Height for purity, diameter for speed, and HETP Calcs

Distillation methods and improvements.

Moderator: Site Moderator

Post Reply
User avatar
Odin
Site Donor
Site Donor
Posts: 6844
Joined: Wed Nov 10, 2010 10:20 am
Location: Three feet below sea level

Height for purity, diameter for speed, and HETP Calcs

Post by Odin »

This is the way we traditionally look at packed columns and their performance. If you want (need) more purity, make the column higher. If you want more liters per hour pushed out, you need a wider diameter column. Of course you need to fill that column with more gases and reflux, so a higher energy input is neede, but simply put: height for purity, diameter for speed.

When I make new column packing, I always want to know it's performance. Part of that performance is to calculate HETP: Height Equivalent of a Theoretical Plate. It tells us how many centimeters of packing equals one distillation.

The procedure I use is a sort of industry standard, that goes like this:
- Put 10% alcohol in the boiler;
- Add enough plates or packing to be able to take that 10% to 96.5%.

To take the 10% low wines or wash to 96.5% needs 13 distillation cycles. But we do not want to just test by top of column temperatures. Instead, we want to take product out for additional testing via a hydrometer. So it's not a closed system. And since we draw off, we need a 14th plate (the one we draw off) and because the action of drawing off liquids creates turbulence, we need one more additional plate or redistillation. 13 plus 1 plus 1 equals 15. Fifteen redistillations to take us from 10% to 96.5%.

This is not my method, this is how companies like Hollstein go about their business. I just apply the procedure to my columns to be able to compare results in, well, a comparable way.

All right, let's say that I take my 3 inch diameter column, that's 125 centimeters high, and pack it with 100 centimeters of my new column packing. I run it and create 96.5% with it. 96.5% means I got at least 15 redistillations (see above). Next thing I do is take out packing, so there is only 75 centimeters of packing in. If I can take the output to 96.5%, this means again 15 redistillations took place. Or more. A first calculation (75:15) teaches that the HETP is no more than 5 centimeters.

Next thing I do, is take out my 50 centimeter column section and fill it with 45 centimeters. All right, we can still draw 96.5%, so we can now calculate that HETP is actually much lower: no more than 3 centimeters.

Testing some more teaches me that I can still draw off 96.5% with just 30 centimeters of packing, but no less. Devide 30 by 15 and HETP is 2 centimeters.

Now here's the funny thing. The less packing I use (the shorter I stack my column), the more reflux I need to maintain 96.5%. The more reflux I need, the less product I can assemble.

There you have it: column height is not just related to purity, but also to speed. If I fill my 3-incher with 125 centimeters of packing, production speed goes up tremendously.

Regards, Odin.
"Great art is created only through diligent and painstaking effort to perfect and polish oneself." by Buddhist filosofer Daisaku Ikeda.
User avatar
Kareltje
Distiller
Posts: 2176
Joined: Tue Feb 09, 2016 4:29 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by Kareltje »

Yes. I am not surprised.
In Dutch: het moet uit de lengte of de breedte.
What goes in, comes out.

A pot still is simple: the vapour is taken aside and condensed. Standard diagram of temperature, %ABV liquid and %ABV vapour.

In other stills there is a path or room where the vapour condenses and partly flows back in the fluid or in a halfway fluid/vapour-equilibrium. The condens is according to the temperature, so lower than the vapour. So the vapour is enriched.
The more times this happens, the more the vapour will be enriched.
So a large room, a long pipe or a lot of objects will enhace this process.

But I suppose there are limits on each of this aspects.
User avatar
still_stirrin
Master of Distillation
Posts: 10337
Joined: Tue Mar 18, 2014 7:01 am
Location: where the buffalo roam, and the deer & antelope play

Re: Height for purity, diameter for speed, and HETP Calcs

Post by still_stirrin »

Interesting concept, analysis, and demonstration Odin. Indeed, your logic seems reasonable...in concept.

And I'm confident that a major part of your demonstration is the ability to control the processes, including the low wines %ABV, the heat input, vapor temperature measurement at the different tower heights, and packing density, which would all seem to be key data inputs in support of your deduction.

This thread is very helpful to understand the HETP relationship, especially in light of the "colorful" discussion on another thread regarding SPP and its effect on the HETP.

Whether intended or not, I am sure that this thread will spur controversy with some. But it brings to light the relationship between diameter, height, packing, and heat input on production speed and purity.

Thanks.
ss
My LM/VM & Potstill: My build thread
My Cadco hotplate modification thread: Hotplate Build
My stock pot gin still: stock pot potstill
My 5-grain Bourbon recipe: Special K
User avatar
shadylane
Master of Distillation
Posts: 10363
Joined: Sat Oct 27, 2007 11:54 pm
Location: Hiding In the Boiler room of the Insane asylum

Re: Height for purity, diameter for speed, and HETP Calcs

Post by shadylane »

Odin wrote:....Now here's the funny thing. The less packing I use (the shorter I stack my column), the more reflux I need to maintain 96.5%. The more reflux I need, the less product I can assemble.

There you have it: column height is not just related to purity, but also to speed. If I fill my 3-incher with 125 centimeters of packing, production speed goes up tremendously.

Regards, Odin.
So long story short :lol:

"Height for purity" is the easiest way to get the job done,
But even a short column can produce 96.5% if you drive it really sllooowwww.
User avatar
Odin
Site Donor
Site Donor
Posts: 6844
Joined: Wed Nov 10, 2010 10:20 am
Location: Three feet below sea level

Re: Height for purity, diameter for speed, and HETP Calcs

Post by Odin »

still_stirrin wrote:Interesting concept, analysis, and demonstration Odin. Indeed, your logic seems reasonable...in concept.

And I'm confident that a major part of your demonstration is the ability to control the processes, including the low wines %ABV, the heat input, vapor temperature measurement at the different tower heights, and packing density, which would all seem to be key data inputs in support of your deduction.

This thread is very helpful to understand the HETP relationship, especially in light of the "colorful" discussion on another thread regarding SPP and its effect on the HETP.

Whether intended or not, I am sure that this thread will spur controversy with some. But it brings to light the relationship between diameter, height, packing, and heat input on production speed and purity.

Thanks.
ss
I hope this post does not create any controversy, SS. I do hope people find it helpfull that, if they want to produce more, one decision is to go wider, but they can also decide to go higher with their column builds. Now in the old times with scrubbers and 3 drops per second collection at low wattages, that was not really an option. If you already have a 1.5 meter column to get to 96.5, .... but with the newer packings and higher energy inputs most people nowadays run their packed columns, the decision to go 90 centimeters or 125 centimeters can be like not important, but it does pack a punch in performance.

Still Stirrin, I do these tests with manual needle valves as well. What you need is a steady power input, that's really all that's to it. And a good probe at the top of your column to tell you what's going on.

Shady, amazing summary. To the point and thanks for that.

Regards, Odin.
"Great art is created only through diligent and painstaking effort to perfect and polish oneself." by Buddhist filosofer Daisaku Ikeda.
User avatar
fizzix
Site Donor
Site Donor
Posts: 3698
Joined: Tue Dec 05, 2017 4:08 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by fizzix »

Odin wrote: ...To take the 10% low wines or wash to 96.5% needs 13 distillation cycles..
Help my monkey brain with how 13 cycles was determined. Or can someone point me in the direction to teach myself, please?
(Thank you Odin and zapata for paving the path to this thread.)
User avatar
Single Malt Yinzer
Trainee
Posts: 973
Joined: Fri Sep 23, 2016 3:20 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by Single Malt Yinzer »

Figure 5: https://www.extension.purdue.edu/extmed ... E-117.html" onclick="window.open(this.href);return false;" rel="nofollow
HETP.jpg
Calcs here: https://en.wikipedia.org/wiki/Theoretical_plate" onclick="window.open(this.href);return false;" rel="nofollow
User avatar
fizzix
Site Donor
Site Donor
Posts: 3698
Joined: Tue Dec 05, 2017 4:08 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by fizzix »

I see now! Each "stair step" in your diagram above is equivalent to a simple distillation, or plate, and your graph has 13 stair steps for alcohol.
Hopefully this interpretation is correct. And it appears to answer my question.

Great reading links and graph. Appreciated.
User avatar
fizzix
Site Donor
Site Donor
Posts: 3698
Joined: Tue Dec 05, 2017 4:08 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by fizzix »

BUMP!
This thread is a gem and I wanted to advertise it one more time. Enlightening... slowly, over days.
User avatar
Odin
Site Donor
Site Donor
Posts: 6844
Joined: Wed Nov 10, 2010 10:20 am
Location: Three feet below sea level

Re: Height for purity, diameter for speed, and HETP Calcs

Post by Odin »

Yeah, your interpretation is correct.

It takes 13 distillation cycles to bring a normal wash to 95/96%. But in reality you need 15. One to draw off. One to compensate for turbulence caused by the draw off.

So in theory 13 redistillations to reach azeo, but in practice 15.

But that's "just" azeo. It turns out that if you do more redistillations than 15, you get better heads and tails separations. Until 40. So for vodka 40 redistillations is ideal, because it minimizes heads and tails smearing, resulting in an eazy 80% loss in tastes (that are heads and tails associated).

Regards, Odin.
"Great art is created only through diligent and painstaking effort to perfect and polish oneself." by Buddhist filosofer Daisaku Ikeda.
User avatar
fizzix
Site Donor
Site Donor
Posts: 3698
Joined: Tue Dec 05, 2017 4:08 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by fizzix »

YES! Picked that up about 15! That's the beauty of this thread. Thanks Odin.
(Now for what you just added....)
User avatar
bluefish_dist
Distiller
Posts: 1502
Joined: Fri Jan 16, 2015 10:13 am
Location: Eastern Ia

Re: Height for purity, diameter for speed, and HETP Calcs

Post by bluefish_dist »

shadylane wrote:
Odin wrote:....Now here's the funny thing. The less packing I use (the shorter I stack my column), the more reflux I need to maintain 96.5%. The more reflux I need, the less product I can assemble.

There you have it: column height is not just related to purity, but also to speed. If I fill my 3-incher with 125 centimeters of packing, production speed goes up tremendously.

Regards, Odin.
So long story short :lol:

"Height for purity" is the easiest way to get the job done,
But even a short column can produce 96.5% if you drive it really sllooowwww.
I would disagree that every column can produce azeo even if driven slow. Once you get something that can do 15 equivalent plates, then yes, I agree. I have run 5 plates and could not get there.
I have seen how running taller is faster. Let the height work for you instead of making lots of reflux do the work. I would really love to try some of odins packing to see if I could get close to 40 plates. I figure I have 24 now and it works better than it was as 19.
Formerly
Dsp-CO-20051
User avatar
Odin
Site Donor
Site Donor
Posts: 6844
Joined: Wed Nov 10, 2010 10:20 am
Location: Three feet below sea level

Re: Height for purity, diameter for speed, and HETP Calcs

Post by Odin »

Yeah, like I state above. Shorter column can get you there (but only if it can still help you get those 13+ redistillations), but slower. Basically, you need to run a shorter column faster for more gas creation. More gas for more reflux for more purity. But with a taller column - yes - let the column do the work!

Odin.
"Great art is created only through diligent and painstaking effort to perfect and polish oneself." by Buddhist filosofer Daisaku Ikeda.
User avatar
jonnys_spirit
Site Donor
Site Donor
Posts: 3629
Joined: Wed Oct 21, 2015 7:58 am
Location: The Milky Way

Re: Height for purity, diameter for speed, and HETP Calcs

Post by jonnys_spirit »

Thank you for this information. I am planning a 3" CCVM column and was considering a 36" packed section as "good enough" but now I'll go a bit higher and spend the $$ so I can simply run it faster and get mo'betta output sooner :)

Cheers!
-j
————
i prefer my mash shaken, not stirred
————
User avatar
der wo
Master of Distillation
Posts: 3817
Joined: Mon Apr 13, 2015 2:40 am
Location: Rote Flora, Hamburg

Re: Height for purity, diameter for speed, and HETP Calcs

Post by der wo »

There is nothing to criticize what Odin wrote. But perhaps a bit more details about the collection rate could be helpful. How much exactly influences the number of virtual plates the collection rate (the speed) at a given purity? How much time do you save with a longer column?

Say we have a packed column still with 10 virtual plates. So according to http://homedistiller.org/theory/theory/strong" onclick="window.open(this.href);return false;" rel="nofollow (other calcs may differ a bit), when we have 10% in the boiler, after 10 plates or distillations we have:
55, 81, 87.5, 90.2, 91.9, 93, 93.9, 94.6, 95.1, 95.6%.

But a big difference between 10 real potstill distillations and 10 virtual plates with reflux condenser is, that with a potstill you would get the 95.6% at FULL output, but with a 10-plates-reflux still at ZERO output. And what happens when you start to collect? You could describe it like that you loose virtual plates. When you choose a reflux ratio of 1:1, you will loose one half of the plates. What means you have only 5 plates anymore, what means you get only 91.9% (see the numbers above).
Now say you want 95.1%. Then you need 9 plates. This means with your 10 plates still you are only allowed to loose 1 of the plates. This means you are able to collect only 1/10 (you have to send back into the column 9/10).
But if your still would have 18 plates, you could collect 9/18 (=5/10) to keep 9 plates "active". This means 5 times more product at the same purity! 5 times faster! Without feeding the pot with more energy! What means during the collection phase 5 times less energy and time costs for the same product!

And what does the column length mean for using a reflux still for aromatic spirits? It means, that the more virtual plates your still has the more and sudden will the purity rise when you start to send something back into the column. Small portions of reflux lead to very high abv (tailsless) products. The height alone doesen't rise the purity, but the influence of reflux is higher the more plates you have.
Say you have 10% in the boiler and you want to collect 87.5%. Then you need 2 plates (see numbers above). With your 10 plates still collect 8/10, send back 2/10. But with a 20 plates still collect 18/20 and send back 2/20.
Comparing the collection speed 8/10 vs 18/20, there is not much difference between 10 or 20 plates here: The 20 plates still collects 9l in the same time when the 10 plates still collects 8l, that's all. The extra effort of doubling the height of packing is relatively useless for aromatic spirits. More speed is only possible with a wider column and more wattage (double the diameter and you can use 4 times the wattage and you get 4 times faster product).

So in a way the rule more height for more purity remains a good rule: If you want mainly pure products, you should build a high column. Or either it will be not pure enough or very slow. So thinking like "with this height of that packing I get this number of plates, what means according to the calculator that abv" seduces to build not high enough columns. And then sooner or later you will sacrifice purity for speed. Or dilute and run it again (that's what I do with 1m SPP for neutrals: reflux to 90%, diluting, again reflux to 90%. All in all faster and cleaner than one run to 96%).

In practice of course the alcohol strength in the boiler will get lower and lower. So this calculation is incomplete. In reality you want the high abv also when the abv in the boiler has got low, what means you need 1 or 2 more active plates (depends on how complete you want to get all the alcohol and at what abv and speed you want to get the last tails). But the same principles are valid during the whole run.
And packing behaves different not only in dependence of the column diameter and wattage. For example the HETP of a type of packing varies also in dependence of the proportion of vapor and reflux amount in the column. What means in dependence of the reflux ratio. And even in dependence of the abv, because the volume difference of low abv between its vapor and liquid (reflux) is different to the volume difference of high abv alcohol vapor and liquid.
And of course not only the reached abv determines the character of the distillation.
In this way, imperialism brings catastrophe as a mode of existence back from the periphery of capitalist development to its point of departure. - Rosa Luxemburg
User avatar
bunny
Site Donor
Site Donor
Posts: 398
Joined: Fri Jun 26, 2020 8:28 am

Re: Height for purity, diameter for speed, and HETP Calcs

Post by bunny »

Odin wrote: Wed Jan 10, 2018 4:12 am
But that's "just" azeo. It turns out that if you do more redistillations than 15, you get better heads and tails separations. Until 40. So for vodka 40 redistillations is ideal, because it minimizes heads and tails smearing, resulting in an eazy 80% loss in tastes (that are heads and tails associated).

Regards, Odin.

Brought this back as a refresher for longer term members and an introduction to the newer members who missed this.


Plus, I wanted to know what more than 40 re-distillations would gain if anything. I think I can do a lot more, but have no real way to measure the difference personally. I'm sure it wouldn't be a big improvement but maybe an improvement non the less.
User avatar
Demy
Master of Distillation
Posts: 3084
Joined: Fri Sep 04, 2020 1:45 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by Demy »

Good reads in these posts, a few years have passed but some pearls remain.
drmiller100
Rumrunner
Posts: 679
Joined: Wed Mar 11, 2009 9:13 pm

Re: Height for purity, diameter for speed, and HETP Calcs

Post by drmiller100 »

Research I've read suggests 20 or more boil/condenses to go from 10 percent to 95 percent. The difference between 94 to 95 percent is pretty tough at that level.

My research says MY columns needed a lot more than that.

I reached a point where more column packing height didn't make the still any faster but more column height never hurt anything.
I ONLY do azeo. I have no doubt more column packing will change flavors and smells.
Now I know how you claim azeo so easy, it's based on a meat thermometer. :lol:
Post Reply