted from the other thread he was in earlier.
From what I was reading before it was deleted someone thought there was a language barrier or something that I wasn't picking up on.
I did read that you are using rashing rings as your packing. I too use the 1/4"or3/16" rashing rings ( not sure how they measure them) when using my column I never jumped on the SPP wagon.
I'm with Yummy, I don't think fluid on top is all that important but the rest of the column full of packing. Normally I might just see bubbles above my packing and not much more than that.

I voluntarily created a new post here as I realized my post was actually a hijack of the other post. My English is spot on, however at times is it difficult to articulate concepts within a couple a lines.

My rings are 6mm and the length is the same as the diameter.

[Andrew_90]

Andrew, do you have better “performance” when you run a flooded column? I guess it would represent the upper limit of “throughput”, but may not actually give you the highest purity of product.
ss

So before I answer let me state what the thought process around my column was, and to test if my thinking was correct. My column criteria:
a. That it was tall enough to meet a 1:20 ratio of diameter to length.
b. That the packing was as dense as I could get for maximum heat retention and therefore minimum change in column temperature. For this reason I ignored copper mesh, and thinner walled media and opted for 6mm Ceramic Raschig Rings which have a higher thermal mass. I have a very small section of copper mesh due to the undersupply of rings, circa 4" of the column which is located at the base of the column, which also serves to retains the Raschig Rings.
c. To further improve the stability of the column and to hold temperature, I lagged the column with reasonably tightly packed Ceramic Wool Blanket of 1" radial wall thickness with a PVC sheath and endcaps to seal the Ceramic Wool.

The point of this was try remove as many process variables as possible so that I could discount column performance from any product quality issues I may experience. So I do believe I have achieved this and have as efficient a 2" column as I could have hoped for.

Back to SS.
So despite googling I am still a little unsure as to what a flooded column is. Would this be a column that is saturated with liquid with little to no air gaps throughout the entire column?

I assumed that as there was bubbling and liquid on top of my packing (0.5 - 1.0") that the column was flooded. I further assumed that as there was a temperature gradient up the column that the "flooding" would have been more severe the further one travelled down the column. From this I then deduced that there would have been too much interference in the up and down paths to effectively promote good reflux and that perhaps backing off a little on the power ma well have helped in improving reflux action and better fraction stacking.

Someone posted that the top and bottom of the column tends to be flooded but not the majority of the column. This completely blows away any assumption I may have improved the working of a column.

So SS I get 95/96% ABV from my column controlling the bubbling on top of the packing to between 0.5 and 1.0" in height. This from a 50l keg charged with Shady's Sugar Shine stripped to low wines and diluted to 38%. Spirit take off of 1.2 to 1.5l per hour (34 - 51 fl oz.). Power required to run at this offtake is around 1 740W. My RC is very efficient and is a 10" long double helix, the distillate streams down the sides of the sight glass below the T Piece, sometimes I think the volume of the condensed product is too high but have erred on the side of max knockdown. I thought that the descending distillate was flooding the top of the packing .

Rather longwinded.

[Saltbush Bill]

As the cold liquid falls, it absorbs heat from the raising vapor.

And that is why we adjust the water flow to a reflux condenser to a point where the water leaving it is quite hot/warm.
Cooling the falling distillate to much just makes the job a whole lot harder for the still, packing and the rising vapors.

[Andrew_90]

As the cold liquid falls, it absorbs heat from the raising vapor.

And that is why we adjust the water flow to a reflux condenser to a point where the water leaving it is quite hot/warm.
Cooling the falling distillate to much just makes the job a whole lot harder for the still, packing and the rising vapors.

Noted, I clearly run mine way too cool.

[Saltbush Bill]

Andrew the vapour needs to be cooled enough that none is escaping the top of the column "all is turned back to liquid".......but any more than that is unnecessary.

[Andrew_90]

Thanks.

Busy relocating to a new home so doubt if I will be up and running in the next 2 to 3 months. Will apply the knowledge then.

[StillerBoy]

The point of this was try remove as many process variables as possible so that I could discount column performance from any product quality issues I may experience. So I do believe I have achieved this and have as efficient a 2" column as I could have hoped for.

It would seem that there is an expectation the the reflux column will provide the means for a "product quality".. unfortunately, a reflux column or any still it not able to do that..

A still is just a water/alcohol separator, and the "product quality" comes making a wash/mash of high quality, fermenting it properly, and making the right cuts.. the operational variables in the process of operating a unit, within reason, will not make a great difference in the end product or quality.. understanding the operational variables only make the operation of a unit easier and enjoyable on one's self..

The efficiency of a unit comes from many factors, the unit design itself, the column length and size, the packing used, the ability to manage/control the power efficiently, the water temp and flow used in feeding the condenser, the take off and it effects, and the operator understanding of the unit's behavior.. as to the insulation of the column, in a close environment, such as inside a room with normal ambient temp, insulation of the column is really not a requirement, other than some temp reduction to the ambient temp of the room, but a unit operated outdoor, yes, it is a requirement but it does not require to be to the extent you have stated..

As you had stated, there will be a break from distilling for a few months, take the time to fully comprehend what has been stated in this thread, and the other thread on fluid level above the package, as there has been posted many valuable info.. you are young at this hobby, and it will take a few more yrs to get the full understanding of all the variables that takes place while operating a unit, so as to be able to see the subtle changes that occur during operation, and how they are intertwined, and the effect they provide..

Mars

[drmiller100]

There is great stuff in this thread. I think about it differently and maybe my paradigm will help someone.

When making azeo from less than 10 percent with reflux I think of it as temperature.

Heat rises. The wash is boiling at like 200 degrees. Azeo, by definition, is around 172 depending on altitude and barometric pressure.
So the hot vapors come off the boiler and head up the column.
The goal is to "cool off" the vapors. If you reflux all the vapor the alcohol will tend to condense and concentrate at the top. I run LM. If I overcool my liquid down to like 160 it doesn't matter. The liquid goes back down the packing and the top of the column ends up at 172 and I have azeo.

When I start pulling product there is less reflux, so top of column cooling suffers. If I pull to much product the temperature rises and I know I'm no longer getting azeo.