Column Height Vs Diameter ratio

[LWTCS]

http://homedistiller.org/forum/viewtopi ... 4#p6834174

I must say,, for not even having fired a basic potstill,,, your level of dialog is quite impressive. Not "novice " at all.

What is your background that enables you to stand and debate this subject matter. We have had chemistry major's with less awareness.

You seem quite good. How so?

[Lubavitcher]

ummm im not sure if my proficiency issues forth form any specific background, i was always good at science, my mother was a research librarian and my father is a is a big researcher hes in istanbul atm and later this week he will be in shanghai he gives a lot of graduate seminars in managerial science. simply put im good at research because my parents are

to end the topic kiwi and i seem to agree and so does the evidence there is no direct ratio of width to height for purity
a 2incher of the same height as a 4 incher might only preform 10-20% better but not the 100%-200% better claimed by the people that say that a ratio of 1to 20 is necessary.

if you want to look at laboratory data look at the table on page six http://www.cannoninstrument.com/Pro-PakBulletin.pdf" onclick="window.open(this.href);return false;" rel="nofollow the still that is 12" by 3' is half as good as the 6" by 6' indicating that it is the height that determines purity not diameter. just like it says in the distillers bible HD. we should all make a resolution to bury and censor all talk of 1:24 or 1:30 to make ultra pure stuff because its plain wrong and misleading.

[kiwistiller]

Well I'd still like to be proven wrong. A lot of very respected column designers use those numbers, they must be based on something, right? Just because you and I don't understand it doesn't mean it's wrong. We can take it to a different topic if you'd like?

[rad14701]

The ratios, while generic, are based on what has been proven to work... Sure, I can get close to 95% out of my small scale 1.25" column with only ~9" of packed column, but I can run faster and stay more stable with ~24" of packed column... But let's look at some figures...

For a column of a given diameter:

Increased reflux ratio decreases HETP plate height and increases plate count
Invreased heat input (watts) increases HETP plate height and decreases plate count
Increasing column height increases HETP plate count


Taking these facts into consideration:

There is no direct correlation between column diameterand HETP plate height
Once datum has been compiled a generic ratio can prove helpful in determining packed column height
Using generic ratios is not exacting science, but gets you in the ballpark with margin for error when any of the cparameters mentioned above is changed, within reason

Example:
2"/50mm column
2:1 reflux ratio
1500 watts power input
Stainless Steel scrubbers for packing

12"/30cm height (6:1) = 3.8 HETP's 4.24"/10.8cm tall and 90.4% ABV
24"/60cm height (12:1) = 6.6 HETP's 4.24"/10.8cm tall and 93.4% ABV
30"/76cm height (15:1) = 8 HETP's 4.24"/10.8cm tall and 93.9% ABV
36"/92cm height (18:1) = 9.5 HETP's 4.24"/10.8cm tall and 94.5% ABV
48"/120cm height (24:1) = 12.1 HETP's 4.24"/10.8cm tall and 94.9% ABV
60"/150cm height (30:1) = 14.9 HETP's 4.24"/10.8cm tall and 95.2% ABV

With these figures in mind, and considering how we can adjust our reflux ratio and heat input, we can change the number of plates, plate height, and purity. Therefore the column height:diameter ratio can be used to get a somewhat accurate estimation of whether the column is capable of performing within expectations or not.

With a 24" X 2" column, or 12:1 ratio, we can still get darn near 95% ABV by adjusting reflux ratio and heat input. The same goes for a 48" X 2" column. Now, where the top end comes in is with a 60" X 2" column where we can adjust past our maximum attainable ~95.6% ABV, meaning that we have more fudge factor built into the column. And at the lower end, a 12" column only has 3.8 HETP's (less than 5 HETP's) and therefore it is all but impossible to produce spirits in the ~95% ABV range with all the reflux ratio and heat input adjustments in the world unless you collect at less than one drip per second.

So, what does this all mean? It means that those speculative ratios get us close enough to be considered a reliable guideline. We must also consider that depending on which end of the column diameter extremes we go, the more skewed the accuracy will be. A 1.25"/32mm column will be thrown off more than a 3"/76mm column. So stating that anything below 12:1 won't allow for ~95% ABV and a ratio above 30:1 will have diminishing improvements in ABV is a relatively accurate statement.

Edited: To specify SS scrubbers as packing.

[kiwistiller]

Ok, I've read that, but can't help thinking that as diameter has nothing to do with purity, why are we advising people to build stills based on a ratio with it? Even for a 3" still it's pretty far out of whack.

Here is my thoughts:

Invreased heat input (watts) increases HETP plate height and decreases plate count

sorta, it's increased vapour speed really, right?

There is no direct correlation between column diameterand HETP plate height

ok...

Once datum has been compiled a generic ratio can prove helpful in determining packed column height

Right now this step is the one that you've lost me at. Why? If the diameter isn't related to purity, why does a wider still require a taller column? Because that's what expressing it in a ratio implies.

With a 24" X 2" column, or 12:1 ratio, we can still get darn near 95% ABV by adjusting reflux ratio and heat input. The same goes for a 48" X 2" column. Now, where the top end comes in is with a 60" X 2" column where we can adjust past our maximum attainable ~95.6% ABV, meaning that we have more fudge factor built into the column. And at the lower end, a 12" column only has 3.8 HETP's (less than 5 HETP's) and therefore it is all but impossible to produce spirits in the ~95% ABV range with all the reflux ratio and heat input adjustments in the world unless you collect at less than one drip per second.

ok, but the same goes for a 3" column of the same heights, right? or a 2.5"? or a 4"?

So, what does this all mean? It means that those speculative ratios get us close enough to be considered a reliable guideline. We must also consider that depending on which end of the column diameter extremes we go, the more skewed the accuracy will be. A 1.25"/32mm column will be thrown off more than a 3"/76mm column. So stating that anything below 12:1 won't allow for ~95% ABV and a ratio above 30:1 will have diminishing improvements in ABV is a relatively accurate statement.

Wouldn't it be much easier to talk about this speculative guideline in terms that are actually directly related to purity and therefore useful across a wider range of diameters, namely column height? isn't it a more accurate statement to say 'anything below 24" won't allow for ~95% ABV and heights above above 60" will have diminishing improvements in ABV'?

Still not quite understanding this, but thanks for your time so far.
Kiwi
PS datum is a singular

[rednose]

Looks like you folks love the topic, here a PDF I catched up some weeks ago.

That might help you a little, I still don't get it completely but if I have time I'll try to get deeper in.

I admit not to be the brightest light under the sun when it comes to theory and long math formulas.

[kiwistiller]

I couldn't find anything relating to ratios in there joe. what is it specifically that you thought might help?

[rednose]

I refere to the HETP calcs for packed columns under different packing materials.

[rad14701]

Yeah, rednose, I forgot to note that my calculations were all done with SS structured scrubber...

kiwistiller, I'm surprised that you don't see a correlation between the figures and the height:diameter ratio... Changing the diameter of the column, with all other settings left the same, will show that a larger diameter column will result in shorter HETP's and more plates within the length, just as a smaller diameter column will have taller HETP's and less plates... But the band of efficient operation will still be within the 12:1 to 30:1 ratio range... Trust me, I've tried to disprove the ratios but no matter how I slice and dice the numbers, the results are they same... A generic guideline, based on numbers compiled from formulas, remains almost every bit as effective as running off calculations... Based on the 2"/50mm figures, the ratios are pretty spot on... Remember, the reflux ratio and heat input are two adjustable parameters which can be used to skew the numbers for a column of a given diameter and height... And we can't forget the percentage of alcohol in the wash either as that will also change the results... The ratios only assure us that we have enough HETP's for decent operation, without direct relation to the plate count... If we want more exacting numbers we should refer to the calculator or use manual calculations... None of this has anything to do with take off rates, only whether we can get attain ~95% ABV by adjusting reflux and heat...

I guess the next step is to see numbers showing that with a given diameter and height that you can fall out of the ratio range, within the realm of practicality... By the realm of practicality I mean using realistic reflux ratios as well as heat input within the operational range with respect to the columns physical size...

And one last parameter that we may be overlooking is the actual packed density of an individual column... My packed density is going to be different than the next persons packed density and, in fact, that density could change by simply unpacking and repacking the column with the exact same structured material... That takes some of the exact out of the exact science of HETP's as they relate to our hobby... Only an un-compressible and uniform material will give repeatable results within a narrow range...

[Lubavitcher]

There is no direct correlation between column diameterand HETP plate height

ok...
... Why? If the diameter isn't related to purity, why does a wider still require a taller column? Because that's what expressing it in a ratio implies.

answer kiwi. if you don't understand his question i will rephrase it:

if you have a 2 incher that is 60 inches high that is a 1:30 ratio and i want to build a 6 incher with the same level of purity does that mean i should build it 180 inches high? yes or no?

if yes your wrong look at the data again

if no then i ask "why do you express it in a ratio?" because expressing it as a ratio misleads people to think that a 6 incher should build it 180 inches high in order to have the same plate equivalent as a 2incher that is 60 inches high

[Lubavitcher]

Hey rednose thanks for the pdf... its a beautiful find!

from the way i read it, leaving out all the math i see that it mesh is the fool proof way to build a still and that to design a plate still you really need a lot of math. and dont fall into the pit of thinking that perforated plates are easier to clean cuz they arnt and they also can have flooding or weeping problems if not built right

[rad14701]

if no then i ask "why do you express it in a ratio?" because expressing it as a ratio misleads people to think that a 6 incher should be 180 inches high

The operative phrase should be "can|could be 180 inches high"... No "should" is implied...

[Lubavitcher]

i just edited my post (the part i added is underlined) so you can no longer, dodge the question and troll this topic.

and about my grammar are you REAALLY sugesting that my question should read "if you have a 2 incher that is 60 inches high that is a 1:30 ratio and i want to build a 6 incher with the same level of purity does that mean i could build it 180 inches high? "

dont use big words unless you know the meaning i had to study Kant in college and the construction i used was a hypothetical imperative. so i HAD to use the word should in order to satisfies the word earlier in the sentince "want"

this is the base form "i want XYZ so i should ABC"

[rad14701]

Since you started this topic with the title "compact column design" and you think I'm trolling then I guess I won't continue allowing it to be pulled off-topic any further... I won't be lured into an argument within your topic, Lubavitcher... The troll comment was totally uncalled for, especially from someone as new to the HD community as you are...

[kiwistiller]

Steady on, Lubavitcher, no need to be rude. I've actually asked one of the mods to split this into a new topic from when we start talking about ratios, but I guess we can continue on here for now?

So rad, you often say that 24:1 is the ideal ratio. That's a 48" column. Is 96" the ideal height for a 4" column?

I still haven't seen why a wider column needs to be taller. can you explain this too me simply? I'm not particularly interested in your spreadsheet (no offense intended) because I'm fairly sure that your averaging method of calculating is a bit flawed.

Saying that the ratio is appropriate because it is robust over a range of variables is a little flimsy. wouldn't a simple height figure be just as robust? Also, talking of ranges of heat input is a little strange, when the main reason for a bigger column is so that you can apply more power and keep efficiency (or more correctly vapour speed) at the same level.

Changing the diameter of the column, with all other settings left the same, will show that a larger diameter column will result in shorter HETP's and more plates within the length, just as a smaller diameter column will have taller HETP's and less plates...

That's because you haven't kept your vapour speed at the efficient spot in your calcs. adjust power so the vapour speed is always 20"/s (which is close enough to optimal for mass transfer while keeping a good production rate), like a tuned reflux column should have, and you'll see that your figures for HETP remain constant over column diameter. Aside from the obvious, this also shows how you can't rely on calculators to make inferences beyond their original purpose, because we all know that without reflux distribution the wider columns will suffer from channeling and a loss in efficiency (HETP increase towards the bottom of the column).

Yes, they hold up for as 2" still, because I'm picking that the numbers were made up based on a 2" still. They don't hold up for other diameters when the power is increased accordingly. when the power ISN"T increased accordingly, they are even more inaccurate, because the reduced vapour speed will decrease the HETP, meaning that you could have a shorter column if you increase diameter and keep power constant.

So, I'd like you to simply answer this. Should an ideal 4" column be twice the height of a 2" column?

[Lubavitcher]

listen im sory for saying that you were trolling. but thats just how i felt. rad made some statements. kiwi asked him some questions about some contradictions. rad didnt give a answer. i re-asked the questions he didnt give an answer he just spoke about grammar. i felt offended that i would ask a question and he would ignore it... thats all

[rad14701]

Ok, we'll give this another go...

First, Lubavitcher, some of us lend support in these forums during our work hours which means we may not always have time to give a complete answer at a given moment... I could tell by the tone of your posts that you were getting worked up and chose not to "go there" at that exact moment but did notice your attempting to steer my words in a direction I have clearly steered away from more than a few times when this topic has come up...


@ kiwistiller

No, I have never said that a 4" column would need to be twice as tall as for a 2" column... That is why the range is so broad... I have always held the opinion that larger columns than 2" would require a lower ratio than 2" and columns smaller would require a higher ratio than 2"... Your comments on vapor speed help prove that out... You are also correct in that no calculator is going to relate to accurate real life results in practical application... As I eluded to in an earlier post, due to real world parameter differences, calculator results may be no more accurate than simple ratios... We've still got to use some common sense and be willing to experiment a bit...

[kiwistiller]

@ kiwistiller
No, I have never said that a 4" column would need to be twice as tall as for a 2" column... That is why the range is so broad... I have always held the opinion that larger columns than 2" would require a lower ratio than 2" and columns smaller would require a higher ratio than 2"... Your comments on vapor speed help prove that out... You are also correct in that no calculator is going to relate to accurate real life results in practical application... As I eluded to in an earlier post, due to real world parameter differences, calculator results may be no more accurate than simple ratios... We've still got to use some common sense and be willing to experiment a bit...

Ok. So if the ratio changes, why is it better to use a ratio, which becomes inaccurate as diameters change and generally tends to confuse people, as opposed to just talking about column height? Why bring column diameter into at all?

Kiwi

[rad14701]

Ok. So if the ratio changes, why is it better to use a ratio, which becomes inaccurate as diameters change and generally tends to confuse people, as opposed to just talking about column height? Why bring column diameter into at all?

Kiwi

I see a chart in my head and that chart works right along with the calculator... It's hard to explain and, to be honest, I'm tired of trying... Maybe ratios aren't for everyone...

To be honest, I come here to relax...

[kiwistiller]

Well I think it is worth discussing if it is going to be recommended to newbies. If it can't be justified more than you seeing it in your head, maybe we should change the way we explain the requirement of column height to newbies? Maybe in terms like, column height? I've asked snuffy about it as well, he hasn't gotten back to me yet.

I'd like it if you could think on it maybe for a bit, for a day or two or whatever, and see if you can explain it to me (I really want to know). Because surely advice that can't be explained is bad advice?

[Lubavitcher]

HETP (Height Equivalent to a Theoretical Plate)
eg how much height of packing is needed to get the same performance as a theoretically perfect (equilibrium) plate. I've done the estimate using Onda's method (from Coulson, Richardson & Sinnott "Chemical Engineering").

Calculate the effective area (aw) using :

where:

aw = effective interfacial area of packing per unit volume (m2/m3)
a = actual area of packing per unit volume (m2/m3)
sc = critical surface tension for the particular packing material (see table below)
sL = liquid surface tension mN/m
L*w = mass flowrate per unit cross-sectional area, kg/m2s = L'

Critical surface tension
Material

sc [mN/m]
Ceramic 61
Metal (steel) 75
Plastic (polyethylene) 33
Carbon 56

Then calculate the liquid and gas mass transfer coefficients (kL and kG) using :

where:

K5 = 5.23 for packing sizes >15mm and 2.0 for sizes <15mm
V*w = gas mass flowrate per cross-sectional area [kg/m2s] = G'
dp = packing size [m]
kL = gas film mass transfer coefficient,[kmol/m2s atm or kmol/m2s bar]
kG = liquid film mass transfer coefficient [kmol/m2s or kmol/m3 = m/s depending on whether use R=0.08206 atm.m3/kmol K or R=0.08314 bar m3/mol K]

From these you can then calculate the film transfer heights:

where:

P = column operating presure, [atm or bar]
Ct = total concentration, kmol/m3 = r L / molecular weight solvent
Gm = molar gas flowrate per unit cross-sectional area, kmol/m2s
Lm = molar liquid flowrate per unit cross-sectional area, kmol/m2s

From these you can then estimate the height of the overall gas-phase transfer unit:

where:

m = slope of the equilibrium line (I reckon its about 0.49)
Gm/Lm = slope of the operating line

Now... for a section of the packed column in which the operating and equilibrium lines can be considered straight (hey- that's almost us !), theoretical stages can be converted to numbers of transfer units by :

and then

where:

Zp = packed bed height
Nt = number of theoretical stages

this is a quote from HD im sorry all the jpegs and charts dont appear if you look through you will see that there are NO variables for diameters...why? because diameter does not affect HETP!

[HookLine]

The main design parameter in a reflux column is vapour speed. Design your column around that. Keep it between 300-500 mm/s (12-20"/s)*, start with 400 mm/s (16"/s) and fine tune.

*This is for small, mesh/scrubber packed columns (maybe up to 100 mm diameter). Don't know if it is true for other types of packing (eg Raschig rings), or for larger or plate type columns.

[rednose]

If I understand the physics right, the column lenght depends on the density of the packing material to simulate a ideal plate.

If the density of the packing material is too high the column will throw out liquid instead of vapor what already happened to some Bok users, me included.

As more vapor speed you produce as wider your column diameter have to be.

In plated columns the limitation of the high is the spraying effect of the bubbles, liquid should not reach the next plate once the column is stable.

I'm just thinking with a common sense but could be wrong somewhere.

Joe

[rad14701]

One issue that we need to remember, in our practical application, is that the vapor speed slows throughout the height of the column because there is less vapor left by the time it reaches the top of the packed column... We aren't getting as much vapor with the same density or speed at the top as enters the bottom of the column... We have reflux, resistance, and temperature drop, to thank for that... The idea of perfect vapor speed goes out the window as soon as the vapor hits the structured packing...

I have read and re-read the section on HETP's and column design on the parent site and even after all of the calculations and comments there are still a few holes in the theory that have never been fully examined unless Snuffy has done recent research in his mad lab... While Tony, Mike, and others have contributed a lot of useful information, I don't think that any additional follow-up information has been sought...

[kiwistiller]

I haven't heard of vapour speed slowing before rad. do you have a link or something to that info? I thought that due to mass transfer with reflux, it would stay more or less constant?

[Lubavitcher]

im posting a graph i made of the data of the the hetp off of the pro pack pdf.... dont get fooled and think that this is data for alcohol. but all i want to show is how distillation data from a lab proves that HETP in not related to diameter

[Ayay]

Using a column packed with mesh/scrubbers we are limited to a 2"-3" column diameter because that's how far the reflux can spread sideways as it decends in a hobby environment. A column height between 20-30 diameters for 2" -3" columns is best for a neutral at 0.3-1.0 Litres/hour. Any more (pot and strip excluded), then you are somewhere other than homedistiller.org.

If you are using plates then the sideways spread is determined by the hole locations and can go much wider.

Theory and practise mingle together and the result is what you git. Chicken an egg?

[kiwistiller]

I've been talking with Riku and the Mike's from amphora, they say the 4" works fine with mesh. I'll be able to verify myself soon I guess

Any more (pot and strip excluded), then you are somewhere other than homedistiller.org.

I don't quite agree with that. I don't want to make any more, I just want to make it faster. Pot runs are interesting, reflux is a chore to me. want it over and done as fast as possible, so I can get to infusing gin with my pot still

[olddog]

I don't quite agree with that. I don't want to make any more, I just want to make it faster. Pot runs are interesting, reflux is a chore to me. want it over and done as fast as possible,

+1


OD

[Lubavitcher]

A column height between 20-30 diameters for 2" -3" columns is best

diameter has nothing to do with column height. what you meant to say is a 3 foot to 5 foot column will work just fine